from:"sebb"

Re: ooo-forums downtime

2018-01-10 Thread sebb

However such things could (IMO should) be noted in upcoming
maintenance (i.e. scheduled downtime)

On 10 January 2018 at 09:53, Greg Stein  wrote:
> status.apache.org reports services that are *down*, rather than individual
> bugs/problems.
>
> Thanks,
> -g
>
> On Wed, Jan 10, 2018 at 2:03 AM, FR web forum  wrote:
>>
>> I note that give you this information next time
>> Why these outages are not reported in http://status.apache.org
>>
>>
>> - Mail original -
>> > De: "Chris Lambertus" 
>> > À: priv...@infra.apache.org
>> > Cc: "Andrea Pescetti" , dev@openoffice.apache.org
>> > Envoyé: Mercredi 10 Janvier 2018 03:22:04
>> > Objet: Re: ooo-forums downtime
>> >
>> >
>> > I’ve pushed a change which fixes this on the database side. I also
>> > saw the email about the forum being down yesterday, but it was a
>> > transient problem. It would be helpful for me to know what the start
>> > time and duration was for any future reports of slowness or outages.
>> >
>> > -Chris
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > > On Jan 4, 2018, at 5:49 PM, Chris Lambertus  wrote:
>> > >
>> > >
>> > > There may not be anything we can do about this until phpBB is
>> > > upgraded.
>> > >
>> > > I put the wiki database move on hold until I can look into it more.
>> > >
>> > >
>> > >> On Jan 4, 2018, at 2:42 PM, Andrea Pescetti 
>> > >> wrote:
>> > >>
>> > >> CCing Chris for the two messages below. Maybe a SQL version issue?
>> > >> Regards,
>> > >> Andrea.
>> > >>
>> > >> On 04/01/2018 21:59, Hagar Delest wrote:
>> > >>> Note: same for the other languages (Italian anf French tested).
>> > >>>
>> > >>> The DISTINCT 3065 thing seems a common one:
>> > >>>
>> > >>> In French forum:
>> > >>> Erreur générale
>> > >>> SQL ERROR [ mysqli ]
>> > >>>
>> > >>> Expression #1 of ORDER BY clause is not in SELECT list,
>> > >>> references
>> > >>> column 'forumaoo_fr.t.topic_last_post_time' which is not in
>> > >>> SELECT list;
>> > >>> this is incompatible with DISTINCT [3065]
>> > >>>
>> > >>> And in Italian:
>> > >>> SQL ERROR [ mysqli ]
>> > >>>
>> > >>> Expression #1 of ORDER BY clause is not in SELECT list,
>> > >>> references
>> > >>> column 'forumaoo_it.t.topic_last_post_time' which is not in
>> > >>> SELECT list;
>> > >>> this is incompatible with DISTINCT [3065]
>> > >>>
>> > >>> Hagar
>> > >>>
>> > >>>
>> > >>> Le 04/01/2018 à 21:33, Hagar Delest a écrit :
>> >  Hi,
>> > 
>> >  Since the move, we noticed that when performing an advanced
>> >  search
>> >  that should display the results as topics, then we get an error:
>> > 
>> >  General Error
>> >  SQL ERROR [ mysqli ]
>> > 
>> >  Expression #1 of ORDER BY clause is not in SELECT list,
>> >  references
>> >  column 'forumaoo_en.t.topic_last_post_time' which is not in
>> >  SELECT
>> >  list; this is incompatible with DISTINCT [3065]
>> > 
>> >  SQL
>> > 
>> >  SELECT DISTINCT SQL_CALC_FOUND_ROWS t.topic_id FROM
>> >  phpbb_en_topics t,
>> >  phpbb_en_posts p WHERE MATCH (p.post_subject, p.post_text)
>> >  AGAINST
>> >  ('+solved ' IN BOOLEAN MODE) AND t.topic_id = p.topic_id ORDER
>> >  BY
>> >  t.topic_last_post_time DESC LIMIT 250
>> > 
>> >  The same search is fine when set to display results as posts.
>> > 
>> >  Could you have a look please?
>> >  Thanks
>> > 
>> >  Hagar
>> > 
>> > 
>> >  Le 04/01/2018 à 03:30, Chris Lambertus a écrit :
>> > >> On Jan 1, 2018, at 2:24 PM, Chris Lambertus 
>> > >> wrote:
>> > >>
>> > >> Hi folks,
>> > >>
>> > >> At 0100 UTC on Thursday January 4th, I will be moving the
>> > >> OpenOffice
>> > >> PHPBB forum databases to a new server. The forums will be
>> > >> offline
>> > >> during this process, and is expected to take about 2 hours.
>> > >>
>> > >> -Chris
>> > >> ASF Infra
>> > >>
>> > >>
>> > >
>> > > Hi all,
>> > >
>> > > This move has been completed. Spot checks on the forums look
>> > > good.
>> > >
>> > > -Chris
>> > >
>> > 
>> > >>>
>> > >>>
>> > >>>
>> > >>> -
>> > >>> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
>> > >>> For additional commands, e-mail: dev-h...@openoffice.apache.org
>> > >>>
>> > >>>
>> > >
>> >
>> >
>
>

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Trivial German grammar error on get-involved page.

2016-05-10 Thread sebb

The get involved page [1] says:

Diese Seite in Deutsch

I'm not a native German speaker (*), but that looks wrong, IMO it should be

Diese Seite auf Deutsch [ansehen]

Google Translate is confused.

"view This site in German" gives the correct answer, but
"view this site in German" and
"View this site in German" do not


[1] http://openoffice.apache.org/get-involved.html
(*) I'm sure your project has plenty of those ...

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [DOWNLOAD][PROPOSAL] How to download another version than offered in the green box?

2014-05-06 Thread sebb

On 7 May 2014 00:10, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 05/06/2014 02:24 AM, schrieb sebb:

 On 5 May 2014 20:00, Marcus (OOo)marcus.m...@wtnet.de  wrote:

 Am 05/05/2014 01:29 AM, schrieb sebb:

 On 4 May 2014 17:37, Marcus (OOo)marcus.m...@wtnet.de   wrote:


 Another update, seems this weekend I've my productive phase: ;-)

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 What was done:

 - Grey box:

 - Bye, bye grey color - hello green freshness.


 +1


 - The select boxes are prefilled with the browser guessed data.


 +1


 - The information link has now a (i) icon.



 Looks better, but the hover text has gone. It was useful...



 I get the text What is the difference between full installation and
 language pack? when hovering the mouse over the (i) icon.
 Have you done something different?


 I get hover text for the download and report links, but not for the (i).
 This applies to Firefox and Opera 12.17 on WinXP, and Firefox on OS X
 (Mavericks)


 I did a test with IE10 and there is also no hover text. I've now added text
 to the respective element.

Firefox (WinXP/Mavericks) now shows the hover text, so is working fully.
Opera (12.1/WinXP) also shows hover text so is now fully working


 Chrome shows the (i) hover, but the download and report texts are
 missing - I only see the (i) and the link icon.


 Still the same when changing a value of a drop-down box? This will trigger a
 new creation of download links.

Changing the drop-down selection does change the hover texts for the
Download install and Download langpack links
However the fixed text Download ... does not appear, and the (I) is
shifted left.


 I do see the download hovers if the mouse is moved over the blank
 space to the left of the (i)
 The drop-downs are correctly preset, but there is no info in the light
 green box.


 These will be set together with the download links.

They are not populated when I change the dropdown contents, though the
invisible download hover texts do change.


 This is true for Chrome on WinXP and OS/X, and for Safari and Opera 20.0
 on OS/X
 [I think Opera 20 is based on Chrome]

 IE8/WinXP is completely broken; no content in the drop-down boxes and
 no download or report text.

 So none of the Win/XP or OS/X browsers are fully working for me.


 Hm, that's not good. Need to work on this.

Should we start a new e-mail thread?

This is getting tricky to follow with all the quoted text.
Or perhaps some other medium might be easier to use, e.g. Bugzilla?

 Thanks

 Marcus


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [DOWNLOAD][PROPOSAL] How to download another version than offered in the green box?

2014-05-05 Thread sebb

On 5 May 2014 20:00, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 05/05/2014 01:29 AM, schrieb sebb:

 On 4 May 2014 17:37, Marcus (OOo)marcus.m...@wtnet.de  wrote:

 Another update, seems this weekend I've my productive phase: ;-)

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 What was done:

 - Grey box:

- Bye, bye grey color - hello green freshness.

 +1

- The select boxes are prefilled with the browser guessed data.

 +1

- The information link has now a (i) icon.


 Looks better, but the hover text has gone. It was useful...


 I get the text What is the difference between full installation and
 language pack? when hovering the mouse over the (i) icon.
 Have you done something different?

I get hover text for the download and report links, but not for the (i).
This applies to Firefox and Opera 12.17 on WinXP, and Firefox on OS X
(Mavericks)
Chrome shows the (i) hover, but the download and report texts are
missing - I only see the (i) and the link icon.
I do see the download hovers if the mouse is moved over the blank
space to the left of the (i)
The drop-downs are correctly preset, but there is no info in the light
green box.
This is true for Chrome on WinXP and OS/X, and for Safari and Opera 20.0 on OS/X
[I think Opera 20 is based on Chrome]

IE8/WinXP is completely broken; no content in the drop-down boxes and
no download or report text.

So none of the Win/XP or OS/X browsers are fully working for me.


- Added a Report broken link - pointing to the debug webpage.


 I clicked on the report button.
 The test page that appears needs to have some introductory text to
 tell users what to do.


 Yes, like I wrote right below. ;-) For the moment it's just to see that
 something is working.


  The icon alone is maybe not meaningful enough, therefore an
  additional text link.
  (Of course the debug webpage needs further work to really work as
  report function.)
- All not primary important data are now in the sub-box.
  Now it looks more clean.

 - Green box:

- All not primary important data are now in the sub-box.
  Now it looks more clean and the focus is set for downloading only.

 - Links:

- I've moved and deleted links and also adjusted the headlines.
- In the nav bar as well as in the colored boxes.
- Of course it's just a suggestion and can be discussed/changed.


 I think the download links could be more descriptive (there's plenty
 of room) - e.g.

 Download full installation package (including language pack)
 and
 Download language add-on pack only


 This is were I need some help *what* to write - text or button or both or
 something else ...
 So, thanks for your suggestion.


 As an aside, the information page about language packs could perhaps
 describe how dictionaries compare with language packs. Do language
 packs include any dictionaries? Or do they just include the text
 strings used in the application?
 I found the page quite difficult to follow.


 OK, I will improve the text.


 What is not yet working:

 - The version drop-down-box has still no effect for others than 4.1.0.
It's a bigger effort as I've to take into account every
available build for every release in the past.
 - The [Reset] button is now not working as expected, however maybe
it's indeed no longer necessary?

What do you think?


 The reset button could be dropped.


 OK

 Thanks for your feedback.

 Marcus



 What are the next steps:

 - Make the version drop-down-box work for older releases.
 - Improve the styling of the both download links.
 - Combine the both green boxes.

 - And further suggestions from you.

 Thanks in advance for testing. :-)

 Marcus



 Am 05/03/2014 12:41 AM, schrieb Marcus (OOo):

 I've done some further work.

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 What is working:

 - Filesizes are shown for selected full install + langpack.
 - Checksum links are shown for selected full install + langpack.
 - When choices are done a little help icon is shown. Click on it to get
 to a help text for full install vs. langpack.
 - Some release data is shown like in the green box.

 What is not yet working:

 - The version dropdown box has no effect yet but it shows already how
 it could work (likely).

 - I've not tested with all browsers. So, maybe something is not working.

 What are the next steps:

 - Reduce the additional links in the green box and re-work in the right
 hand nav bar - delete duplicates, change order/positions, delete if
 not needed here, etc.

 In general, this should result in a clean-up.

 - Make the version drop-down box work to choose from older releases.

 After that I'll publish a new update to the webpage and go on with your
 further comments and suggestions.

 So, please have a look and don't hesitate to comment.

 Thanks

 Marcus



 Am 04/29/2014 12:40 AM, schrieb Marcus (OOo):


 Thanks a lot for your input.

 A few of them (delete some

Re: [DOWNLOAD][PROPOSAL] How to download another version than offered in the green box?

2014-05-04 Thread sebb

On 4 May 2014 17:37, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Another update, seems this weekend I've my productive phase: ;-)

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 What was done:

 - Grey box:

   - Bye, bye grey color - hello green freshness.
+1
   - The select boxes are prefilled with the browser guessed data.
+1
   - The information link has now a (i) icon.

Looks better, but the hover text has gone. It was useful...

   - Added a Report broken link - pointing to the debug webpage.

I clicked on the report button.
The test page that appears needs to have some introductory text to
tell users what to do.

 The icon alone is maybe not meaningful enough, therefore an
 additional text link.
 (Of course the debug webpage needs further work to really work as
 report function.)
   - All not primary important data are now in the sub-box.
 Now it looks more clean.

 - Green box:

   - All not primary important data are now in the sub-box.
 Now it looks more clean and the focus is set for downloading only.

 - Links:

   - I've moved and deleted links and also adjusted the headlines.
   - In the nav bar as well as in the colored boxes.
   - Of course it's just a suggestion and can be discussed/changed.


I think the download links could be more descriptive (there's plenty
of room) - e.g.

Download full installation package (including language pack)
and
Download language add-on pack only

As an aside, the information page about language packs could perhaps
describe how dictionaries compare with language packs. Do language
packs include any dictionaries? Or do they just include the text
strings used in the application?
I found the page quite difficult to follow.

 What is not yet working:

 - The version drop-down-box has still no effect for others than 4.1.0.
   It's a bigger effort as I've to take into account every
   available build for every release in the past.
 - The [Reset] button is now not working as expected, however maybe
   it's indeed no longer necessary?

   What do you think?

The reset button could be dropped.


 What are the next steps:

 - Make the version drop-down-box work for older releases.
 - Improve the styling of the both download links.
 - Combine the both green boxes.

 - And further suggestions from you.

 Thanks in advance for testing. :-)

 Marcus



 Am 05/03/2014 12:41 AM, schrieb Marcus (OOo):

 I've done some further work.

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 What is working:

 - Filesizes are shown for selected full install + langpack.
 - Checksum links are shown for selected full install + langpack.
 - When choices are done a little help icon is shown. Click on it to get
 to a help text for full install vs. langpack.
 - Some release data is shown like in the green box.

 What is not yet working:

 - The version dropdown box has no effect yet but it shows already how
 it could work (likely).

 - I've not tested with all browsers. So, maybe something is not working.

 What are the next steps:

 - Reduce the additional links in the green box and re-work in the right
 hand nav bar - delete duplicates, change order/positions, delete if
 not needed here, etc.

 In general, this should result in a clean-up.

 - Make the version drop-down box work to choose from older releases.

 After that I'll publish a new update to the webpage and go on with your
 further comments and suggestions.

 So, please have a look and don't hesitate to comment.

 Thanks

 Marcus



 Am 04/29/2014 12:40 AM, schrieb Marcus (OOo):

 Thanks a lot for your input.

 A few of them (delete some obviously not needed text, make the choice
 order of the drop-dwn boxes irrelevant) are already committed. But the
 most need more time.

 Please don't expect to have this full-featured for the AOO 4.1.0
 release. I need more time to do all this. Luckily the need work-free
 days are coming. :-)

 Marcus



 Am 04/27/2014 08:19 PM, schrieb Marcus (OOo):

 Hi all,

 while enjoying the Easter time I made some progress about a long-term
 problem:

 - Downloading the offered file via the green box is fast and easy
 - Downloading a different file than the offered is not

 Via the big table on the other.html is now easy to maintain for us but
 has no improvement for the user as it is large and complex.

 Therefore I've thought of making another offer directly on the main
 download webpage:

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 Please don't look too close to look  feel - sure the grey color is not
 the best one ;-) - this can and will be improved for sure.

 For now please have a look for the functionality via the drop-down boxes
 and the then created links below them.

 What do you think? is this a direction we can/should go in general?
 Or do you have complete different thoughts?

 Thanks in advance.

 Marcus


 -
 To unsubscribe,

Re: [DOWNLOAD][PROPOSAL] How to download another version than offered in the green box?

2014-05-03 Thread sebb

On 3 May 2014 10:30, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 05/03/2014 04:05 AM, schrieb sebb:

 How about presetting the fields according to the autodetected values,
 where possible?


 actually I've planned this when the green and grey box will be combined but
 ...


 Then if the auto-detection works for some of the fields it will reduce
 the work for the user.
 Also if the autodetection works, but the user wants to download for a
 differnt OS or language it also reduces the work.


 ... this is reasonable enough to implement it in the grey box first. Also
 for our tests so far. Will try to set this up.


 BTW, it might perhaps be worth adding a link to the screen so people
 can report auto-detection failures.
 This could be a ? link like the What is the difference between full
 install ... but instead invites the user to visit the browser
 detection test page.


 Do you mean this page?
 http://ooo-site.staging.apache.org/download/test/analyze.html

 If so, then yes, could be.


 On OS X  Safari and Chrome auto-detect as US English.


 Then it could already help to get 2 outputs from the page above. ;-)
 Please would you? Just copy  paste the table without formatting as plain
 text.


Looks like the problem is that Chrome and Safari return
navigator.language = en-us
AFAICT they don't pick up the OS X language setting.

It looks like the recommended way to detect the language is to use the
Accept-Language header.
The navigator.language property appears to relate to the browser code
rather than the OS on which it runs.

I downloaded the enGB version of Firefox, but Safari was pre-installed
and I did not get a choice with Chrome.


 Only Firefox correctly detects as British English
   In all 3 cases the OS type is correctly recognised.


 Great. :-)

 Thanks for your comment.

 Marcus



 On 2 May 2014 23:41, Marcus (OOo)marcus.m...@wtnet.de  wrote:

 I've done some further work.

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 What is working:

 - Filesizes are shown for selected full install + langpack.
 - Checksum links are shown for selected full install + langpack.
 - When choices are done a little help icon is shown. Click on it to get
to a help text for full install vs. langpack.
 - Some release data is shown like in the green box.

 What is not yet working:

 - The version dropdown box has no effect yet but it shows already how
it could work (likely).

 - I've not tested with all browsers. So, maybe something is not working.

 What are the next steps:

 - Reduce the additional links in the green box and re-work in the right
hand nav bar - delete duplicates, change order/positions, delete if
not needed here, etc.

In general, this should result in a clean-up.

 - Make the version drop-down box work to choose from older releases.

 After that I'll publish a new update to the webpage and go on with your
 further comments and suggestions.

 So, please have a look and don't hesitate to comment.

 Thanks

 Marcus



 Am 04/29/2014 12:40 AM, schrieb Marcus (OOo):

 Thanks a lot for your input.

 A few of them (delete some obviously not needed text, make the choice
 order of the drop-dwn boxes irrelevant) are already committed. But the
 most need more time.

 Please don't expect to have this full-featured for the AOO 4.1.0
 release. I need more time to do all this. Luckily the need work-free
 days are coming. :-)

 Marcus



 Am 04/27/2014 08:19 PM, schrieb Marcus (OOo):


 Hi all,

 while enjoying the Easter time I made some progress about a long-term
 problem:

 - Downloading the offered file via the green box is fast and easy
 - Downloading a different file than the offered is not

 Via the big table on the other.html is now easy to maintain for us
 but
 has no improvement for the user as it is large and complex.

 Therefore I've thought of making another offer directly on the main
 download webpage:

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 Please don't look too close to look  feel - sure the grey color is not

 the best one ;-) - this can and will be improved for sure.

 For now please have a look for the functionality via the drop-down
 boxes
 and the then created links below them.

 What do you think? is this a direction we can/should go in general?
 Or do you have complete different thoughts?

 Thanks in advance.

 Marcus


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [DOWNLOAD][PROPOSAL] How to download another version than offered in the green box?

2014-05-02 Thread sebb

How about presetting the fields according to the autodetected values,
where possible?

Then if the auto-detection works for some of the fields it will reduce
the work for the user.
Also if the autodetection works, but the user wants to download for a
differnt OS or language it also reduces the work.

BTW, it might perhaps be worth adding a link to the screen so people
can report auto-detection failures.
This could be a ? link like the What is the difference between full
install ... but instead invites the user to visit the browser
detection test page.

On OS X  Safari and Chrome auto-detect as US English.
Only Firefox correctly detects as British English
 In all 3 cases the OS type is correctly recognised.

On 2 May 2014 23:41, Marcus (OOo) marcus.m...@wtnet.de wrote:
 I've done some further work.

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 What is working:

 - Filesizes are shown for selected full install + langpack.
 - Checksum links are shown for selected full install + langpack.
 - When choices are done a little help icon is shown. Click on it to get
   to a help text for full install vs. langpack.
 - Some release data is shown like in the green box.

 What is not yet working:

 - The version dropdown box has no effect yet but it shows already how
   it could work (likely).

 - I've not tested with all browsers. So, maybe something is not working.

 What are the next steps:

 - Reduce the additional links in the green box and re-work in the right
   hand nav bar - delete duplicates, change order/positions, delete if
   not needed here, etc.

   In general, this should result in a clean-up.

 - Make the version drop-down box work to choose from older releases.

 After that I'll publish a new update to the webpage and go on with your
 further comments and suggestions.

 So, please have a look and don't hesitate to comment.

 Thanks

 Marcus



 Am 04/29/2014 12:40 AM, schrieb Marcus (OOo):

 Thanks a lot for your input.

 A few of them (delete some obviously not needed text, make the choice
 order of the drop-dwn boxes irrelevant) are already committed. But the
 most need more time.

 Please don't expect to have this full-featured for the AOO 4.1.0
 release. I need more time to do all this. Luckily the need work-free
 days are coming. :-)

 Marcus



 Am 04/27/2014 08:19 PM, schrieb Marcus (OOo):

 Hi all,

 while enjoying the Easter time I made some progress about a long-term
 problem:

 - Downloading the offered file via the green box is fast and easy
 - Downloading a different file than the offered is not

 Via the big table on the other.html is now easy to maintain for us but
 has no improvement for the user as it is large and complex.

 Therefore I've thought of making another offer directly on the main
 download webpage:

 http://ooo-site.staging.apache.org/download/test/index_droplist.html

 Please don't look too close to look  feel - sure the grey color is not
 the best one ;-) - this can and will be improved for sure.

 For now please have a look for the functionality via the drop-down boxes
 and the then created links below them.

 What do you think? is this a direction we can/should go in general?
 Or do you have complete different thoughts?

 Thanks in advance.

 Marcus


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Server status: What's Up; What's Down?

2014-04-24 Thread sebb

On 24 April 2014 18:21, Joe Schaefer joe_schae...@yahoo.com wrote:
 We had to restore blogs from backups, so some of the latest
 data was lost.  The service is stable again, sorry for the inconvenience
 this has caused.

Status page is currently showing red for ASF blogs...

 Buildbot/CMS is still not building off commit-triggers with no predictable 
 ETA, but we're hoping the resolution will come in days versus weeks.

This e-mail has only been sent to dev@ooo, but I assume it applies to
all CMS servers/blogs?


 On Thursday, April 24, 2014 1:16 PM, Rob Weir robw...@apache.org wrote:

 On Thu, Apr 24, 2014 at 12:52 PM, Kay Schenk kay.sch...@gmail.com wrote:
 On Thu, Apr 24, 2014 at 6:56 AM, Rob Weir robw...@apache.org wrote:

 On Tue, Apr 22, 2014 at 10:56 AM, Rob Weir robw...@apache.org wrote:
  CMS appears to be back up.
 

 I assume this is still down?  I'm able to check in files, but not publish.


  Blog is down.
 

 Blog appears to be back up.  Almost.

 https://blogs.apache.org/OOo/

 1) Images seem to be missing

 2) The most recent blog post (Apache OpenOffice Weekly News #1) which
 was posted on Monday is gone :-(


 This is too bad. Should you re-post?


I don't want to go through the effort of funding and reposting all the
images from past blog posts, and also recreating the last blog post,
if this is still work in progress.   If there is data loss that we
need to recover from, then I'd like to hear it officially from Infra.
But maybe this is actually still being worked on and we haven't lost
anything.

-Rob



  Just got a report that downloads were timing out.  Appears to be on SF
 side.
 
  How are the buildbots?
 
  Regards,
 
  -Rob

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org




 --
 -
 MzK

 Life is either a daring adventure, or nothing.
 -- Helen Keller


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org





-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Incubator still has SVN tree

2014-03-24 Thread sebb

On 23 March 2014 08:52, Andrea Pescetti pesce...@apache.org wrote:
 sebb wrote:

 There is still an SVN tree  under Incubator:
 https://svn.apache.org/repos/asf/incubator/ooo/
 I assume this was all transferred to the main ASF SVN area at
 https://svn.apache.org/repos/asf/openoffice/
 If  so, please can someone tidy up the Incubator tree?


 It was indeed all transferred, and the old copy made read-only by Infra. It
 wasn't deleted since we still had releases and documentation around that
 referred to the Incubator.

 I believe it is now safe to delete it. I don't know who can do it. I didn't
 try, but maybe we need to contact Infra since it is read-only.

I can remove the read-only marker if you wish, in which case anyone on
the IPMC - or an ASF member - can remove it.


 Our project site https://openoffice.apache.org/ only had one page
 referring to the Incubator. It was very outdated, we didn't have anything
 linking to it and I've now removed it. See
 http://svn.apache.org/viewvc/openoffice/site/trunk/content/release-process.mdtext?revision=1418982view=markup
 in case we need to restore it.

 Regards,
   Andrea.

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Incubator still has SVN tree

2014-03-24 Thread sebb

On 24 March 2014 13:51, Herbert Duerr h...@apache.org wrote:
 On 23.03.2014 09:52, Andrea Pescetti wrote:

 sebb wrote:

 There is still an SVN tree  under Incubator:
 https://svn.apache.org/repos/asf/incubator/ooo/
 I assume this was all transferred to the main ASF SVN area at
 https://svn.apache.org/repos/asf/openoffice/
 If  so, please can someone tidy up the Incubator tree?


 It was indeed all transferred, and the old copy made read-only by Infra.
 It wasn't deleted since we still had releases and documentation around
 that referred to the Incubator.


 There are still many e.g. Wiki pages that reference the incubator locations.
 I updated some of them but there are still plenty left, e.g. about the
 branch aw080, netbeans, OOXML, etc.

 I suggest to update them to their new location at
 https://svn.apache.org/repos/asf/openoffice
 For links that should continue to point into the incubator directory, e.g.
 because the branch was integrated and removed at the new location SVN offers
 the possibility to reference older branches using the
 https://svn.apache.org/repos/asf/!svn/bc/155/incubator/ooo/trunk/
 notation.

Although that should work, I think it could be confusing.

IMO it would be better to find the appropriate entry in the new SVN layout.
If there is no such entry, is there really a need to keep the
reference to the incubator-only SVN tree?

 Herbert


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: How to download your program?

2014-03-24 Thread sebb

I suspect that what the OP meant to write was:

I was in need of an office program for my job applications, and a
search on google directed me to your openoffice.I had great
difficulties downloading the software on your download site:
http://www.openoffice.org/download/index.html I could not find a link
where i could download this, only extensions and templates, while this
could be interesting they are worthless without  the main program.

(I have editted the above to add a couple of spaces only)

If so, then perhaps the autodetection was not working for the OP.




On 24 March 2014 08:10, Raphael Bircher rbirc...@apache.org wrote:

 Am 24.03.2014 um 05:51 schrieb øyvind sørensen:

 I was in need of an office program for my job applications, and a search on 
 google directed me to your openoffice.I had great difficulties downloading 
 the software on your download site: 
 http://www.openoffice.org/download/index.htmlIcould

 Maybe you try without the iCloud.

 http://www.openoffice.org/download/index.html

 not find a link where i could download this, only extensions and templates, 
 while this could be interesting they are worthless without  the main 
 program.So i downloaded kingsoft office instead, because they were easily 
 obtainable, but i had problems utilizing their program.Then i did more 
 (several) google search for openoffice download, and after dismissing CNET 
 and other ways to get your software i finally got a downloadable solution. I 
 do not know what i did or how i got there, but i can say that I am now happy 
 with your software and have got the documents i needed written and PDF-ified
 I write this so that other can enjoy the benefit of using your very great 
 software.Please make a link to download Apache openoffice.exe on yor 
 website.

 I tro, håp og kjærlighet Øyvind Sørensen



 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Incubator still has SVN tree

2014-03-22 Thread sebb

There is still an SVN tree  under Incubator:

https://svn.apache.org/repos/asf/incubator/ooo/

I assume this was all transferred to the main ASF SVN area at

https://svn.apache.org/repos/asf/openoffice/

If  so, please can someone tidy up the Incubator tree?

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: New committer: Tsutomu Uchino (hanya)

2014-01-01 Thread sebb

On 28 December 2013 23:42, Alexandro Colorado j...@oooes.org wrote:
 Hi Hanya, good to see more developers on the PMC.

[Pendantic mode]

Please note that the PMC is the Project Management Committee.

Not all developers who work on the AOO project are members of the PMC.

 Hope your insights will keep this project afloat and thriving.

Indeed.


 On Sat, Dec 28, 2013 at 3:30 PM, Vladislav Stevanovic 
 stevanovicvladis...@gmail.com wrote:

 Hello Tsutomu!
 We do not know each other, but I noticed your effort in Ooo (Aoo) project
 (as Hanya). I am glad that you are here (in formal way), where you belong
 for a long time...

 Regards,
 Wlada



 2013/12/28 Hagar Delest hagar.del...@laposte.net

  Welcome!
  Nice to see another regular forum user (contributing a lot for macros)
  part of the PMC.
  :-)
 
  Hagar
 
 
  Le 28/12/2013 10:38, Andrea Pescetti a écrit :
 
   The Project Management Committee (PMC) for Apache OpenOffice has asked
  Tsutomu Uchino to become a committer and we are pleased to announce
  that he has accepted and taken the ID hanya.
 
  A warm welcome to Tsutomu !
 
  Regards,
 
  Andrea, on behalf of the Apache OpenOffice PMC
 
  -
  To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
  For additional commands, e-mail: dev-h...@openoffice.apache.org
 
 
  -
  To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
  For additional commands, e-mail: dev-h...@openoffice.apache.org
 
 




 --
 Alexandro Colorado
 Apache OpenOffice Contributor
 http://www.openoffice.org
 882C 4389 3C27 E8DF 41B9  5C4C 1DB7 9D1C 7F4C 2614

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Downloading problems with some Browsers

2013-12-31 Thread sebb

On 30 December 2013 16:25, Rory O'Farrell ofarr...@iol.ie wrote:
 On the en-Forum there have been some reports of Download interrupted, from 
 memory these are mostly from users of Chrome browser. We normally advise use 
 (either permanently or on a one off basis) of Firefox.

Not sure that is wise.

Firefox may not suffer from interrupted downloads, however it does
suffer from silent partial downloads. There is a long-standing bug on
this, which shows little sign of being addressed:

https://bugzilla.mozilla.org/show_bug.cgi?id=237623

I have experienced this bug myself with some servers in the past.

When I used tools such as wget and curl these handled temporary breaks
in the download and resume correctly - or gave up with an error.

Whereas under some conditions Firefox just terminated the download
with no error being reported.

I have not experienced it recently; however that may be more due to
improvements in server behaviour rather than a fix to the Firefox bug.

Hashes and sigs can of course be used to check whether the download
has completed OK.

 A current case is
 https://forum.openoffice.org/en/forum/viewtopic.php?f=15t=66618#p296227
 which links to a possible flaw in the file server.

 The suggested flaw may be worth investigating.

 --
 Rory O'Farrell ofarr...@iol.ie

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Reporting a problem with the OpenOffice website

2013-12-28 Thread sebb

On 27 December 2013 09:14, Andrea Pescetti pesce...@apache.org wrote:
 On 26/12/2013 Andrea Pescetti wrote:

 On one browser where I can reproduce the amusing pattern wiki works /
 forum doesn't, the certificate manager contains the
 *.apache.org certificate for openoffice.org:443


 We now have the final answer from the Apache Infrastructure team: it all
 depends on the client (the browser and operating system you are using).

 If you don't have SNI support (most common problematic case: Internet
 Explorer 8 on Windows XP) you will get the warning. This is due to how the
 certificate is configured and apparently can't be changed. See
 https://issues.apache.org/jira/browse/INFRA-7131

 So (but this is a personal remark, not checked with Infra) also the fact
 that on some systems the issue appears intermittently could be due to broken
 SNI support (like: your client uses the first certificate it sees, and if it
 is the right one everything is OK, while if it is the wrong one you get the
 warning).

AIUI the problem is that the same physical host (IP address) is used
for www.apache.org and wiki.openoffice.org.

The HTTP server has to decide which certificate to serve to the
browser; it does this based on the SNI (server name indication)
provided by the client.
Given that SNI was invented, I assume there is no other way for the
server to know the target host name without it, i.e. it needs to
happen before the Host: header is sent.

AFAIK the server does not return multiple certificates.

This would suggest that the behaviour should be predictable - WinXP+IE
does not work, just about every other browser does work.
I've certainly not noticed a variation.

 This probably closes the conversation.

Well, it might be worth documenting this on the Wiki somewhere so
people can be directed to it if they ask.


 Regards,
   Andrea.

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Reporting a problem with the OpenOffice website

2013-12-26 Thread sebb

On 26 December 2013 10:28, Andrea Pescetti pesce...@apache.org wrote:
On 23/12/2013 sebb wrote:

...
http://wiki.services.openoffice.org/wiki/Documentation/FAQ/Installation

Confirm with IE, which reports a certificate for *.apache.org (Thawte)

I've seen it happening from time to time, regardless of the browser I use,
but randomly: at times, loading a wiki page gives the certificate mismatch
error, but I am unable to reproduce it consistently. Almost always the
certificate is correct and the browser does not complain.

Today I could investigate it better. At the time of writing this, if I open
http://wiki.services.openoffice.org/wiki/Documentation/FAQ/Installation
with any of the browsers I commonly use, the certificate I get is the right
one (for *.openoffice.org, issued 11 June 2013). Though, if I fire up an old
Konqueror I get the *.apache.org certificate, issued on 20 December 2011,
and thus the mismatch warning.

Experience with wget is quite complex too.

With wget 1.12:
$ wget https://wiki.openoffice.org/wiki/Documentation/FAQ/Installation
ERROR: certificate common name “*.apache.org” doesn’t match requested host
name “wiki.openoffice.org”.

I get the same with 1.11.4 on WinXP.

With wget 1.14:
$ wget https://wiki.openoffice.org/wiki/Documentation/FAQ/Installation
[works as expected]

On minotaur, I get

ERROR: cannot verify wiki.openoffice.org's certificate, issued by
'/C=US/O=DigiCert Inc/CN=DigiCert Secure Server CA':
Unable to locally verify the issuer's authority.

Adding --no-check-certificate allows wget to work with a Warning:

WARNING: cannot verify wiki.openoffice.org's certificate, issued by
'/C=US/O=DigiCert Inc/CN=DigiCert Secure Server CA':
Unable to locally verify the issuer's authority.

With wget 1.13.4 from ooo-wiki2-vm (so this is repeatable by others):
$ wget https://wiki.openoffice.org/wiki/Documentation/FAQ/Installation
ERROR: certificate common name `*.apache.org' doesn't match requested host
name `wiki.openoffice.org'.

To avoid misunderstandings: this page does not currently contain any
insecure content. This is a separate problem, see below.

Might it be that the *.apache.org certificate is served as a fallback to
some older clients that for some reason do not support the *.openoffice.org
one?

Could be true.

Does not happen in Firefox for me either, though I do get a shield
icon [1] indicating that there is some http: reference that needs to
be changed to https:
[IE also reports the same issue, if one ignores the cert. error]

I investigated this on the VM as far as my permissions allow. It should be
due to the hardcoded http in the file GoogleCoop/GoogleCoop.php

To work around this, I've temporarily removed the Search within the FAQs
function that was including Google CSE via http, see
https://wiki.openoffice.org/w/index.php?title=Documentation%2FFAQ%2FInstallationdiff=232487oldid=188626

Does this solve the insecure content problem for you? It does for me.

Yes, Firefox, Opera are now happy, as is IE once on gets past the cert error.

The right fix would then to change http to https in
GoogleCoop/GoogleCoop.php but I can't modify that file.

Regards,
Andrea.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Reporting a problem with the OpenOffice website

2013-12-26 Thread sebb

On 26 December 2013 20:39, jan i j...@apache.org wrote:
On 26 December 2013 19:17, sebb seb...@gmail.com wrote:

On 26 December 2013 10:28, Andrea Pescetti pesce...@apache.org wrote:
On 23/12/2013 sebb wrote:

...
http://wiki.services.openoffice.org/wiki/Documentation/FAQ/Installation

Confirm with IE, which reports a certificate for *.apache.org (Thawte)

I've seen it happening from time to time, regardless of the browser I
use,
but randomly: at times, loading a wiki page gives the certificate
mismatch
error, but I am unable to reproduce it consistently. Almost always the
certificate is correct and the browser does not complain.

Today I could investigate it better. At the time of writing this, if I
open
http://wiki.services.openoffice.org/wiki/Documentation/FAQ/Installation
with any of the browsers I commonly use, the certificate I get is the
right
one (for *.openoffice.org, issued 11 June 2013). Though, if I fire up
an old
Konqueror I get the *.apache.org certificate, issued on 20 December
2011,
and thus the mismatch warning.

Experience with wget is quite complex too.

With wget 1.12:
$ wget https://wiki.openoffice.org/wiki/Documentation/FAQ/Installation
ERROR: certificate common name “*.apache.org” doesn’t match requested
host
name “wiki.openoffice.org”.

I get the same with 1.11.4 on WinXP.

It works for me, with firefox and IE on win7, but wget 1.14 fails on the
same vm.

I have a theory an would like it confirmed

@andrea, @sebb, please have a look in certificate manager/servers, and
check is wiki.openoffice.org is there, if its in there please tell me in
which group?

I'm not sure how to check this, sorry.

With wget 1.14:
$ wget https://wiki.openoffice.org/wiki/Documentation/FAQ/Installation
[works as expected]

On minotaur, I get

ERROR: cannot verify wiki.openoffice.org's certificate, issued by
'/C=US/O=DigiCert Inc/CN=DigiCert Secure Server CA':
Unable to locally verify the issuer's authority.

Adding --no-check-certificate allows wget to work with a Warning:

WARNING: cannot verify wiki.openoffice.org's certificate, issued by
'/C=US/O=DigiCert Inc/CN=DigiCert Secure Server CA':
Unable to locally verify the issuer's authority.

To avoid misunderstandings: this page does not currently contain any
insecure content. This is a separate problem, see below.

Might it be that the *.apache.org certificate is served as a fallback to
some older clients that for some reason do not support the *.
openoffice.org
one?

Could be true.

If you trace tcp/ip, is seems there are 2 requests for certificate from
wget. I dont think the reason is the certificate itself, but more the root
chain. There was a problem similar to this when the certificate was
installed.

rgds
jan I.

I investigated this on the VM as far as my permissions allow. It should
be
due to the hardcoded http in the file GoogleCoop/GoogleCoop.php

To work around this, I've temporarily removed the Search within the
FAQs
function that was including Google CSE via http, see

https://wiki.openoffice.org/w/index.php?title=Documentation%2FFAQ%2FInstallationdiff=232487oldid=188626

Does this solve the insecure content problem for you? It does for me.

Yes, Firefox, Opera are now happy, as is IE once on gets past the cert
error.

The right fix would then to change http to https in
GoogleCoop/GoogleCoop.php but I can't modify that file.

Regards,
Andrea.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Reporting a problem with the OpenOffice website

2013-12-23 Thread sebb

On 18 December 2013 14:39, Donald Whytock dwhyt...@apache.org wrote:
 On Tue, Dec 17, 2013 at 10:24 PM, grumpy rp...@q.com wrote:


 From this page:
 http://www.openoffice.org/download/common/java.html

 taking the link Installation FAQ which goes here:
 http://wiki.services.openoffice.org/wiki/Documentation/FAQ/Installation

 results in this message:

 There is a problem with this website's security certificate.

  The security certificate presented by this website was issued for a
 different website's address.


 Confirmed in IE8.  Though I don't get it in Chrome.


Confirm with IE, which reports a certificate for *.apache.org (Thawte)

Does not happen in Firefox for me either, though I do get a shield
icon [1] indicating that there is some http: reference that needs to
be changed to https:
[IE also reports the same issue, if one ignores the cert. error]

Firefox reports a different certificate, *.openoffice.org (Digicert)

Looks like the server is serving up the wrong cert for IE.

Opera does not complain.

[1] 
https://support.mozilla.org/en-US/kb/how-does-content-isnt-secure-affect-my-safety?as=uutm_source=inproduct

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Reporting a problem with the OpenOffice website

2013-12-23 Thread sebb

On 23 December 2013 14:05, sebb seb...@gmail.com wrote:
On 18 December 2013 14:39, Donald Whytock dwhyt...@apache.org wrote:
On Tue, Dec 17, 2013 at 10:24 PM, grumpy rp...@q.com wrote:

From this page:
http://www.openoffice.org/download/common/java.html

taking the link Installation FAQ which goes here:
http://wiki.services.openoffice.org/wiki/Documentation/FAQ/Installation

results in this message:

There is a problem with this website's security certificate.

The security certificate presented by this website was issued for a
different website's address.

Confirmed in IE8. Though I don't get it in Chrome.

Confirm with IE, which reports a certificate for *.apache.org (Thawte)

Firefox reports a different certificate, *.openoffice.org (Digicert)

Looks like the server is serving up the wrong cert for IE.

Opera does not complain.

Have found out that Opera downgrades the site to insecure, but this is
not very obvious.
This appears to be because of the insecure (http:) content.

It looks like Firefox suppresses the insecure content (with an option
to allow it) and treats the page as secure.

However Opera allows the insecure content, but treats the resulting
page as insecure. Not easy to spot.

The solution in both cases should be simple - fix the insecure parts
of the page.
Note: this does not mean all links, only ones that are part of the
page, e.g. script src=http://;

Of course this won't fix the IE certificate issue, but once that is
fixed, IE will complain about the insecure content...

[1]
https://support.mozilla.org/en-US/kb/how-does-content-isnt-secure-affect-my-safety?as=uutm_source=inproduct

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: JIRA instead of Bugzilla?

2013-11-08 Thread sebb

On 8 November 2013 00:43, Kay Schenk kay.sch...@gmail.com wrote:
 On Thu, Nov 7, 2013 at 2:28 PM, Rob Weir robw...@apache.org wrote:

 On Tue, Oct 22, 2013 at 11:14 AM, janI j...@apache.org wrote:
  On 22 October 2013 16:41, Oliver-Rainer Wittmann
  orwittm...@googlemail.comwrote:
 
  Hi,
 
  On 22.10.2013 10:04, Rainer Bielefeld wrote:
 
  Hi,
 
  it's really daunting that nobody cares!
 
 
  I care, but only as a user of our Bugzilla instance being frustrated
 when
  I need Bugzilla in the morning (European time zone).
 
  It seems that we need to involve ASF Infra as I do not believe that this
  scheduled outage every day is controlled by us.
 
 
  Just checked, there are no outstanding issues with aoo-bz, except its
 very
  slow because it has not yet had the db moved. The scheduled outage is
  unknown, but could be the backup which runs very early morning (europe
  time).
 
  rgds
  jan I.
 
  Ps. once again it was suggested that we move to jira.

By whom?

AFAIK, JIRA requires more resources than Bugzilla.

 
 

 So for argument's sake, and speaking purely hypothetically, what are
 the pros and cons of moving to JIRA?  It is worth at least discussing
 whether this would be something worth looking into.


 ==Con==

 1. Assume migration of new bugs would be imperfect.  But maybe not so
 bad.  We have many attachments, comments, etc., but the comments are
 all plain text, not rich text.

 2. New tool to learn for volunteers.  But many of us know JIRA also.

 3. Would require some time to migrate, from Infra and from BZ admins

 4. ???


The handling of attachments is poor comnpared with Bugzilla (which is
very straightforward).
There does not seem to be a way to add comments at the same time as an
attachment.

Marking an issue as a duplicate requires a separate operation to link
to the duplicate issue.

AFAIK importing issues requires quite a long down-time.
This would presumably affect both Bugzilla and JIRA, as Bugzilla would
need to be be read-only for the duration.
It may be quicker if JIRA is set up as a separate instance.

 ==Pro==

 1. Performance/stability?   I assume that is why Infra was suggesting this?


AFAIK, JIRA needs more resources and is less stable than Bugzilla.

Make sure this information is checked with Infra before it is relied on.

 2. Agile features that help with release planning


It is easier to find issues that relate to a particular version.

 3. Anything else ???

UI looks nicer, but has been known to change without warning between releases.


 Regards,

 -Rob

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 This is neither a pro nor a con, just a comment.

 We make use of saved searches and can make them public -- I did a quick
 look at the ref for Jira and I can't tell how Jira's mechanisms  work in
 this fashion.

 I have used both also, but did not do  anything very sophisticated with
 Jira in the past.


 --
 -
 MzK

 “Unless someone like you cares a whole awful lot,
  Nothing is going to get better. It's not.”
   -- Dr. Seuss, The Lorax

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: RTF

2013-10-20 Thread sebb

On 19 October 2013 11:17, Patrick pdema...@yahoo.com wrote:
 Hello,

 I've received this .rtf file (attachment) with a little picture on the left
 side that i can't see with Open Office. I'm using Ubuntu because I hate
 Microsoft and all their products :) ... So, in conclusion, my friend (on
 Microsoft Office Word) opened this file and the picture was here. I'm asking
 now, could this problem be resolved in the future Open Office release?

Attachments are stripped from this mailing list.

I suggest you create a Bugzilla issue [1] where you can describe the
problem (please include details of the version you are using), and
then attach the RTF file that exhibits the problem. Note that Bugzilla
issues are public (as is this mailing list), so please ensure that the
RTF file does not contain any private/sensitive material before
attaching it.

Thanks!

[1] http://openoffice.apache.org/bug-tracking.html

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [Proposal] Update Icons for AOO 4.1

2013-10-17 Thread sebb

On 10 October 2013 04:01, Samer Mansour samer...@gmail.com wrote:
 Hello,

 I'm proposing to have the icons (and related assets) updated for AOO 4.1
 release. I would like to take the responsibility to see this gets done.

 Here is the icons that need to be updated:
 https://cwiki.apache.org/confluence/display/OOOUSERS/AOO4.1+-+Desktop+Icons

 Related asset:
 https://cwiki.apache.org/confluence/display/OOOUSERS/AOO4.1+-+Application+And+Launcher

The Splash screen needs some changes; please see:
https://issues.apache.org/ooo/show_bug.cgi?id=123491

Also, the About screen says:

This product was created by the OpenOffice community.
I think that should say:
This product was created by the Apache OpenOffice community.

 You can see two examples of ideas on the first link. You can also suggest
 your own ideas verbally by commenting on the wiki page or visually by
 attaching it to the wiki. (Please don't reply ideas in this mailing list,
 let us know if you have trouble)

 I am putting a deadline for the idea submission and discussion in 30 days,
 November 9th, 2013.
 Once we reach the deadline, we will have a separate discussion for
 optimizing for the best user experience.

 This is not a contest or a call for public proposals. This is regular,
 needs to be done, no-bikeshedding-please work. When the deadline Nov 9th
 arrives, if there is more than one viable solution, we will try to reach
 consensus without a end user vote.
 eg. We will not be doing what we did with the logo, that was a special case
 because it is the face of AOO.

 Samer Mansour

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Where can I find the new AOO4-Logo as a vector format?

2013-09-27 Thread sebb

On 27 September 2013 10:11, David Gerard dger...@gmail.com wrote:
 On 27 September 2013 09:23, Jörg Schmidt joe...@j-m-schmidt.de wrote:

 But a note:
 The M in TM is shown cut off and the representation of TM is different 
 in Internet Explorer and Firefox, once serifs, once without serifs, at an 
 official logo should not be.



 Display artifact in Firefox. It's fine in Inkscape or on Wikimedia Commons:

 http://en.wikipedia.org/wiki/File:Aoo4-main-tm-logo-rgb.svg


For 
http://www.openoffice.org/marketing/art/galleries/logos/main/aoo4-main-tm-logo-rgb.svg

I get the same display artifact in Opera 12.16 / Chrome 29.0.1547.76
m / and Firefox 23.0.1

However the Wikipedia logo looks OK to me in all 3 browsers, so there
is something else happening here.

IE 8 does not seem to want to display SVG for me.

 - d.

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Mails from iss...@openoffice.apache.org

2013-09-13 Thread sebb

FWIW: AFAICT there are currently 123 251 bugs in the AOO Bugzilla instance.

On 12 September 2013 19:33, sebb seb...@gmail.com wrote:
 On 12 September 2013 18:53, Rob Weir robw...@apache.org wrote:
 On Thu, Sep 12, 2013 at 12:30 PM, Mechtilde o...@mechtilde.de wrote:
 Hello

 What happens? I get nearly 6500 mails today with date 07.08.2013.

 My mail provider ask me, if all things are right ;-)

 Kind regards


 I haven't been hacked.

 These issues were changed back in August as you can see from their
 modification dates.   I know to disable mail notifications when doing
 bulk changes.  I did so in this case.

 To do that I set mail_delivery_method to none, which the Bugzilla
 admin documentation describes as:

 'none' will completely disable email. Apache OpenOffice (AOO)
 Bugzilla continues to act as though it is sending mail, but nothing is
 sent or stored. 

 However, when doing routine maintenance, running the sanity check
 which checks for inconsistencies in the database, it reports on issues
 that have unsent notifications and offers to send them.   I assumed
 (falsely) that this the previous none mail processing behaved as
 described, namely that it acted the same as sending mail, but with
 nothing sending.  In other words, that it would record that a
 notification had been meant.

 So what we're seeing are old notifications that were prevented before,
 but are now being sent.

 I've set the mail_deliver_method to none for now, which hopefully will
 stop the notifications for now.

 Might be worth taking this up with Infra in case there is a better way
 to do bulk updates in future.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Mails from iss...@openoffice.apache.org

2013-09-12 Thread sebb

On 12 September 2013 18:53, Rob Weir robw...@apache.org wrote:
 On Thu, Sep 12, 2013 at 12:30 PM, Mechtilde o...@mechtilde.de wrote:
 Hello

 What happens? I get nearly 6500 mails today with date 07.08.2013.

 My mail provider ask me, if all things are right ;-)

 Kind regards


 I haven't been hacked.

 These issues were changed back in August as you can see from their
 modification dates.   I know to disable mail notifications when doing
 bulk changes.  I did so in this case.

 To do that I set mail_delivery_method to none, which the Bugzilla
 admin documentation describes as:

 'none' will completely disable email. Apache OpenOffice (AOO)
 Bugzilla continues to act as though it is sending mail, but nothing is
 sent or stored. 

 However, when doing routine maintenance, running the sanity check
 which checks for inconsistencies in the database, it reports on issues
 that have unsent notifications and offers to send them.   I assumed
 (falsely) that this the previous none mail processing behaved as
 described, namely that it acted the same as sending mail, but with
 nothing sending.  In other words, that it would record that a
 notification had been meant.

 So what we're seeing are old notifications that were prevented before,
 but are now being sent.

 I've set the mail_deliver_method to none for now, which hopefully will
 stop the notifications for now.

Might be worth taking this up with Infra in case there is a better way
to do bulk updates in future.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Fw: Reporting a problem with the OpenOffice website

2013-09-11 Thread sebb

I think Max meant to write [Microsoft] Word file and by explorer
he meant Windows Explorer not Internet Explorer.

You wrote that you were having problems opening certain existing documents.

Using Windows Explorer, right-click on one such file.
Choose Open With ... and then Choose program 
This will show a list of programs (may take a short while).
Select the program which you want Windows to use to open the file in future.
For documents, choose Microsoft Word
For spreadsheets, Microsoft Excel
e.t.c.

You will need to do this once for each different type of file
(document, spreadsheet, presentation, etc.)

==

If you are *still* having problems, right-click and select Properties.
Near the top of the properties display there should be a section with
the entries:

Type of File:
Opens with:

Make a note of the settings and ask again.


On 11 September 2013 03:10, Mel Goddard mdgoddard3...@i-zoom.net wrote:
 ??


 -Original Message- From: Mel Goddard
 Sent: Tuesday, September 10, 2013 9:40 AM

 To: Max Merbald
 Subject: Re: Reporting a problem with the OpenOffice website

 I'm afraid that you have me at a disadvantage.  I can find no word file in
 my internet 'explorer' [msm.com]
 Got any other ideas?
 Please excuse the time delays, I have many other things that need my
 attention, besides 'playing' with a computer; in our household, (as others)
 a computer is a convenience of communication, and not a 'way of life'.


 Mel

 -Original Message- From: Max Merbald
 Sent: Monday, September 09, 2013 3:53 PM

 To: dev@openoffice.apache.org
 Cc: Mel Goddard
 Subject: Re: Fw: Reporting a problem with the OpenOffice website


 Hi Mel,

 that sounds like you accidentially assigned OpenOffice to open all word
 dokuments for you when you installed it. To return to normal,
 right-click on a word file in the explorer, then chose open with and
 then Chose default program You'll get a list of programs to chose from
 and Word should be among them. You should be back to normal then.

 Max


 Am 09.09.2013 15:47, schrieb Mel Goddard:

 Before I do download it, perhaps you can tell me WHY I would need
 OpenOffice  to open my files for me.
 That's akin to me telling you that you cannot drive your car unless I turn
 the key to start it for you.
 I have now wasted almost a full day to try to solve a problem that
 prevents me from doing a 30 second job.

 Mel

 -Original Message- From: Mel Goddard
 Sent: Monday, September 09, 2013 9:00 AM
 To: Max Merbald
 Subject: Re: Reporting a problem with the OpenOffice website

 Max;

 The problem is, that I'm trying to open an existing document to add to it.
 Your program is preventing me from doing it.
 I've never had any problem before, until a couple of days ago.
 So since I've never even 'heard'  of you before; I did uninstall it, and
 then 'somebody' else is trying to get in here.
 All I want to do is to access an existing document! That's all.

 My computer 'guru' tells me that you're legit, so I'm going to reinstall
 your program and see: What the Hell.

 I'll let you know of the follow up.
 Mel
 -Original Message- From: Max Merbald
 Sent: Monday, September 09, 2013 8:38 AM
 To: dev@openoffice.apache.org
 Cc: Mel Goddard
 Subject: Re: Reporting a problem with the OpenOffice website

 Hello Mel,

 unfortunately you didn't tell us what was your actual problem with Open
 Office, and that makes it rather difficult to help you. If you do not
 wish to have OpenOffice on your computer just uninstall it. I don't
 believe that yelling would help you.

 Max


 Am 09.09.2013 03:30, schrieb Mel Goddard:

 GET OUT OF MY COMPUTER!
 I DIDN'T ASK FOR YOU, I DON'T NEED YOU, AND YOU ARE INTERFERING WITH MY
 WORK ON THIS THING!
 GET OUT NOW!
 mdgoddard3...@i-zoom.net

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org



 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org



 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Menu language problem with Wiki

2013-09-11 Thread sebb

On 11 September 2013 17:49, sebb seb...@gmail.com wrote:
 Just clicked on OpenOffice Wiki: Documentation area link under
 http://www.openoffice.org/support/#community-support.

 The page that is displayed has English body text, but Dutch (I think)
 for the menu bar.

 Clicking other languages (under Ander tale) changes the body text,
 but not the menu.

 However, when I try the same in Opera and Chrome, the menu stays in English.

 Very odd.

Sorry, my fault - I'd forgotten I'd been playing with the languages
setting in Firefox.
It was actually Afrikaans.

However, given that the menu is available in languages other than
English, why does it not change when the user selects a different
language?

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Menu language problem with Wiki

2013-09-11 Thread sebb

Just clicked on OpenOffice Wiki: Documentation area link under
http://www.openoffice.org/support/#community-support.

The page that is displayed has English body text, but Dutch (I think)
for the menu bar.

Clicking other languages (under Ander tale) changes the body text,
but not the menu.

However, when I try the same in Opera and Chrome, the menu stays in English.

Very odd.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [CONF] Apache OpenOffice Community Louis Vuitton Evora 100% Authentic 80% Off Free Shipping

2013-09-06 Thread sebb

On 6 September 2013 17:07, Alexandro Colorado j...@oooes.org wrote:
 We are getting some spam attacks on the cWiki, wonder if we can do
 some scanning to detect similar pages, and also remove this nick.

On the Moin Wikis, anyone can create a Wiki account, but that does not
grant them write access.
They then have to apply to the dev list to get added to the
ContributorsGroup page. [1]
If anyone abuses their permissions (AFAIK that has not happened) the
login can easily be removed from the page.

[1] 
http://wiki.apache.org/general/OurWikiFarm#per_wiki_access_control_-_tighten_your_wiki_just_a_little.2C_benefit_just_a_lot
 -- Forwarded message --
 From: lily regland (Confluence) conflue...@apache.org
 Date: Fri, 6 Sep 2013 06:08:00 + (UTC)

Please don't forward spam unless it is essential to the reply, and
then remove as much as possible first.
Thanks!

snipped spam

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

User questions on dev mailing list

2013-09-05 Thread sebb

The dev list sees the occasional thread which IMO really belongs on
the user list/forum

Although it is convenient for the poster to have the question answered
on the dev list, I wonder if that is the best approach overall.

The disadvantages of not referring the poster to the user list are several:

- other users don't benefit from the thread
- the dev list is cluttered with the off-topic threads
- the poster may continue to use the dev list rather than the user
list/forum where there are lots more potential responders

Just a thought.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: User questions on dev mailing list

2013-09-05 Thread sebb

On 5 September 2013 12:15, Peter Junge peter.ju...@gmx.org wrote:
 On 9/5/2013 5:24 PM, sebb wrote:

 The dev list sees the occasional thread which IMO really belongs on
 the user list/forum

 Although it is convenient for the poster to have the question answered
 on the dev list, I wonder if that is the best approach overall.

 The disadvantages of not referring the poster to the user list are
 several:

 - other users don't benefit from the thread
 - the dev list is cluttered with the off-topic threads
 - the poster may continue to use the dev list rather than the user
 list/forum where there are lots more potential responders


 +1 to all of the above. Unfortunately it's not possible to already
 redirecting to the user list during moderating. I would guess at least 80%
 of user topics here are moderated in. Didn't we discuss mechanisms to
 restrict postings to dev@ a couple of months ago?

A moderator can always reject a posting to the developer list with a
polite message asking them to subscribe/post elsewhere.

For example:

%%% Start comment
This message is off-topic for this mailing list.

Please subscribe to the Apache OpenOffice user mailing list and post
there - thanks.
See:
http://openoffice.apache.org/mailing-lists.html
%%% End comment


 Just a thought.


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: User questions on dev mailing list

2013-09-05 Thread sebb

On 5 September 2013 12:00, Rob Weir robw...@apache.org wrote:
 On Thu, Sep 5, 2013 at 5:24 AM, sebb seb...@gmail.com wrote:
 The dev list sees the occasional thread which IMO really belongs on
 the user list/forum

 Although it is convenient for the poster to have the question answered
 on the dev list, I wonder if that is the best approach overall.

 The disadvantages of not referring the poster to the user list are several:

 - other users don't benefit from the thread
 - the dev list is cluttered with the off-topic threads
 - the poster may continue to use the dev list rather than the user
 list/forum where there are lots more potential responders

 Just a thought.


 This is all true.  We also get user questions to the private mailing
 list, and via the press alias address.  And we also get product
 support questions to the Bugzilla admin alias and the ezmlm admin
 alias.

 But I don't think we actively encourage users to post questions to do
 these things.  But it is a puzzle why this happens.

 The main contact page for the project, on the footer of every web page is 
 this:

 http://www.openoffice.org/contact_us.html

 We've adjusted this over time, to make sure that the support options
 were first and more prominent.  But we still get misdirected emails,
 to the dev, private, etc. lists.  But with a million downloads a week
 there will always be some small number of users who don't read that
 page carefully, or maybe think they will get a faster response if they
 send to a different list.

[And they are proved right when their off-topic questions are promptly
answered ...]

The second heading currently says:

If you want to contact the Apache OpenOffice developer team...

which may explain some of the misdirected questions.
[Who better to ask than a developer if there is a problem?
Maybe I'll get a faster response?]

I think that section should have some explanation as to when it is
appropriate to contact the team directly.
Possibly move it further down the page as well.

 -Rob

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: User questions on dev mailing list

2013-09-05 Thread sebb

A related issue is: why so many emails with the (irrelevant) subject:

Reporting a problem with the Open Office website


I assume there must be a mailto: link somewhere that is producing this.
The surrounding text may need to be tweaked to reduce these.


On 5 September 2013 22:48, Kay Schenk kay.sch...@gmail.com wrote:
 On Thu, Sep 5, 2013 at 6:29 AM, sebb seb...@gmail.com wrote:

 On 5 September 2013 12:00, Rob Weir robw...@apache.org wrote:
  On Thu, Sep 5, 2013 at 5:24 AM, sebb seb...@gmail.com wrote:
  The dev list sees the occasional thread which IMO really belongs on
  the user list/forum
 
  Although it is convenient for the poster to have the question answered
  on the dev list, I wonder if that is the best approach overall.
 
  The disadvantages of not referring the poster to the user list are
 several:
 
  - other users don't benefit from the thread
  - the dev list is cluttered with the off-topic threads
  - the poster may continue to use the dev list rather than the user
  list/forum where there are lots more potential responders
 
  Just a thought.
 
 
  This is all true.  We also get user questions to the private mailing
  list, and via the press alias address.  And we also get product
  support questions to the Bugzilla admin alias and the ezmlm admin
  alias.
 
  But I don't think we actively encourage users to post questions to do
  these things.  But it is a puzzle why this happens.
 
  The main contact page for the project, on the footer of every web page
 is this:
 
  http://www.openoffice.org/contact_us.html
 
  We've adjusted this over time, to make sure that the support options
  were first and more prominent.  But we still get misdirected emails,
  to the dev, private, etc. lists.  But with a million downloads a week
  there will always be some small number of users who don't read that
  page carefully, or maybe think they will get a faster response if they
  send to a different list.

 [And they are proved right when their off-topic questions are promptly
 answered ...]

 The second heading currently says:

 If you want to contact the Apache OpenOffice developer team...

 which may explain some of the misdirected questions.
 [Who better to ask than a developer if there is a problem?
 Maybe I'll get a faster response?]

 I think that section should have some explanation as to when it is
 appropriate to contact the team directly.
 Possibly move it further down the page as well.


 Maybe the contact the Apache OpenOffice developer team needs to be
 removed from the Contact Us page entirely.  Honestly, given the context,
 I'm not sure this belongs here. Maybe delete that section and add a heading
 at the end called Mailing Lists and route them to the project mailing list
 page. That might encourage users to  make a better choice for contacting.




  -Rob
 
  -
  To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
  For additional commands, e-mail: dev-h...@openoffice.apache.org
 
 
  -
  To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
  For additional commands, e-mail: dev-h...@openoffice.apache.org
 

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org




 --
 -
 MzK

 Truth is stranger than fiction, but it is because Fiction is obliged
  to stick to possibilities. Truth isn't.
  -- Following the Equator, Mark Twain

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: 4.0.1 release and distribution.

2013-09-03 Thread sebb

On 3 September 2013 08:05, Andrea Pescetti pesce...@apache.org wrote:
 On 03/09/2013 sebb wrote:

 Are you suggesting pointing users to the archives for current sigs and
 hashes?
 If so, I don't think that's a good idea.


 Maybe it isn't, but using dist is worse:
 - Everything that lives on dist is temporary
 - You end up with tons on liks to update (and some cannot really be fixed,
 such as URLs in PDFs obtained from security announcements, or URLs
 copy-pasted from them into blog posts)

That is unlikely to apply to sigs and hashes.

 - The bandwidth required for downloading hashes is obviously very tiny

But the user population is huge compared with other ASF products.

 So I really prefer to use permanent URLs (from archive) for hashes.

AFAIK all other ASF products link to the ASF mirrors for current releases.


 Updating the links to point to the archive server should be done for
 both artifacts and sigs/hashes at the same time.


 Artifacts in our case are much more manageable since we have one link in
 other.html ; in that case, it's clearly best to use the dist link until the
 version is available and the archive link when it is no longer available on
 dist.

I don't follow how the sigs/hashes are any different from the
downloads they protect.
Either they are current, and on the ASF mirrors, or they have been
superseded and are on the archive host.


 Regards,
   Andrea.

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: 4.0.1 release and distribution.

2013-09-02 Thread sebb

On 2 September 2013 09:26, Jürgen Schmidt jogischm...@gmail.com wrote:
 On 9/1/13 4:45 PM, janI wrote:
 On 1 September 2013 15:31, Rob Weir robw...@apache.org wrote:

 On Sun, Sep 1, 2013 at 5:42 AM, janI j...@apache.org wrote:
 On 1 September 2013 11:27, janI j...@apache.org wrote:

 Hi.

 I had a talk on #asfinfra today, regarding our upcomming 4.0.1 release.

 Sync on mirrors takes about a week, and the mirror can in general not
 hold
 4.0 and 4.0.1


 Is this a change?  The current published advice is that the mirrors
 take no more than 24 hours to sync:

 https://www.apache.org/dev/release-publishing.html#distribution


 no actually not, the problem is the size of our distribution, with 4.0 it
 took 8 days and one Chinese mirror has not updated fully yet.



 Therefore the current suggestion is to
 a) remove 4.0 from mirrors GA - 1week = 12 september
 b) update 4.0.1 to mirrors GA = 19 september.

 The downside is that mirrors will have the 4.0 ready for download for
 upto
 a week.


 The problem is we don't know for certain a GA date a week in advance.
 We can just estimate.  But as we saw with 4.0.0, a last minute defect
 can delay things by a week or more.


 We have a choice, we can wait until GA (as we did last time, where it took
 several weeks after GA before downloads were in place), or take a chance. I
 opt for the chance, I think it is important to have the mirrors in place
 when we announce our release.

 I think most of the downloads are going over SourceForge so having the
 mirrors a little bit later in sync shouldn't be a big problem.

 I would prefer one simple to follow approach. And the synch is done in a
 way that is suitable for the mirrors.

I think that rules out links, because of the problem of maintaing
parity between artifacts and sigs/hashses.

 A further question how do we count the downloads from the ASF mirrors or
 can we count them at all?


http://www.apache.org/dev/release.html#downloads





 So in practice this means that there could be more than a week where
 4.0.0 is not on the mirrors.  Maybe this is not a problem?


 I hope not, we loose some downloads, but hopefully the users will try again.



 The two things to watch out for (and you have probably already
 considered these, but  I'll mention them just in case):

 1)  We should not remove the 4.0.0 hashes and signature files from
 /dist.  These are referenced even when the binaries are downloaded
 from SourceForge.


 @henkp: can you make sure of that, please


 2) We need to make sure SourceForge is not rsyncing from /dist and
 mirror the 4.0.0 removal.


 I assume that will be the case.


 And I assume 4.0.1 goes to archive then?


 If I remember right, we (andrea) have to move it to the archive (I presume
 you mean 4.0).

 no, it goes automatically to archive as far as I know

http://www.apache.org/dev/release.html#how-to-archive

and

http://www.apache.org/dev/release.html#when-to-archive
[For projects with smaller footprints the old and new releases should
normally overlap whilst the mirrors catch up.]

 Juergen


 rgds
 jan I.



 -Rob


 An alternative suggestion, is to rename 4.0 to 4.0.x on the mirrors and
 have a 4.0 symlink pointing at 4.0.x.

 That way, we simply replace the 4.0.x file, after GA, and mirrors do not
 have a time without a package.

 I personally like the rename idea, so can we do lazy consensus on that ?

 I have updated issue 6654, and Henkp is copied on this mail.

 rgds
 jan I


 Second try, sorry for the first mail.

 I had a talk on #asfinfra today, regarding our upcomming 4.0.1 release.

 Sync on mirrors takes about a week, and the mirror can in general not
 hold
 4.0 and 4.0.1

 Therefore the current suggestion is to
 a) remove 4.0 from mirrors GA - 1week = 12 september
 b) update 4.0.1 to mirrors GA = 19 september.

 The downside is that mirrors will have the 4.0 ready for download for
 upto
 a week (SF will be faster).

 For 4.1 we should consider an alternative way.

 Use 4.1.x on the mirrors and have a 4.1 symlink pointing at 4.1.x.

 That way, we simply replace the 4.1.x file, after GA, and mirrors do not
 have a time without a package.


 That's an interesting approach that could work.

 -Rob

 I have updated issue 6654, and Henkp is copied on this mail.

 rgds
 jan I




 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [discussion] smaller footprint on dist.

2013-09-02 Thread sebb

On 2 September 2013 09:33, Jürgen Schmidt jogischm...@gmail.com wrote:
 On 9/2/13 12:55 AM, sebb wrote:
 On 1 September 2013 23:48, janI j...@apache.org wrote:
 Hi.

 I am considering how we can have a smaller footprint on dist and mirrors.

 One of the ideas is to have a pure exe, pure language oxt and pure
 dictionaries on the physical disk, and the have a more intelligent download
 page.


 Would it be possible (and preferable) to make a download page, where the
 user.

 a) selected version (exe)
 b) selected (multiple) languages (language pack without dictionary)
 c) selected (mutiple) dictionaries.

 The choices should be combined into a filename == exe+lang(s)+dict(s).
 Which is sent to the server as a download request.

 On the server we would have a backend script that packed the items together
 just like postprocess/instsetoo does today, so the user would get 1 file.

 I know how to make the backend script, but the UI could be a problem ?

 I think it is no new idea and the first question is how we handle the
 signatures?

WIth the current build, the combined packages are created upfront and
signed by the RM.
Provided that the packages are put together in the same way from the
same component parts, the sigs will still be valid.

So the RM would still need to create the combined packages, but each
could be junked after creating the sig.
However one might want to keep them in a separate area for SourceForge
which does seem to be able to handle the volume.
Or indeed use the concatenation approach if that can be combined witrh
rsync (or whatever SF uses to fetch the stuff).

Obviously hashes can be created the same way.

If a separate download / installer is used, the problem does not arise
as the parts/sigs don't change.

There's another issue with concatenation by the server: if this
requires server configuration, then mirrors will have to do it as
well.
That may be a step too far.

It would be technically possible to use a different strategy for the
ASF mirrors as opposed to SF downloads.
E.g. SF continues as now, ASF mirrors use two downloads or download/installer.
Since the vast bulk of downloads are likely to be from SF, would that
be practical?


 I just posted to a different thread much the same thoughts.

 Another way to solve UI issue might be to generate a (small)
 downloader in each language + OS combination and have that do the
 download.
 Though I've just realised that would not allow a download manager to
 be used for the main download(s).

 Again no new ideas and work for such a special downloader/instaler was
 already ongoing in former times but never made it in the public repo.

 We should take some important points into account. One click
 installation for end users is very important.
 Offline distribution is also important for users with low bandwidth.

 Juergen




 rgds
 jan I.

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org



 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: 4.0.1 release and distribution.

2013-09-02 Thread sebb

On 2 September 2013 22:45, Andrea Pescetti pesce...@apache.org wrote:
 On 01/09/2013 Henk P. Penning wrote:

 For 4.0.0 I added the binaries slowly ; some 4 to 5 languages
 a day is what the rsync servers can handle. So, it takes 5 days
 for the quick mirrors plus 2 days for the straglers. Note that
 there are some sites that are still trying to catch up.

 Soon after GA date (or earlier), refs to 4.0.0 sigs and sums should
 point to 'archive.apache.org/' (instead of www.apache.org).


 I would very much avoid introducing another bottleneck in the process.
 Voting on the release already takes 3 days, and we won't start copying
 anything before we have the final result, as we learned from 4.0. For 4.0,
 after the release was approved, we managed to have it uploaded to the
 mirrors (the SF mirrors) rather quickly, and at the same time the
 checksums/signatures, being very small files, were quickly uploaded to dist.
 This may have taken 2 days, but not more.

 With this change we would have 3 days for voting plus 7 days for copying to
 mirrors before we can announce. And the benefits would be very marginal: the
 number of users who download from the Apache mirrors is negligible.

 So, in short, if this can be done in a way that does not slow down the
 post-approval upload period from 2 days to 7 days, fine; otherwise, it is
 better to repeat what was done for 4.0: within 2 days binaries on SF and
 checksums on dist or, even better, already on archive -which is
 automatically populated from dist but has a 24-hour delay- so that we don't
 need to update the links later;

Are you suggesting pointing users to the archives for current sigs and hashes?
If so, I don't think that's a good idea.

Note that the main release area (dist) is mirrored in US and EU,
whereas AFAIK http://archive.apache.org/dist/ is not.
AIUI http://archive.apache.org/dist/ is intended for archives only.

Updating the links to point to the archive server should be done for
both artifacts and sigs/hashes at the same time.

 then, gradually, binaries uploaded to dist
 too (this time the delay may be a matter of days instead of weeks).

 Regards,
   Andrea.


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [discussion] smaller footprint on dist.

2013-09-02 Thread sebb

On 2 September 2013 18:26, janI j...@apache.org wrote:
 On 2 September 2013 15:48, Andrea Pescetti pesce...@apache.org wrote:

 janI wrote:

 Would it be possible (and preferable) to make a download page, where the
 user.
 a) selected version (exe)
 b) selected (multiple) languages (language pack without dictionary)
 c) selected (mutiple) dictionaries.
 The choices should be combined into a filename == exe+lang(s)+dict(s).
 Which is sent to the server as a download request.
 On the server we would have a backend script that packed the items
 together
 just like postprocess/instsetoo does today, so the user would get 1 file.


 Obviously this is not for 4.0.1. What I like of this proposal is that the
 user still downloads one file, which is optimal for the user interface.

 A beginning could be to simply assemble the same downloads we have now
 (i.e., the user has no choice: he will get, say, the Italian version with
 Italian language and dictionary; only, this will be generated rather than
 pre-built).

 Then there are a lot of things to consider:

 1) Digital signatures: the assembled installer must respect them, and this
 seems hard to do.

 As far as I have been able to find out (with the good help of infra
 colleagues) is:
 - Only exe have a digital signature

That does not sound right.
All other ASF downloads (source archives, binary archives etc) have
PGP signatures, which are created by the Release Manager.

Or maybe you mean something else by digital signature?

 meaning this has no impact.

 But it DO have an impact on checksums, where we need to store all
 combinations (lots of files, each very small).



 2) Server-side processing: this would likely require some load on the
 mirrors and some infrastructure standardization. I don't know what's the
 status on Apache mirrors.

 The server side, processing would happen on our server, and the files
 would still be located on the mirrors

 Basically the server side scropt, would split the file request into
 multiple requests.

If our server does the concatenation, surely it will have to
intercept all the data from the mirror?
That would put a huge network load on the server, no?


 3) Respecting the priorities. Apache is a secondary mirror system, since
 the Apache mirrors don't have enough space/bandwidth to reliably offer
 downloads. So whatever is done should not cause technical issues with our
 primary mirror system (SourceForge), that never had space/bandwidth
 problems. Note that also the Apache Archives never reported problems so far
 about the space needed to archive old/current releases.


 The problem is only partial the space itself, much more the size of each
 release. With a distributed system (like suggested) we can independently
 release language packs, and the user sees them as integrated in the main
 AOO release.

 rgds
 jan i.


 Regards,
   Andrea.


 --**--**-
 To unsubscribe, e-mail: 
 dev-unsubscribe@openoffice.**apache.orgdev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org



-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Removing JavaScript from other.html

2013-09-02 Thread sebb

AFAICT, there is *already* a really good candidate for the
download/other.html page:

https://cwiki.apache.org/confluence/display/OOOUSERS/Development+Snapshot+Builds

The tables in the middle are easy to follow and AFAIK contain all the
required information.

No Javascript needed for the table itself.

Can't the script which was used to create that content be tweaked a
bit to create downloads/other.html?

On 2 September 2013 22:26, Marcus (OOo) marcus.m...@wtnet.de wrote:
Am 09/02/2013 10:33 PM, schrieb Alexandro Colorado:

On Mon, Sep 2, 2013 at 1:09 PM, Marcus (OOo)marcus.m...@wtnet.de wrote:

Am 09/02/2013 06:51 PM, schrieb Alexandro Colorado:

It also looks a bit 'dumb' that we dont have a link to others from the

download page within our no script tag:

Of course it doesn't make sense to refer to a page with JS when JS is
disabled in the broeser. That's why there is no link to the webpage.

downloads, need to provide a download link, that is just common sense.
There is another problem with othes.html but not offering ANY link, is no
solution.

Sure, that's the reason why we are discussing how to ensure this for more
users than now.

Marcus

On Mon, Sep 2, 2013 at 11:49 AM, Alexandro Coloradoj...@oooes.org

wrote:

On 9/2/13, Andrea Pescettipesce...@apache.org wrote:

In the last few weeks we've seen users unable to see the big
Download
button in
http://www.openoffice.org/**download/http://www.openoffice.org/download/due
to broken (but still
used) browsers that failed to parse the JavaScript correctly. And

http://www.openoffice.org/**download/other.htmlhttp://www.openoffice.org/download/other.htmlcannot
be used as a
fallback due to the same issue.

The JavaScript in other.html is used only to generate the page
content,
and the result is independent of the user's browser: as Marcus
explained, it is there for convenience in creating the page.

How are we using thenoscript tag within others? and which kind of
content is it trying to plug?

Would it make sense to do the following?

1) Add an All Apache OpenOffice downloads link in the
right-hand-side
column of
http://www.openoffice.org/**download/http://www.openoffice.org/download/near
the top: this way, we
ensure that browsers with poor JavaScript support still display the
link.

Problaby the quicker solution, still is just a patch instead of just
doing good webdev.

2) Rename other.html to other_js.html

This is a bad idea, I've seen many alternative clones being
unmantained by webmasters.

3) Modify other.html by pasting the actual download table (can be

retrieved, for example, with Firebug from other_js.html) in its HTML.

This way we add a manual step (step 3) once per release, but we can be
sure that virtually all users can download OpenOffice in all cases
(working JavaScript, no JavaScript, broken JavaScript).

Regards,
Andrea.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: 4.0.1 release and distribution.

2013-09-01 Thread sebb

On 1 September 2013 14:31, Rob Weir robw...@apache.org wrote:
 On Sun, Sep 1, 2013 at 5:42 AM, janI j...@apache.org wrote:
 On 1 September 2013 11:27, janI j...@apache.org wrote:

 Hi.

 I had a talk on #asfinfra today, regarding our upcomming 4.0.1 release.

 Sync on mirrors takes about a week, and the mirror can in general not hold
 4.0 and 4.0.1


 Is this a change?  The current published advice is that the mirrors
 take no more than 24 hours to sync:

 https://www.apache.org/dev/release-publishing.html#distribution

 Therefore the current suggestion is to
 a) remove 4.0 from mirrors GA - 1week = 12 september
 b) update 4.0.1 to mirrors GA = 19 september.

 The downside is that mirrors will have the 4.0 ready for download for upto
 a week.


 The problem is we don't know for certain a GA date a week in advance.
 We can just estimate.  But as we saw with 4.0.0, a last minute defect
 can delay things by a week or more.

 So in practice this means that there could be more than a week where
 4.0.0 is not on the mirrors.  Maybe this is not a problem?

 The two things to watch out for (and you have probably already
 considered these, but  I'll mention them just in case):

 1)  We should not remove the 4.0.0 hashes and signature files from
 /dist.  These are referenced even when the binaries are downloaded
 from SourceForge.

 2) We need to make sure SourceForge is not rsyncing from /dist and
 mirror the 4.0.0 removal.

 And I assume 4.0.1 goes to archive then?

 -Rob


 An alternative suggestion, is to rename 4.0 to 4.0.x on the mirrors and
 have a 4.0 symlink pointing at 4.0.x.

 That way, we simply replace the 4.0.x file, after GA, and mirrors do not
 have a time without a package.

 I personally like the rename idea, so can we do lazy consensus on that ?

 I have updated issue 6654, and Henkp is copied on this mail.

 rgds
 jan I


 Second try, sorry for the first mail.

 I had a talk on #asfinfra today, regarding our upcomming 4.0.1 release.

 Sync on mirrors takes about a week, and the mirror can in general not hold
 4.0 and 4.0.1

 Therefore the current suggestion is to
 a) remove 4.0 from mirrors GA - 1week = 12 september
 b) update 4.0.1 to mirrors GA = 19 september.

 The downside is that mirrors will have the 4.0 ready for download for upto
 a week (SF will be faster).

 For 4.1 we should consider an alternative way.

 Use 4.1.x on the mirrors and have a 4.1 symlink pointing at 4.1.x.

 That way, we simply replace the 4.1.x file, after GA, and mirrors do not
 have a time without a package.


 That's an interesting approach that could work.

Unfortunately, I think there are too many disadvantages.

The sigs/hashes are not mirrored. So they would be updated
immediately, but the mirrors would still have old binaries.
Even if they were mirrored, it would be impossible to ensure that the
4.1.x sig/hashes were updated at the same time as the binary.

Mirrors may end up with two copies of the file, one for the link, and
one for the actual file.
I'm not sure if this is a problem with the rsync protocol or the rysnc
settings our end or their end.

Also what happens when you want to release 4.2.x?

IMO a better solution would be to split the download so the core code
(which is by far the biggest chunk) is only produced for a single
language (or none) for each OS.  The user also has to download
whichever language packs they want.

This could potentially be done automatically with a small installation
loader that then downloads the core code and the chosen language pack.
SInce the loader would be tiny (in comparison) it would be OK to
produce customised versions for all the release combinations.
Maybe multiple downloads can even be done with HTML these days; or
maybe the server could be given a special URL that identifies the OS
and language and it generates a download containing both.

This could also have major advantages for development as well.


 -Rob

 I have updated issue 6654, and Henkp is copied on this mail.

 rgds
 jan I

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [discussion] smaller footprint on dist.

2013-09-01 Thread sebb

On 1 September 2013 23:48, janI j...@apache.org wrote:
 Hi.

 I am considering how we can have a smaller footprint on dist and mirrors.

 One of the ideas is to have a pure exe, pure language oxt and pure
 dictionaries on the physical disk, and the have a more intelligent download
 page.


 Would it be possible (and preferable) to make a download page, where the
 user.

 a) selected version (exe)
 b) selected (multiple) languages (language pack without dictionary)
 c) selected (mutiple) dictionaries.

 The choices should be combined into a filename == exe+lang(s)+dict(s).
 Which is sent to the server as a download request.

 On the server we would have a backend script that packed the items together
 just like postprocess/instsetoo does today, so the user would get 1 file.

 I know how to make the backend script, but the UI could be a problem ?

I just posted to a different thread much the same thoughts.

Another way to solve UI issue might be to generate a (small)
downloader in each language + OS combination and have that do the
download.
Though I've just realised that would not allow a download manager to
be used for the main download(s).


 rgds
 jan I.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: template troubles in xx pseudo NL website

2013-08-30 Thread sebb

On 30 August 2013 15:28, Ariel Constenla-Haile arie...@apache.org wrote:
 On Fri, Aug 30, 2013 at 08:26:51AM -0400, Rob Weir wrote:
 But one question:  what is the strategy on locale detection on the
 download page for NL pages?  It looks like you are hardcoding it to
 es, but preserving the OS detection code?

 I see:

 var NL_LANGUAGE = es;

 Is that all that is needed?  We can otherwise just reuse the same scripts?

 yes, var NL_LANGUAGE = ; will default to en-US.

 The download page, with all that javascript in it, is rather hard to
 translate. Isn't it possible to use a cgi script (server-side vs.
 client-side js)?

If the ASF servers can support server-side processing then it should
be possible to use that for the OS auto-detection as well.

However, if client-side processing is required, maybe the NL text
strings should be extracted into a JSON object, making the javascript
generic.
That would make it much easier to maintain - only a single copy.

Failing that, it should be fairly simple (?) to extract the NL strings
into JS variables and group them at the start of the page.


 Regards
 --
 Ariel Constenla-Haile
 La Plata, Argentina

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: KEYS

2013-08-30 Thread sebb

On 30 August 2013 14:48, Rob Weir robw...@apache.org wrote:
 Moving conversation over to the dev list...

 On Sun, Aug 25, 2013 at 8:18 PM, Ariel Constenla-Haile
 arie...@apache.org wrote:
 On Sun, Aug 25, 2013 at 05:31:35PM -0400, Rob Weir wrote:
  I wanted to verify the
  Apache_OpenOffice_4.0.0_Linux_x86-64_install-rpm_en-US.tar.gz.asc
  against my download.  I downloaded the KEYS using: wget
  http://www.apache.org/dist/openoffice/KEYS Then I imported the keys.
 
  But when I ran gpg --verify it said:
 
  $ gpg --verify
  Apache_OpenOffice_4.0.0_Linux_x86-64_install-rpm_en-US.tar.gz.asc
  gpg: Signature made Tue 16 Jul 2013 05:39:05 PM CDT using RSA
  key ID B8E50356 gpg: Can't check signature: No public key
 
  The Key ID B8E50356 is not in the set I downloaded from your KEYS
  file.  Why is it not in there??
 

 Hi Ariel,  is B8E50356 your key?

 Yes, it is a bug that my key is not in
 http://www.apache.org/dist/openoffice/KEYS

 This file should be a copy of

That's debatable, see below.

 https://people.apache.org/keys/group/openoffice.asc or
 https://people.apache.org/keys/group/openoffice-pmc.asc (in case only
 PMC members are supposed to sign artifacts).


 Does anyone know:  can we (may we?) do this now?  Or is this something
 to fix in 4.0.1 release?

The KEY used to sign an artifact MUST be in the KEYS file that is
linked from the download page(s).
This is something that should be checked as part of a release vote.

The files under https://people.apache.org/keys/group/ are
automatically generated from LDAP.
As such they only contain keys from current entries.
However KEYS files may still be needed to validate archive releases
where the key is not in LDAP (or the key is in LDAP but the owner is
no longer in the relevant TLP or PMC group).

For the above reasons, at present I don't think it makes sense to
blindly copy the file.

The file http://www.apache.org/dist/openoffice/KEYS has historically
been manually maintained.
New keys are added to the file as required.
Old keys are never deleted, as they may have been used for signing
archive releases.
[I guess there might be a case for deleting a compromised key]

So I suggest you just add the missing key(s) - with header info please
- to the dist/oo/KEYS file.


 -Rob



 @OP: please import the keys from
 https://people.apache.org/keys/group/openoffice-pmc.asc

  Allen -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.14
  (GNU/Linux)
 
  iQEcBAEBAgAGBQJSGABoAAoJEK3AFbtYOknnVI0IALxbJlIW58Ll3R8aryWQXX4k
  GJ1+Gh5cWFDYvFq9Cetz86vnxDuCaiVMxEOwnRc+PtBQHWHpzRuSKTG16fOs/5JD
  SGykhVkgdkRodpiuQKE8n/kV8+/aEaa+9WpxVdn+eqhTsi3nc570JQbOaw0sCOrY
  Nrdwm5Urm7w6wcP240g5UD4pjfXqAieEEe/0FdJQepikt7VFlRjsvRYVekSDHkUL
  t5XgL3LQAaTt47vMM9EyPMxK2RfIG2dXUQ54phtgFs9CUt2yqVF4s8mA2Ha+moPu
  rc2mS4vrKeswCO6ywyfDtaQnbaZrLxPG0y9Ql0hcUv5CEHE0eRxnJgkkTYzVUaI=
  =0QtH -END PGP SIGNATURE-

 I suggest you configure Enigmail in Thunderbird to sign using PGP/MIME
 instead of the old-fashioned inline-PGP, in Thunderbird's Account
 Settings go to OpenPGP Options and enable Use PGP/MIME by default, as
 explained here http://www.rainydayz.org/content/81-account-settings


 Regards
 --
 Ariel Constenla-Haile
 La Plata, Argentina

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [RELEASE]: new snapshot build for AOO 4.0.1

2013-08-30 Thread sebb

On 30 August 2013 21:36, Kay Schenk kay.sch...@gmail.com wrote:
 On Fri, Aug 30, 2013 at 12:34 PM, Fernando Cassia fcas...@gmail.com wrote:

 On Fri, Aug 30, 2013 at 10:27 AM, Jürgen Schmidt jogischm...@gmail.com
 wrote:
 
  The wiki page is updated and for Linux you can find archive builds as
  well directly in the directory (not linked via the wiki).

 Thanks for the heads up , Jürgen. Can you point us to the download
 dir? Im not sure what wiki or what page within such wiki youre
 referring to. I'd like to test Windows builds if possible.

 FC



 First a BIG THANK YOU to Ariel for all his great Linux builds in the past!
 I'm now downloading my first Linux build from the new build area.

 the link to developer snapshots is:

 https://cwiki.apache.org/confluence/display/OOOUSERS/Development+Snapshot+Builds

Great page, very easy to follow, and it does not need Javascript!
That's exactly the sort of page I think would be suitable for
downloads/other.html

However there seem to be two tables for language packs:

Snapshot Apache OpenOffice 4.0.1 - language packs

and

Apache OpenOffice - language packs

The former links appear to be full installations; that table seems to
be a repeat of

Snapshot Apache OpenOffice 4.0.1 - full installation sets

so could probably be dropped.






 --
 During times of Universal Deceit, telling the truth becomes a
 revolutionary act
 - George Orwell

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org




 --
 -
 MzK

 When in doubt, cop an attitude.
 -- Cat laws

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: template troubles in xx pseudo NL website

2013-08-29 Thread sebb

On 29 August 2013 14:49, Rob Weir robw...@apache.org wrote:
 As I mentioned earlier I'm trying to make a self-contained English
 language website that can copied and translated to make new (or
 refreshed) NL websites.  This includes the 25 or so core pages that
 should be translated for every language.

 You can see the site here in:

 https://svn.apache.org/repos/asf/openoffice/ooo-site/trunk/content/xx/

 or a live copy here:

 http://www.openoffice.org/xx/

 As you can see, it is not picking up the topnav at all.  Ditto for the
 side navigation on pages like:

 http://www.openoffice.org/xx/why/

 Any idea what I'm missing?  I've added a ssi.mdtext file into the
 templates directory, e.g.,:

 https://svn.apache.org/repos/asf/openoffice/ooo-site/trunk/templates/xx/ssi.mdtext

When downloading http://www.openoffice.org/, I get:

===
!DOCTYPE html

html
head
link href=/css/ooo.css rel=stylesheet type=text/css
===

However, for http://www.openoffice.org/xx/ I get:

===
[an error occurred while processing this directive]
html
===

Note that in both cases the xmlns attribute is dropped from the html file.

Perhaps try removing the DOCTYPE line and xmlms attribute and see if
that has any effect

 Regards,

 -Rob

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Where do I download Open Office 4.0?

2013-08-27 Thread sebb

On 26 August 2013 18:47, Donald Whytock dwhyt...@apache.org wrote:
 On Fri, Aug 23, 2013 at 10:34 AM, sebb seb...@gmail.com wrote:

 On 23 August 2013 14:55, Rob Weir robw...@apache.org wrote:
  On Fri, Aug 23, 2013 at 8:47 AM, Rob Weir robw...@apache.org wrote:
  Hi Kit,
 
  Rory gave you another link that should work.  But so we can debug
  things more on our end, could you let us know was operating system you
  are using, and what browser version?
 
 
  Response, off list, was that she was using Win 7 64 bit and IE 10.
 

 Whatever client is used to access the page, ISTM that the download box
 should still be present.


 It isn't present for me under IE8 on WinXP, but this is my office system
 and javascript is probably turned off.

Even if javascript is turned off, the green download box should still be shown.

At present, for me it says:

Download Apache OpenOffice
Javascript needs to be enabled to download Apache OpenOffice.


 Don

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [IDEA] Back to School with AOO blog post

2013-08-23 Thread sebb

On 22 August 2013 18:54, Rob Weir robw...@apache.org wrote:
 On Tue, Aug 6, 2013 at 10:40 AM, Rob Weir robw...@apache.org wrote:
 It is that time of year now:  back to school.

 I wonder if this could make a good blog post.   Of course, AOO being
 free is core part of the story.  But even better would be if we had a
 top 10 list of ways in which AOO is great for school.

 Maybe specific templates or extensions that are especially useful for school?

 Good for students as well as for teachers?

 Any special tips or techniques about using AOO for school work?

 If we can come up with 10 nice ideas, I can write it up in a blog post.

 Pleas share any ideas you might have.


 Here is a draft blog post, using a number of the ideas suggested in this 
 thread:

 https://blogs.apache.org/preview/OOo/?previewEntry=back_to_school_with_apache


Minor nits:

s/multiple computer./multiple computers./

You can use OpenOffice perpetually, - that does not read well to me.
Perpetually means ceaselessly/continuously to me - which is not
something we should encourage! (kids also need to get out and about)

Perhaps put something like:

You can use OpenOffice, with no subscription fees - ever. 

Although the picture is fun, it is obviously US and white (and not recent).
I'm not sure it gives the correct message for a multilingual modern appliction.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Where do I download Open Office 4.0?

2013-08-23 Thread sebb

On 23 August 2013 14:55, Rob Weir robw...@apache.org wrote:
 On Fri, Aug 23, 2013 at 8:47 AM, Rob Weir robw...@apache.org wrote:
 Hi Kit,

 Rory gave you another link that should work.  But so we can debug
 things more on our end, could you let us know was operating system you
 are using, and what browser version?


 Response, off list, was that she was using Win 7 64 bit and IE 10.


Whatever client is used to access the page, ISTM that the download box
should still be present.


 -Rob



 Thanks!

 -Rob

 On Fri, Aug 23, 2013 at 7:52 AM, Kit Spackman bosspec...@hotmail.com wrote:
 Hello there,

 I've been a satisfied user of Open Office for some time, and I'm currently 
 using V 3.4.1. I'm TRYING to download V 4.0 but the linked 'Download' page 
 on your web site at 
 http://www.openoffice.org/download/?utm_source=AOO3_4_1_en-GButm_medium=Clientutm_campaign=Upgrade
  has all sorts of links to add-ons etc, plus an invite to me to spread the 
 word about Open Office, but it DOESN'T have a button download the update! 

 I've searched all over the site but cannot find anywhere to download it, 
 please tell me where I can do this.

 Regards
 Kit Spackman


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Bug 49699 - Remove SPAM attachments

2013-08-21 Thread sebb

On 21 August 2013 06:31, Rainer Bielefeld i...@bielefeldundbuss.de wrote:
 Hi,

 This query
 https://issues.apache.org/ooo/buglist.cgi?f1=attachments.submittero1=equalsquery_format=advancedresolution=---v1=non-migrated%40openoffice.orglist_id=82681

 shows 201 Bugs with attachments created by user Unknown, a small sample
 mostly showed SPAM contents in these attachments (Bugs 57603, 17954, 34866).

 But some might contain appropriate sample documents (Bug 9231, 13394, ...),
 so unfortunately any of these attachments will have to be checked manually.

 SPAM should be deleted!

Not sure any but super users have permission to delete attachments.
But administrators should be able to mark them as obsolete, which
would stop them showing up by default.
That would be a start.

 Best regards

 Rainer

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Should download pages require Javascript?

2013-08-20 Thread sebb

On 19 August 2013 19:44, Rory O'Farrell ofarr...@iol.ie wrote:
 On Mon, 19 Aug 2013 20:14:08 +0200
 Andrea Pescetti pesce...@apache.org wrote:

 On 19/08/2013 sebb wrote:
  Note that the page http://www.openoffice.org/download/other.html also
  requires Javascript!

 This is not so good. The noscript option should direct the user to
 http://www.apache.org/dyn/aoo-closer.cgi/openoffice/
 (users who disable JavaScript are likely to be able to browse the
 FTP-like structure they will see there).

  Whatever method is chosen, I think it should be possible to download
  AOO without the use of Javascript.

 It should also be possible to download OpenOffice in the cases where
 JavaScript parsing breaks, i.e., we should have alternative download
 links that are always visible (working JavaScript, broken Javascript, no
 JavaScript).

 Regards,
Andrea.


 I think the automatic selection of a best fit OpenOffice for the user is 
 good.

Agreed.

 Forum experience suggests that many unskilled users will get confused if 
 offered the full range of options of versions for operating sytems, languages 
 and language packs.

Yet that is exactly what the page
http://www.openoffice.org/download/other.html does.

All the versions and all the language packs are laid out in one huge table.
Probably easier to use than the ASF mirror structure, but still not easy.

However, although the page requires Javascript, it does not use
Javascript to narrow the choices for the user.
AFAICT the Javascript is only used to build the table so it might as
well be static HTML that any browser can use.

If it were turned into a static page, I think it might be a better
alternative to the ASF mirror structure.

Maybe there should be a separate discussion about how to organise the
download pages when the autodetection does not work.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Should download pages require Javascript?

2013-08-20 Thread sebb

On 20 August 2013 01:20, Rob Weir robw...@apache.org wrote:
 On Mon, Aug 19, 2013 at 6:39 PM, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 08/19/2013 09:51 PM, schrieb sebb:

 On 19 August 2013 20:27, Rob Weirrobw...@apache.org  wrote:

 On Mon, Aug 19, 2013 at 2:14 PM, Andrea Pescettipesce...@apache.org
 wrote:

 On 19/08/2013 sebb wrote:


 Note that the page http://www.openoffice.org/download/other.html also
 requires Javascript!



 This is not so good. The noscript option should direct the user to
 http://www.apache.org/dyn/aoo-closer.cgi/openoffice/
 (users who disable JavaScript are likely to be able to browse the
 FTP-like
 structure they will see there).


 Whatever method is chosen, I think it should be possible to download
 AOO without the use of Javascript.



 It should also be possible to download OpenOffice in the cases where
 JavaScript parsing breaks, i.e., we should have alternative download
 links
 that are always visible (working JavaScript, broken Javascript, no
 JavaScript).


 I think this is the key insight.  There are actually three cases to
 consider:

 1) Java script disabled

 2) Javascript is supported, but not working with our page

 3) Javascript working fine.

 Some of the more recent reports are about #2.  These are older
 versions of Internet Explorer, e.g., I.E. 6.  Anoscript  block will
 not help in this case.


 But if it is possible to detect the broken Javascript without
 crashing, then it would be possible to treat the browser as if it did
 not have Javascript.


 Right, has anybody an idea how to detect such broken JS engines?


 From this test it looks like the main issue is Internet Explorer before I.E. 
 8:

 http://browsershots.org/http://www.openoffice.org/download/

 One approach is to see if you can code around that error and get it to
 work correctly on older I.E. installs.

 Another approach is to use one of these techniques to detect older
 I.E. and then fall back to a non-script page:

 http://msdn.microsoft.com/en-us/library/ms537509%28v=vs.85%29.aspx

It might also be possible to use the HTTP headers User-Agent and
Accepts-Language as a backstop for when Javascript is not available.
This might need some Infra config support.

 -Rob

 Marcus



 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Should download pages require Javascript?

2013-08-19 Thread sebb

The download page at present requires Javascript.
Without Javascript, there is no obvious way to download AOO 4.0 at all.

Note that the page http://www.openoffice.org/download/other.html also
requires Javascript!

It's nice to be able to use Javascript to guess which download the
user wants, but I don't think that this convenience should be at the
expense of preventing users from finding the download link if they
don't have Javascript (or they have a broken version of Javascript).

How about having drop-down lists for the OS and Language?

Javascript could be used to preset the values, but at least the user
would remain in control, e.g. if the Javascript failed to recognise
their system or they wanted to download a different language from the
one supported by their browser.

Whatever method is chosen, I think it should be possible to download
AOO without the use of Javascript.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Example of spreadsheet formula testing

2013-08-19 Thread sebb

On 19 August 2013 11:41, Regina Henschel rb.hensc...@t-online.de wrote:
 Hi Jan,

 janI schrieb:

 On 19 August 2013 12:24, Regina Henschel rb.hensc...@t-online.de wrote:

 [..]

 But for both kind of testing there exists a problem with expected
 values. For example, calculation of PMT needs expm1 and log1p, or
 calculation of LINEST needs lot of matrix calculation. It is not
 impossible
 to calculate this with Java, but who will do it? And when you use
 constant
 expected values, from where do you get them and how to ensure, that they
 are valid?


 Just one simple question, how do you do manually today ? use the same
 constants.


 I use my MuPad 3.1 and sometimes calculate values on
 http://www.wolframalpha.com/

 Or I compare the values to those from Gnumeric and Excel. Having the same
 value does not mean that they are correct, but the other way round, having
 different values means that I have to investigate it.



 We should not try to invent a total new road, just do what we do today
 more
 efficiently.


 Suggestions?

If there are a *lot* of similar calculations that need to be tested,
then it might be worth investing some time looking at at ways to
expresss these simply as a script, and then write a parser to drive
the existing test code from the script, rather than having to update
the Java test code.

The advantage would be that just about anyone can create test cases,
and they are easy to check by hand.

However the parser may be a lot of work; I don't know.

Maybe there is already a suitable script language which already has
suitable examples?

Also there are various script interpreters that already undestand some
calculations, for example Commons JEXL.
It should be possible to use them to check the test cases, and
possibly even use as a basis for parsing the test cases.

 Kind regards
 Regina




 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org

 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Strange phrase on download page: with other users

2013-08-19 Thread sebb

On 19 August 2013 11:42, Rory O'Farrell ofarr...@iol.ie wrote:
 On Mon, 19 Aug 2013 11:16:34 +0100
 sebb seb...@gmail.com wrote:

 There are two sentences on the download page which read very oddly to me:

 1) Extend your Apache OpenOffice functionality with a large and
 growing library of extensions and dictionaries with other users.

 2) Extend your Apache OpenOffice creativity with a large and growing
 library of templates with other users.

 I don't understand the phrase with other users.
 Well, obviously I understand the words, but they don't make sense to
 me in that context.

 Can anyone enlighten me what the phrase is intended to mean here?

 If the phrase were dropped entirely, the sentences would make sense to me.

 In both cases, if the with is replaced by from the phrases make sense.

Thanks, now fixed!


 --
 Rory O'Farrell ofarr...@iol.ie

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Usability: dictionary updates and profile migration

2013-08-19 Thread sebb

On 19 August 2013 16:39, Rob Weir robw...@apache.org wrote:
 On Mon, Aug 19, 2013 at 11:16 AM, Andrea Pescetti pesce...@apache.org wrote:
 Rob Weir wrote:

 On Thu, Aug 8, 2013 at 3:15 AM, Andrea Pescetti wrote:

 Not seeing comments, I'll publish the updated dictionary in a couple of
 days. In the end this is not breaking anything, it just produces a Fatal
 Error (that is actually not fatal, and solved by simply restarting
 OpenOffice) during profile migration. ...

 I don't know if it is related, but we're starting to get a few reports
 from confused users on Facebook.  It sounds like they are getting
 dictionary upgrade notifications but are then lost when trying to
 download/install the update.   It is coming down as a ZIP file rather
 than an OXT.


 This is a known problem with Internet Explorer (unrelated to this specific
 extension): at times, IE tries to be smart and fixes the filename
 extension into ZIP rather than OXT as you correctly observed. I don't have
 more precise information, but this has happened for years and the common
 advice given on lists is use another browser (obviously, capable users can
 rename the file as OXT).


 If I.E. is suggesting a ZIP content type then this means they've
 resorted to content sniffing and found the ZIP magic number in the
 file header.   Usually a browser only resorts to this if every other
 attempt to determine the content type has failed, e.g., HTTP response
 headers, file name, etc.   This is something that should be fixable.

The request uses a lot of redirects; the last two respond as follows
(some headers dropped):

HTTP/1.1 302 Found
Content-Disposition: attachment; filename=dict-en.oxt
Location: 
http://heanet.dl.sourceforge.net/project/aoo-extensions/17102/1/dict-en.oxt
Content-Type: text/html
Content-Length: 0
Server: lighttpd/1.4.26


HTTP/1.1 200 OK
Server: Apache/2.2.14
Accept-Ranges: bytes
Content-Length: 6529865
Content-Type: application/octet-stream

I did an experiment with IE 8 - it downloads as .zip from the
SourceForge, but downloads OK from a copy on people.apache.org.

The relevant response headers from people.a.o are:

HTTP/1.1 200 OK
Server: Apache/2.2.24 (Unix) mod_ssl/2.2.24 OpenSSL/1.0.1e DAV/2
Accept-Ranges: bytes
Content-Length: 6529865
Content-Type: application/vnd.openofficeorg.extension

So it might be sufficient to ensure that the correct content-type is returned.

I think this will need to be done by SourceForge on all their systems.

 -Rob



 Regards,
   Andrea.

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Reporting a problem with the OpenOffice website

2013-08-17 Thread sebb

On 17 August 2013 06:35, Rory O'Farrell ofarr...@iol.ie wrote:
 The OpenOffice Forum server will be offline until sometime on Sunday evening 
 (UTC) for major OS and phpBB upgrades

Can't the Forum URLs be temporarily redirected to a page that explains this?

 --
 Rory O'Farrell ofarr...@iol.ie

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Reporting a problem with the OpenOffice website

2013-08-17 Thread sebb

On 17 August 2013 09:01, Andrea Pescetti pesce...@apache.org wrote:
 sebb wrote:

 On 17 August 2013 06:35, Rory O'Farrell wrote:

 The OpenOffice Forum server will be offline until sometime on Sunday
 evening (UTC) for major OS and phpBB upgrades

 Can't the Forum URLs be temporarily redirected to a page that explains
 this?


 Sure, but this would require coordination and actually the redirection would
 need to be at the DNS level, since the operating system and web server are
 being upgraded too.

 As Jan wrote, nobody sent him the text of a notice to put online instead of
 the error message you see now (which will only appear for brief periods
 anyway; most of the time httpd will be down) and he didn't put anything.

I have yet to find that the entire site [1] is down, but the forums
are responding with SQL error messages.

So having a redirect page would be useful to prevent people sending
unnecessary reports.

Also people are more likely to accept that an entire site is down,
whereas a site that is partly broken gives a worse impression.

Please consider adding a maintenance page.

[1] http://forum.openoffice.org/

 I've now added a note to http://www.openoffice.org/contact_us.html , the
 source of the mails we are receiving, to explain that the forums are under
 maintenance.

 Regards,
   Andrea.


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: The Names of OpenOffice (How users are finding our website)

2013-08-16 Thread sebb

On 16 August 2013 12:56, Rob Weir robw...@apache.org wrote:
 And update of data gathered from website visitors.  Of the ones who
 come to our website from a search engine results list (and that is 48%
 of our total visitors), the top ten search queries, along with a count
 of recent visitors, are:

 1. open office (326,369)
 2. openoffice (213,374)
 3. openoffice download (32,188)
 4. openoffice.org (21,786)
 5. オープンオフィス (13,476)
 6. open office mac (11,307)
 7. apache openoffice (10,576)
 8. open office download (8,964)
 9. openoffice for mac (7,395)
 10. download open office (7,006)

 Note the strong drop after the first two queries.

 (And what is #5?  Japanese?  What does it say?)

Google Translate says it is Japanese for open Office (sic)

 So what does this all mean?

 A. Users are not consistent about whether the name is one word or two.
  Maybe they hear about the name by ear?  Or maybe this is just the
 pull of standard language rules.  The noun is office and open is
 an adjective.  It is hard to overcome years of schooling to think of
 an artificial name like OpenOffice.

 B.  The core name in their mind is OpenOffice/Open Office without
 the .org or the Apache.  This is what they are searching for when
 they look for us.

 Now, one might have a theory that uses searching for open office end
 up on our website by mistake.  Maybe they were searching for something
 else.  For example, this term is also used to refer to an office
 seating plan without walls, where everything is open in a big room.
 This is also an open office.  However, if I look at only
 search-directed traffic that actually leads to a download of AOO, the
 query open office and openoffice are also at the very top of the
 list.

 Regards,

 -Rob

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Microsoft Censors OpenOffice Download Links

2013-08-16 Thread sebb

On 16 August 2013 19:44, Rory O'Farrell ofarr...@iol.ie wrote:
 On Fri, 16 Aug 2013 14:39:16 -0400
 Rob Weir robw...@apache.org wrote:

 On Fri, Aug 16, 2013 at 1:27 PM, Donald Whytock dwhyt...@apache.org wrote:
  On Fri, Aug 16, 2013 at 12:32 PM, Rob Weir robw...@apache.org wrote:
 
  On Fri, Aug 16, 2013 at 12:18 PM, Hagar Delest hagar.del...@laposte.net
  wrote:
   Objet : Re: Microsoft Censors OpenOffice Download Links
   Not to speak for them, but I suspect they would point out the fact
   that we there are over 100 Apache projects, and they all seem to do
   fine with distribution via the mirrors.
  
   Personally, I'd wonder where this rates with us in terms of priority.
   Compare to, say, forum stability improvements, code signing for our
   installers, and further buildbot coverage, where do torrents rate?
  
   Of course it's not a priority.
   But think about the mechanism of torrent: once it's initiated, it
  spreads by itself without any input needed. I'm not sure we need powerful
  resources for the seeds, we can even limit the number of uploads I guess.
  And then let the torrent spread among users.
  
   A forum was not in the field of the ASF scope. The AOO forum is still
  doing and rather well, there is a lot of cooperation and feedback when
  information is forwarded from on side to the other. So why not make a
  torrent a first for ASF?
  
   Please remember that you're handling an office suite, it's not a niche
  program, it's something that is heavily popular, you tell it yourself when
  you inform the list about the millions downloads. Ubuntu offers torrents
  for example.
  
 
  AOO is popular.  Torrents are not.  I bet that 1% of downloads were
  of torrent, when OOo had them.
 
  Remember, a common question from users is I just downloaded
  OpenOffice and now I cannot find it.  So skill level of typical user
  is not ideal for explaining how to download via P2P.
 
   If ASF does not want to do new things because no other ASF project has
  even tried, then I'm rather worried about the future. Especially when on
  the other side LibreOffice has a so efficient team, very good at marketing
  their project.
  
 
  1. Maybe ask LibreOffice how many torrent downloads they see?  That
  would be an interesting number to know.
 
  2. This is not a question of avoiding doing something new.  It is a
  question of prioritization based on cost and benefit.
 
  3. Torrents are not even new. They are old technology.
 
  4. There is nothing to prevent someone from seeding a torrent for AOO
  today, right now if you thought it was important.  It does not need to
  come from Apache.
 
 
  Infra could conceivably create torrents for every ASF distro file, probably
  on an automated basis.  Were that to happen, the effort by the AOO TLP
  would be nil and the effort proportionally related to AOO would be
  negligible.
 
  Of course, this could take some significant setup effort on Infra's part,
  and if only the AOO torrents were ever used someone might say, Why are we
  doing this for only one TLP?
 
  It would be best if ASF could do it so as to add legitimacy to the torrent.
   Otherwise, if AOO itself was doing it, it would need to be on a
  respected/respectable torrent server, such that we could point to it and
  say, That is the official AOO torrent.
 

 I think that's the key.  If it is to be considered official then we
 need sufficient control to ensure that it has not been tampered.  What
 we do right now is have Release Candidates on Apache servers, which
 are voted on and then copied onto another Apache server for archives,
 and then rsynced from that Apache server by SourceForge.  And all
 along we have the original digital signature files that can be
 verified.  So it is around as secure as we can go without taking the
 builds themselves right from Apache-hosted buildbots, which is the
 next logical step.

 But honestly my low motivation for this is based on the fact that
 we're talking about a 150MB file, not a 4GB ISO image.  The typical
 user can download AOO in less time than it took me to write this note.
  For me it takes longer to install AOO than to download it.  So in the
 grand order of annoyances related to AOO, the download time does not
 seem to rate very highly.

 That's my personal opinion.  But the nice thing about Apache is this
 doesn't prevent someone else from moving this forward if they have the
 motivation.  Everyone is able to scratch their own itch here.

 Regards,

 -Rob


  Maybe a cheap 10gig VM?
 
  Don

 With respect, Rob, we don't all live in fibre access broadband areas. I count 
 myself lucky in having 200KB/sec access; there are those who are still stuck 
 with 56KB modems on bad dial-up lines. My OpenOffice download is typically 
 12-14 minutes.

But would a torrent improve the download times over slow links?
Surely the limiting factor is the link speed in that case, as any
download servers would have a much higher bandwidth.





 --
 Rory O'Farrell

Re: Issue 122927 -- spreadsheet formula compatibility

2013-08-15 Thread sebb

On 15 August 2013 14:47, Rob Weir robw...@apache.org wrote:
 On Thu, Aug 15, 2013 at 9:41 AM, janI j...@apache.org wrote:
 On Aug 15, 2013 3:06 PM, Rob Weir robw...@apache.org wrote:

 https://issues.apache.org/ooo/show_bug.cgi?id=122927

 It boils down to how an IF() statements are evaluated.

 Remember, the typical form is IF(Condition;X;Y) where you give a
 return value for the case where Condition is TRUE and another value
 when Condition is FALSE.

 But it is also possible to leave out the last parameter and have a
 formula like this:

 IF(Condition;X)

 So what does the formula evaluate to if Condition is FALSE?

 The behavior in 4.0.0, returning FALSE, is correct according to the
 ODF 1.2 specification and is the same as what Excel does.  However, it
 is different than what earlier versions of OpenOffice did, namely
 returning 0.0.

 We obviously cannot do both.  I think the AOO 4.0.0 behavior is
 correct and should remain.

 I dont understand why we cannot do both, most programming languages
 interpret falase==0 and true==1, that allows the use of boolean functions
 in calculations.


 If the user takes the results of the IF() calculation and uses it in
 another formula, then FALSE is automatically treated as 0 in any other
 formula where a number is expected.  You are correct in your
 assumption there.   So no one gets a wrong answer in a calculation
 because of the change.

 What is different is what appears in the cell that actually has the
 IF() statement in it.  AOO 4.0 and Excel show FALSE.  Earlier versions
 of AOO showed 0.   In this sense we can have one default behavior or
 the other, but not both.

Could you not add a setting that controls the behaviour?

For a fresh install AOO 4.x will show FALSE.
But if the user sets the appropriate backwards compatibilty option, it
will show 0.

 -Rob


 rgds
 jan i

 I'd like to close the issue as NOTABUG.  But I'd like to get a few
 more thoughts on this first.

 Regards,

 -Rob

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Issue 122927 -- spreadsheet formula compatibility

2013-08-15 Thread sebb

On 15 August 2013 15:21, Rob Weir robw...@apache.org wrote:
 On Thu, Aug 15, 2013 at 10:07 AM, sebb seb...@gmail.com wrote:
 On 15 August 2013 14:47, Rob Weir robw...@apache.org wrote:
 On Thu, Aug 15, 2013 at 9:41 AM, janI j...@apache.org wrote:
 On Aug 15, 2013 3:06 PM, Rob Weir robw...@apache.org wrote:

 https://issues.apache.org/ooo/show_bug.cgi?id=122927

 It boils down to how an IF() statements are evaluated.

 Remember, the typical form is IF(Condition;X;Y) where you give a
 return value for the case where Condition is TRUE and another value
 when Condition is FALSE.

 But it is also possible to leave out the last parameter and have a
 formula like this:

 IF(Condition;X)

 So what does the formula evaluate to if Condition is FALSE?

 The behavior in 4.0.0, returning FALSE, is correct according to the
 ODF 1.2 specification and is the same as what Excel does.  However, it
 is different than what earlier versions of OpenOffice did, namely
 returning 0.0.

 We obviously cannot do both.  I think the AOO 4.0.0 behavior is
 correct and should remain.

 I dont understand why we cannot do both, most programming languages
 interpret falase==0 and true==1, that allows the use of boolean functions
 in calculations.


 If the user takes the results of the IF() calculation and uses it in
 another formula, then FALSE is automatically treated as 0 in any other
 formula where a number is expected.  You are correct in your
 assumption there.   So no one gets a wrong answer in a calculation
 because of the change.

 What is different is what appears in the cell that actually has the
 IF() statement in it.  AOO 4.0 and Excel show FALSE.  Earlier versions
 of AOO showed 0.   In this sense we can have one default behavior or
 the other, but not both.

 Could you not add a setting that controls the behaviour?

 For a fresh install AOO 4.x will show FALSE.
 But if the user sets the appropriate backwards compatibilty option, it
 will show 0.


 In theory yes, but in practice users don't really think about this as
 a per-installation setting.

Maybe not, but if it was documented it could solve problems for some users.
And it is a fairly simple solution.

 They want their spreadsheet to look the
 same as it was when it was created, even if it was created in a
 different version of OpenOffice,

That I can understand.

 or in a different spreadsheet application altogether.

In which case the default may be something else entirely - e.g.
omitted trailing values are not allowed.

 So a more targeted fix would be to trigger backwards compatibility
 mode whenever you read a spreadsheet that was created in older
 versions of AOO.

In which case, there needs to be some way to change the behaviour in
case the user wants to update to the new behaviour without recreating
the spreadsheet.

 Even better is to have a declarative approach where the behaviors are
 encoded in the document itself as metadata.  This approach has been
 discussed, but is not yet standardized.

That would require older sheets to be updated, unless the default for
missing metadata varied between versions.

When context-sensitive solutions work, it's great.
However when they fail to work, it's usually not at all obvious what
the problem is - nor the solution.

 -Rob

 -Rob


 rgds
 jan i

 I'd like to close the issue as NOTABUG.  But I'd like to get a few
 more thoughts on this first.

 Regards,

 -Rob

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Some thoughts on quality

2013-08-14 Thread sebb

On 14 August 2013 23:10, Rob Weir robw...@apache.org wrote:
 On Wed, Aug 14, 2013 at 4:24 PM, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 08/14/2013 10:05 PM, schrieb Hagar Delest:

 Le 14/08/2013 21:39, Rob Weir a écrit :

 Maybe we need to call an earlier build the RC so it will get more
 attention? We had a complete test build that we were testing for
 over a month. But maybe it is ignored unless we call it an RC? In
 other words, there were many opportunities for users to help try 4.0
 before it was released, but maybe there opportunities were not well
 known.


 I think that it was too difficult to get to the dev versions and it was
 too much complicated. There was no clear link to download a dev version
 (I had to rely on the url in the messages from the dev list).


 This was intended.

 We hadn't published the dev builds via Apache or SF mirrors but only on 2
 people accounts. Apache policy says it's not allowed to publish them for a
 wider audience to save the servers from a high traffic load. It's the job of
 the mirrors to handle this.


 What I see (from a standard user point of view) for a RC:
 - When a dev version is almost done, rename it RC and make it known
 (blog and we would relay the announcement in the forums of course)
 - Have a link visible under the main download button of the download


 Both can be done, depending where the install files are located.


 page (perhaps a similar button as a dedicated entry)
 - Make that RC installable in parallel with a stable version
 - No file association allowed for that RC by design


 IMHO the last both points doesn't apply to a RC [1] as it wouldn't be a RC
 anymore. One of the RC attributes is to change it into the final release
 with, e.g., just a file name change. But this has to be done without any
 code changes. Otherwise you have to change code parts, build again, test
 again, ... ;-)

 But a Beta release could go this separated way.


 Right.  A release is a release is a release.  The basic requirements
 for every release still apply:

 1) 3 PMC +1 votes
 2) Must include source files
 3) Digital signatures, hash files, etc.

 But we can have a beta release, that follows these rules, and it
 would be acceptable.  We can then host on the mirrors, publicize, etc.

There would likely be some restrictions on how many extra downloads
are permitted.
For example, the ASF mirrors probably could not cope with a set of
betas of all the languages for all the OSes in addition to the current
GA release.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Unnecesary filestructure on images

2013-08-12 Thread sebb

On 12 August 2013 17:05, janI j...@apache.org wrote:
On Aug 12, 2013 3:25 PM, Rob Weir robw...@apache.org wrote:

On Sun, Aug 11, 2013 at 7:49 PM, Alexandro Colorado j...@oooes.org wrote:
On Sun, Aug 11, 2013 at 6:33 PM, Rob Weir robw...@apache.org wrote:

On Sun, Aug 11, 2013 at 1:35 AM, Alexandro Colorado j...@oooes.org
wrote:
I think the image structure on the website is a bit messy, there has
been
some cleanup done by kschenk but I think there is still a lot of
clean up
work to be done.

For example, the new logo, was simply draged and drop to the AOOLogos
folder with a huge name. I understand the name was needed to
identify it
between the rest of the competitive logos. But now that is selected,
the
current name is unecessary long.

Apache_OpenOffice_Logo_ChrisR_selected_2013-06_optim_300w.png

http://svn.apache.org/viewvc/openoffice/ooo-site/trunk/content/images/AOO_logos/Apache_OpenOffice_Logo_ChrisR_selected_2013-06_optim_300w.png?view=log

Right. That work is incomplete. I checked it in originally, after
the logo vote, so we could start working on the product integration
immediately. But note that the above logo is not the one we actually
used in AOO 4.0 !!

The one we actually used is this one:

https://svn.apache.org/repos/asf/openoffice/branding/Apache_OpenOffice_Logo_ChrisR_selected_2013-06_Inkscape_kg.svg

This was Chris R's contest logo with some minor technical changes.
Kevin G. used this and generated the PNG/JPG files for AOO 4.0, which
I helped check in.

My intent was to take that SVG and rename it to master-logo-40.svg

Again I think we do need a convention for a logo.svg as opposed to
ending with a logo-30.svg logo-40.svg logo-50.svg. An just incrementally
replace with the future logos as we update the SVG.

Here's the complication: The old logos are still relevant some some
purposes. For example, the PMC receives ongoing requests to approve
use of the old OpenOffice.org logo. Why would that happen? Often it
is a request by publishers who are making an e-book version of an
older print book. If their original request did not include the
e-book rights then they come back to us (and owners of every other
image they use) to request additional permissions.

So it may be possible, going forward, to store logos as SVN revisions
under the same name. But we cannot retroactively do this with
pre-Apache logos. And even if we could, this is harder for users of
the logo to access. It is much easier to have something like
logo-330.svg available via HTTP.

Of course you can have a hybrid approach:

1) When a new logo is introduced, svn copy the old one into a
/old-logos directory with a new descriptive name. This preserves the
version history.

2) New logo then is checked in as a new revision of logo-master.svg.

or do as I am used to

new logo is checked in as
logo-master.png
and
logo-4.x.png

this way we have history and actual logo. and most important the links
never change.

That's OK provided that the logo size and type (svg/png/gif etc.) don't change.
In other words, if it's a drop-in replacement (binary compatible in
code terms) then keeping the same name is sensible.

If either changes, this could cause incompatibilities, in which case a
name change may cause fewer issues in the long run.
And it should be more obvious that references have to change (as well
as being obvious which ones have been updated).

rgds
jan i

Regards,

-Rob

or something clean like that. However, I have not had any luck
getting this logo to load into Inkscape or Adobe Illustrator. I get
errors. And I have not had any luck getting Kevin to send a version
that will load.

So we're stuck right now with a logo that does load into Inkscape, but
is slightly different than the one we used in AOO 4.0.

At the same time we have old logos which had been untouch. I think
the
webdevs have small understanding of a svn is builted so that the
files
are
updated without having different versions laying arround. Over this
example, ooo-logo.png and
AOO4_website_logo.png

http://svn.apache.org/viewvc/openoffice/ooo-site/trunk/content/images/AOO_logos/AOO4_website_logo.png?view=log
exist.

All these proliferation of logos, usually will built up to become
incredibly messy to work. I suggest to put the information such as
author,
version, status, etc. on the comments of the commit and not on the
filename. Likewise to take the time to look for the source of the
image,
since there is an SVG/ folder to link the source of them, and
finally if
there are different images (sizes) to have a common convention.

A more logical naming scheme would be good, I agree. But this has
been waiting for resolution of which SVG we should actually be using.

Regards,

-Rob

If you want to review the images please go here:

Re: [discuss] drop support for Java 5 and Java 6 for Windows

2013-08-08 Thread sebb

On 8 August 2013 12:23, Rob Weir robw...@apache.org wrote:
 On Thu, Aug 8, 2013 at 5:43 AM, sebb seb...@gmail.com wrote:
 On 8 August 2013 02:26, Rob Weir robw...@apache.org wrote:
 On Wed, Aug 7, 2013 at 5:23 PM, Kay Schenk kay.sch...@gmail.com wrote:
 On Wed, Aug 7, 2013 at 11:24 AM, janI j...@apache.org wrote:

 On 7 August 2013 18:55, Andrea Pescetti pesce...@apache.org wrote:

  Oliver-Rainer Wittmann wrote:
 
  Important note for discussion: it is all about platform Windows.
  On my work to update the AOO build environment for Windows I recognized
  that it is hard to get an official JDK 1.5 (Java 5) or JDK 1.6 (Java 6)
  for Windows. Thus, I decided to go with JDK 1.7. The resulting AOO
  installation on Windows no longer works together with an JRE 6. It does
  not recognize an installed JRE 6 as an valid Java runtime environment.
 
 
  May we frame the problem in more technical terms, just to know what is
  broken? For example, why is this affecting only Windows and why is Java 
  6
  not recognized in your build? Could the problem be in detection rather
 than
  in the actual compatibility?
 
  Java issues were extensively discussed in earlier times, so here's a
 quick
  summary that also answers most of the questions in this thread:
  - As of 4.0, OpenOffice can be built with Java 5, 6 or 7
  - Whatever you use for building, the resulting binary has a Java
  baseline of 1.5 as per http://wiki.openoffice.org/**
  wiki/Policies/Java_Usage
 http://wiki.openoffice.org/wiki/Policies/Java_Usage(means: runs with
 Java 5, 6 or 7)
  - We built 4.0 with Java 6 (on Linux at least; not 100% sure about other
  platforms)
 
  In general, I agree that we should build on the most secure platform
  available. But, based on the above, what is the relationship between
  building on Java 7 and running on Java 6? To reuse Rob's Windows XP
  argument, sure we should build on a supported (by Microsoft) Windows
  version, but, if at all possible/reasonable, we shouldn't break
  compatibility with Windows XP.
 

 I am sorry if this posting is obvious to everyone, but reading the 
 remarks,
 make me think there are some confusion about what we mean with using java
 for development and runtime.

 One of the strength of java is program once, run everywhere . This is
 accomplished by by 2 magic trix (compared to eg. C++).
 1) Java does not compile to machine code but to pcode (a virtual machine),
 therefore you can build the program on linux, and run the build on window
 (or even one of the big mainframes).
 2) Java also does late binding (think of a very smart dll), so libraries
 are not part of your build.

 This means you can use a java development 1.7 on any platform, to make a
 build that runs on any platform and (nearly) any java runtime version. As
 an example I use areca backup, its a java program, the exact same jar 
 files
 run on vista,xp,win7,ubuntu and even android, areca is programm towards
 java 1.4, and I have 1.6 and 1.7 installed depending on platform.

 The problem is the classes and the API. If our code use just a single java
 1.7 specific call, the runtime must be at least 1.7. This is however no
 problem today, our code is build for the classes and api available in java
 runtime 1.5, so it will run there.

 Oracle have promised to keep the API and classes for 1.4 and forwards
 stable, and available in new versions. They are pretty good at living up 
 to
 the promise

 So in theory we can change build environment to java 1.7 and not tell 
 user,
 as long as we only use 1.5 API and classes. As part of a release cycle, we
 should of course test once with runtime 1.5.

 I wrote in theory because in the real world, we might want to (in future
 releases) use the 1.7 api for e.g. performance reasons, when that time
 comes we would have to make a wrapper class, just like we have in C++ to
 cover differences Linux/windows.

 Sorry again, if I misread the postings, but this is very much different
 from the XP scenario.

 rgds
 jan I.



 Thank you for this great explanation! So basically, review the AOO java 
 API.


 It is a bit more complicated than that.   The Java language itself has
 evolved, not just the libraries. There are bytecode changes as well.
 The difference between Java 1.7/1.6 is not very big, but there are
 more significant differences if you need to maintain compatibility
 with Java 1.5.  Not impossible, but it would be extra effort.

 AIUI the compiler just has to be told to generate the appropriate code:

 javac -source 1.5 -target 1.5

 The source will of course have to be 1.5 compatible.
 But is there very much Java code?

 And remember, the cost of supporting old platforms is not just the
 dev work.  It also involves QA and support..  If we say we support
 something then we really ought to be testing in, not just saying that
 we not aware of any problems.  The OpenOffice brand should mean that
 users can run on any supported platform and have a good experience.
 IMHO we should not say we

Re: [Build Environment] Windows - build using Java 7 (JDK 1.7) without having Visual Studio

2013-08-07 Thread sebb

On 7 August 2013 09:33, Rory O'Farrell ofarr...@iol.ie wrote:
 I'm deleting the quotes of the earlier discussion as it is not relevant to my 
 point here.

 I have no experience of building with Java and care little about the security 
 holes, so I bow to the expertise of those who know better.

 If changes are being made to the OpenOffice/Java interaction, would it be 
 possible that these be done in such a way that no matter whether installed on 
 a 32 or 64 bit windows, OpenOffice interfaced correctly with the installed 
 Java, so doing away with the requirement that a 32 bit Java be installed.

+1

AOO should work with what is installed, [though there is probably no
need to support Java 4 or earlier any more.]
Just because Java 5 or 6 is no longer current does not mean it is not
being used.
In particular, businesses tend to stay on older versions of Java (and
OS) until the cost of updating is less than the cost of not updating.

 This would prevent many queries on the Forum of why OO doesn't work with an 
 installed Java.


 --
 Rory O'Farrell ofarr...@iol.ie

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [Build Environment] Windows - build using Java 7 (JDK 1.7) without having Visual Studio

2013-08-07 Thread sebb

On 7 August 2013 11:55, janI j...@apache.org wrote:
 On 7 August 2013 12:51, sebb seb...@gmail.com wrote:

 On 7 August 2013 09:33, Rory O'Farrell ofarr...@iol.ie wrote:
  I'm deleting the quotes of the earlier discussion as it is not relevant
 to my point here.
 
  I have no experience of building with Java and care little about the
 security holes, so I bow to the expertise of those who know better.
 
  If changes are being made to the OpenOffice/Java interaction, would it
 be possible that these be done in such a way that no matter whether
 installed on a 32 or 64 bit windows, OpenOffice interfaced correctly with
 the installed Java, so doing away with the requirement that a 32 bit Java
 be installed.

 +1

 +1


 AOO should work with what is installed, [though there is probably no
 need to support Java 4 or earlier any more.]
 Just because Java 5 or 6 is no longer current does not mean it is not
 being used.
 In particular, businesses tend to stay on older versions of Java (and
 OS) until the cost of updating is less than the cost of not updating.


 correct, but at the same time companies are typically security aware, and
 java 6 (and below) have some serious know issues.

 I would have thought that its possible to develop/built with java 7 and
 still have java 5,6 as runtime environment, but looking at the code it does
 not seem easy (we would have to get the java version, and use different API
 calls).

Java is upwards compatible (except parts of JDBC).
So it should be possible to restrict the code to the Java 5 API and it
would then work on Java 6 and 7 and 8 etc.

 rgds
 jan I.



  This would prevent many queries on the Forum of why OO doesn't work with
 an installed Java.
 
 
  --
  Rory O'Farrell ofarr...@iol.ie
 
  -
  To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
  For additional commands, e-mail: dev-h...@openoffice.apache.org
 

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org



-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Proposal -- AOO 4.0.1 Release

2013-08-07 Thread sebb

On 7 August 2013 12:55, Jürgen Schmidt jogischm...@gmail.com wrote:
 On 8/7/13 1:51 PM, janI wrote:
 On 7 August 2013 13:07, Jürgen Schmidt jogischm...@gmail.com wrote:

 On 8/7/13 11:47 AM, janI wrote:
 On 7 August 2013 11:28, Jürgen Schmidt jogischm...@gmail.com wrote:

 On 8/6/13 6:42 PM, janI wrote:
 On 6 August 2013 17:15, Andrea Pescetti pesce...@apache.org wrote:

 Jürgen Schmidt wrote:

 On 8/6/13 3:05 PM, Andrea Pescetti wrote:

 It is important that we don't fall in the release and forget trap,
 i.e., this bug was already known when 4.0 was released, so it
 doesn't
 need to be evaluated again for 4.0.1. At least, we should
 re-evaluate
 the old proposed blockers: some of them might have become more
 relevant.

 in theory and with an idealistic view I would agree but for practical
 reason I don't. You should not forget that issues have to be fixed as
 well.
 We should really be careful here and should focus on the most serious
 issues only. From my point of view many proposed showstoppers for 4.0
 were no showstopper and why should we prioritize them now.


 We shouldn't prioritize them, just look at them again. My suggestion
 was
 to have regressions and old nominated blockers as PROPOSED blockers
 (status: ?), not as blockers (status: +). Some will have to be
 rejected
 again, obviously; but it is very bad, as a user and a community
 member,
 to
 get an answer like my (made up) example above. Of course, anybody who
 is
 concerned can propose an issue as a blocker, but a quick review makes
 sense
 in my opinion.


  we have volunteers who are ready to
 work and Pootle is not ready yet for their language, or it only
 offers
 3.4.1. See http://markmail.org/message/**4oxacrviktdbmbcv
 http://markmail.org/message/4oxacrviktdbmbcvfor more.

 where are the issues? Where are the volunteers to work on this?
 Nobody
 should plan with other peoples time and willingness


 One issue: https://issues.apache.org/ooo/**show_bug.cgi?id=122910
 https://issues.apache.org/ooo/show_bug.cgi?id=122910

 As for the volunteers, I understand that the Pootle update is a lot of
 work, as I wrote. Fact is, this lot of work is instrumental in
 attracting
 volunteers successfully and will remain the same amount of work
 whether
 done now or after 4.0.1. And doing it now (or soon) is a nice
 opportunity
 for the project for a combination of reasons: OpenOffice 4.0 had great
 exposure, volunteers want to translate it into their language, Summer
 is
 the best period for people to contribute in their spare time, telling
 someone that his efforts will be turned into an official release next
 month
 is very motivating... But indeed so far you are the only one who
 actually
 did this Pootle administration work.


 I can give a hand, with this work, but reading through the mails it
 seems
 we have quite a few open issues (mainly raised by jsc):
 - Should we make 4.01 in pootle or as suggested continue working on
 4.0 ?

 if we create a new project I would use 4.0.1

 I see you have created new project names and used again a new naming
 scheme, why?

 old aoo40

 new a00401

 This makes it not easier to get an overview

 I know, but this was just an experiment to test if I could copy the db
 easily. That did not work, so its the old way, as described below.




 - Do we want to add languages where we have no translation teams ?

 I would only add languages where we have an active translating
 community. We should save all other languages in a secure place and add
 them on demand or we create a further project where we add all inactive
 languages and keep them more or less up-to-date by merging to the latest
 templates


 so you dont agree with andrea, that argues (correctly) its a motivation
 factor to see that part of the language is already translated.

 also keep in mind, that genLang hopefully comes soon, then we need to
 convert the sdf files anyhow, not to loose the information.

 as I mentioned store them in a secure place or an additional project but
 away from the active ones. Simply reduced work and the motivation of
 people who actually do the work is important as well ;-)




 - How do we merge languages changed in pootle and sdf ?

 We should not merge sdf files back. We work with po files and use Pootle
 to manage them and get an overview where we are. Offline translation
 will be merged on Pootle first.

 we need to, first of all we have sdf files that have not been converted
 to
 po, second we have 3.4.1 po files that need to be updated from sdf to 4.0
 level.

 sure we have to do it ones but I talked more about the handling after
 this initial step



 And with your new translation tools sdf files become obsolete
 completely.


 yes, but thats just so much more reason to get all sdf files synchronized
 now.

 I think I said this already. We have to convert them all in po, merge
 against the latest templates from 4.0 and safe them in a secure
 place/project and use new languages on demand


 No problem, I would have

Re: [discuss] drop support for Java 5 and Java 6 for Windows

2013-08-07 Thread sebb

On 7 August 2013 13:47, janI j...@apache.org wrote:
 On 7 August 2013 14:02, Rory O'Farrell ofarr...@iol.ie wrote:

 On Wed, 7 Aug 2013 07:54:55 -0400
 Rob Weir robw...@apache.org wrote:

  On Wed, Aug 7, 2013 at 3:44 AM, Oliver-Rainer Wittmann
  orwittm...@googlemail.com wrote:
   Hi,
  
   I would like to discuss here, if we drop the support for Java 5 and
 Java 6
   for AOO installation on Windows.
   Important note for discussion: it is all about platform Windows.
  
   On my work to update the AOO build environment for Windows I
 recognized that
   it is hard to get an official JDK 1.5 (Java 5) or JDK 1.6 (Java 6) for
   Windows. Thus, I decided to go with JDK 1.7. The resulting AOO
 installation
   on Windows no longer works together with an JRE 6. It does not
 recognize an
   installed JRE 6 as an valid Java runtime environment.
   Thus, it comes into my mind to drop the support for Java 5 and Java 6
 for
   Windows.
  
 
  Another perspective to consider:   What leads to the most secure build
  environment for our binaries?   I don't think we want to ever be
  building binaries that millions of users download, that are built on a
  machine with an unsupported JRE that is no longer receiving security
  patches.  A build machine should be full patched against known
  security issues.  And of course it should be used only for building,
  not for daily email and web browsing.
 
  So I think we should use JDK 1.7 for our builds.  If that breaks Java
  5 and Java 6 compatibility for end users, then this is unfortunate,
  but justifiable.
 
  We'll face a similar issue of this sort in April 2014, when Windows XP
  reaches End of Support by Microsoft.  Do we then continue to support
  AOO on XP?  Do we test with it?  Perhaps.  But we certainly would not
  build our binaries on XP, right?
 
  Regards,
 
  -Rob
 
 
 
   Some discussion already took place in the thread about my update on
 the AOO
   build environment for Windows. Here are the original statements:
   cite
  
From Oliver:
   On a Windows system with JRE 6 the installation of my build does
 not
   recognize installed JRE 6 as an Java runtime environment (Menu -
 Tools
   -
   Option - Java). This is no problem from my point of view as our
 Windows
   users should not have JRE 6 installed anymore on their systems due
 to
   its security risks. Does somebody contradicts?
  
  
   From Andrea:
   As far as I know, this would be a significant limitation. We can now
   build with Java 5, 6 or 7 and the build can work with Java 5, 6 or 7
   (regardless of the version used for building). Restricting this
 would
   require discussion, especially on less common platforms.
  
  
   From Oliver:
   I agree that it would be a restriction, but due to the security
 risks of
   Oracle's JRE 6 I do not think that such a restriction hurts. In
 contrast
   it would 'help' our Windows users to update their Java environment.
  
   Thus, let us start a new thread to discuss this topic.
  
  
   FromJürgen:
   we should think how relevant it is and if we have more work to support
   it. As Oliver pointed out, the latest security problems of Java result
   in probably many updated systems. I don't see that Java 5 or 6 is
   important in the future and we should focus on the future.
  
   /cite
  
   My arguements for a drop of the Java 5 and Java 6 support on Windows
 are:
   - JRE 5 is quite old and no longer officially available
   - JRE 6 is no longer officially available
   - JRE 6 has certain security risks and the corresponding tools on
 Widnows
   are reporting to update to JRE 7
   - Simplify our work as we do not need to test under Windows JRE 5 and
 JRE 6
  
   I currently see no need to support JRE 5 or JRE 6 in our future
 releases for
   Windows.
  
   Let us discuss openly more Pros and Cons on this topic.

 I see no objection to future AOO releases requiring at least Java 1.7;
 ideally I would wish it continued to support XP (I suppose Win 2K is too
 much to hope for?).


 I am all in favour of remove java 1.5 and 1.6 from our build environment,
 as they are a security risk. But we should exploit if its possible to use
 java 1.7 for build and accept java 5,6,7 as runtime.

 I have looked a bit in the code (I am no java guru), and I can see the APIs
 differentiate. We could overcome that with a couple of if statements
 (testing for version), question is do we also want to force our customers
 to use java 1.7 ?

Where is this code?
Not sure I'd class myself as a Java guru, but I have done quite a bit
of work on Java compatibilty issues, so I might be able to help.

 rgds
 jan I.



 --
 Rory O'Farrell ofarr...@iol.ie

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org



-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For

Re: Proposal -- AOO 4.0.1 Release

2013-08-06 Thread sebb

On 6 August 2013 06:51, O.Felka olaf-openoff...@gmx.de wrote:
 Am 05.08.2013 23:34, schrieb David Gerard:

 On 5 August 2013 22:32, janI j...@apache.org wrote:

 When I consider what I hear in the real world, I would prefer a fast
 release, solving the most important issues. We always have the
 possibility
 to make a 4.02 if really needed.



 x.0.x releases monthly are the way to go. I think LibreOffice really
 got this right. If you do this, the users will be happier because they
 anticipate happiness in the *near* future, not a year from now.


 If we get quality we don't need quantity. We don't have to bother the user
 with monthly installations.
 The majority of feedback I remember is: 'We don't want all these new
 features but we want it stable.'
 Monthly updates are a pain in business areas.

So long as it is very clear what the new release contains, does it
matter how frequent they are?
A business will update to the next release if it contains fixes for
problems that are causing them problems, otherwise they will wait.

 Groetjes,
 Olaf



 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [IDEA] Back to School with AOO blog post

2013-08-06 Thread sebb

On 6 August 2013 15:56, Donald Whytock dwhyt...@gmail.com wrote:
 On Tue, Aug 6, 2013 at 10:50 AM, Ian Lynch ianrly...@gmail.com wrote:

 On 6 August 2013 15:40, Rob Weir robw...@apache.org wrote:

  It is that time of year now:  back to school.
 
  I wonder if this could make a good blog post.   Of course, AOO being
  free is core part of the story.  But even better would be if we had a
  top 10 list of ways in which AOO is great for school.
 


 1. Use the same software at home and at school without worrying about
 licensing
 2. Participate in the AOO community and provide useful things for others
 while you learn
 3. Fix broken Word documents
 4. Provide choice and promote transferrable skills by using more than one
 type of tool
 5. Get a qualification referenced to the European Qualifications Framework


 6. Fits nicely in a small/low income community's school budget.

7. Work in multiple languages.

 Don

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Download page could give approximate file size

2013-08-05 Thread sebb

On 4 August 2013 19:45, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/27/2013 10:14 PM, schrieb Marcus (OOo):

 Am 07/27/2013 10:10 PM, schrieb sebb:

 It might be helpful if the download page showed the approximate file
 size downloads.

 This would help users to know how long it might take (and do they have
 the space!) as well as offering an obvious sign if a download is
 truncated by more than a few kB. Not perfect, but would have helped
 the recent downloader.


 Right, this would be much more helpful than having nothing.

 I'll think about how to implement this into the current DL scripting.


 I've created some new scripting and the result can be seen here:

 http://ooo-site.staging.apache.org/download/test/index.html

That's good to have the size in the main download box.
However, given that the size is approximate, I'm not sure it's
necessary to provide it to two decimal places.
Either give the exact size, or round to the nearest MB.

The page shows me:

Windows (EXE) and English (British) (~129.89 MByte).

Once downloaded, the Windows XP property page shows:
129 MB (136,201,626 bytes)

So I think the download box should also show 129MB

Unhelpfully, the SF downloads page [1] shows 136.2MB - looks as though
they divided by 1000 rather than 1024 - though it's unlikely anyone
would/could use that page as the file names are all truncated.  I
don't know if it's possible to change the column sze for the name
column. Would be good if that could be corrected along with the size.

[1] 
http://sourceforge.net/projects/openofficeorg.mirror/files/4.0.0/binaries/en-GB/

Also, the KEYS/sigs/hashes ought to be shown before the alternate
downloads, ideally within the main green box as they relate to the
main download
The other downloads ought to be shown in a bit more detail in a separate box

 Marcus


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Download page could give approximate file size

2013-08-05 Thread sebb

On 5 August 2013 20:38, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 08/05/2013 07:11 PM, schrieb sebb:

 On 4 August 2013 19:45, Marcus (OOo)marcus.m...@wtnet.de  wrote:

 Am 07/27/2013 10:14 PM, schrieb Marcus (OOo):

 Am 07/27/2013 10:10 PM, schrieb sebb:


 It might be helpful if the download page showed the approximate file
 size downloads.

 This would help users to know how long it might take (and do they have
 the space!) as well as offering an obvious sign if a download is
 truncated by more than a few kB. Not perfect, but would have helped
 the recent downloader.



 Right, this would be much more helpful than having nothing.

 I'll think about how to implement this into the current DL scripting.



 I've created some new scripting and the result can be seen here:

 http://ooo-site.staging.apache.org/download/test/index.html


 That's good to have the size in the main download box.
 However, given that the size is approximate, I'm not sure it's
 necessary to provide it to two decimal places.
 Either give the exact size, or round to the nearest MB.

 The page shows me:

 Windows (EXE) and English (British) (~129.89 MByte).

 Once downloaded, the Windows XP property page shows:
 129 MB (136,201,626 bytes)

 So I think the download box should also show 129MB


 OK, I've eliminated the decimal places.


 Unhelpfully, the SF downloads page [1] shows 136.2MB - looks as though
 they divided by 1000 rather than 1024 - though it's unlikely anyone
 would/could use that page as the file names are all truncated.  I
 don't know if it's possible to change the column sze for the name
 column. Would be good if that could be corrected along with the size.


 Please have a look into this issue:
 https://issues.apache.org/ooo/show_bug.cgi?id=122233


 [1]
 http://sourceforge.net/projects/openofficeorg.mirror/files/4.0.0/binaries/en-GB/

 Also, the KEYS/sigs/hashes ought to be shown before the alternate
 downloads, ideally within the main green box as they relate to the
 main download


 I've tried to realize that. But I think it's not possible as there is only
 one link possible as the whole pper green box has to be clickable: for the
 install file or for one of the signature/hash files.

 The most important part is the install file. Everything else is secondary.

In that case, put the KEYS etc. in a separate box immediately under
the clickable box.


 The other downloads ought to be shown in a bit more detail in a separate
 box


 What do you mean with more details? What do you think is missing?

The links are very cramped, and there is no explanation of what they are for.

At present the page reads:

[LARGE GREEN DOWNLOAD BOX]
Get all platforms, languages, language packs | Source code and SDK |
Portable USB versions and third-party ports | Older and legacy
versions: 3.4.1 + 3.3.0 |
Signatures and hashes: KEYS , ASC , MD5 , SHA256 , How to verify? |
Release Notes

It should read something like:

[LARGE GREEN DOWNLOAD BOX]
Signatures and hashes: KEYS , ASC , MD5 , SHA256 , How to verify? |
Release Notes
blank line
then another different colour box containing something like
Looking for other downloads?
* Builds of OpenOffice 4.0.0 for other platforms and languages
* Additional Language packs for your OpenOffice installation
* The source code and the SDK (Software Development Kit)
* Portable USB versions and third-party ports of OpenOffice
* Older and Legacy versions [should probably link to a separate page
that explains the difference between 3.4.1 and 3.3.0]

==

The rest of the page is laid out with plenty of space between items.
At present the light green box looks out of place, and is tricky to
read unless you know exactly what you are looking for (and you
understand the jargon).

 Thanks

 Marcus


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Download page could give approximate file size

2013-08-05 Thread sebb

On 5 August 2013 22:23, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 08/05/2013 10:24 PM, schrieb sebb:

 On 5 August 2013 20:38, Marcus (OOo)marcus.m...@wtnet.de  wrote:

 Am 08/05/2013 07:11 PM, schrieb sebb:

 On 4 August 2013 19:45, Marcus (OOo)marcus.m...@wtnet.de   wrote:


 Am 07/27/2013 10:14 PM, schrieb Marcus (OOo):

 Am 07/27/2013 10:10 PM, schrieb sebb:



 It might be helpful if the download page showed the approximate file
 size downloads.

 This would help users to know how long it might take (and do they
 have
 the space!) as well as offering an obvious sign if a download is
 truncated by more than a few kB. Not perfect, but would have helped
 the recent downloader.




 Right, this would be much more helpful than having nothing.

 I'll think about how to implement this into the current DL scripting.




 I've created some new scripting and the result can be seen here:

 http://ooo-site.staging.apache.org/download/test/index.html



 That's good to have the size in the main download box.
 However, given that the size is approximate, I'm not sure it's
 necessary to provide it to two decimal places.
 Either give the exact size, or round to the nearest MB.

 The page shows me:

 Windows (EXE) and English (British) (~129.89 MByte).

 Once downloaded, the Windows XP property page shows:
 129 MB (136,201,626 bytes)

 So I think the download box should also show 129MB



 OK, I've eliminated the decimal places.


 Unhelpfully, the SF downloads page [1] shows 136.2MB - looks as though
 they divided by 1000 rather than 1024 - though it's unlikely anyone
 would/could use that page as the file names are all truncated.  I
 don't know if it's possible to change the column sze for the name
 column. Would be good if that could be corrected along with the size.



 Please have a look into this issue:
 https://issues.apache.org/ooo/show_bug.cgi?id=122233


 [1]

 http://sourceforge.net/projects/openofficeorg.mirror/files/4.0.0/binaries/en-GB/

 Also, the KEYS/sigs/hashes ought to be shown before the alternate
 downloads, ideally within the main green box as they relate to the
 main download



 I've tried to realize that. But I think it's not possible as there is
 only
 one link possible as the whole pper green box has to be clickable: for
 the
 install file or for one of the signature/hash files.

 The most important part is the install file. Everything else is
 secondary.


 In that case, put the KEYS etc. in a separate box immediately under
 the clickable box.


 I've rearranged the content of the sub-green box. Now the hash links are on
 top.


 The other downloads ought to be shown in a bit more detail in a separate
 box



 What do you mean with more details? What do you think is missing?


 The links are very cramped, and there is no explanation of what they are
 for.


 Yes, that's right. A new format how these are presented to the users is
 needed.


 At present the page reads:

 [LARGE GREEN DOWNLOAD BOX]
 Get all platforms, languages, language packs | Source code and SDK |
 Portable USB versions and third-party ports | Older and legacy
 versions: 3.4.1 + 3.3.0 |
 Signatures and hashes: KEYS , ASC , MD5 , SHA256 , How to verify? |
 Release Notes

 It should read something like:

 [LARGE GREEN DOWNLOAD BOX]
 Signatures and hashes: KEYS , ASC , MD5 , SHA256 , How to verify? |
 Release Notes
 blank line
 then another different colour box containing something like
 Looking for other downloads?
 * Builds of OpenOffice 4.0.0 for other platforms and languages
 * Additional Language packs for your OpenOffice installation
 * The source code and the SDK (Software Development Kit)
 * Portable USB versions and third-party ports of OpenOffice
 * Older and Legacy versions [should probably link to a separate page
 that explains the difference between 3.4.1 and 3.3.0]


 OK. Let's see how this can be solved. This will be a bigger task. I'll put
 this into the Wiki for improvement.


 ==

 The rest of the page is laid out with plenty of space between items.
 At present the light green box looks out of place, and is tricky to


 In general, because it's also green I don't think that it is out of place.
 The connection to the normal green box - and therefore to the most current
 release - should be obvious and not that it is something different. But ...

By looks out of place I meant that the styling/layout of the box
does not agree with the rest of the page.


 read unless you know exactly what you are looking for (and you
 understand the jargon).


 ... yes, that's right. A new format could be better.

 Marcus


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Release 3.4.1 storage and incubator removal.

2013-07-31 Thread sebb

On 31 July 2013 21:31, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/31/2013 03:06 PM, schrieb Rob Weir:

 On Wed, Jul 31, 2013 at 7:33 AM, janIj...@apache.org  wrote:

 Moving conversation to dev@


 On 31 July 2013 13:28, Jürgen Schmidtjogischm...@gmail.com  wrote:


 On 7/31/13 1:07 PM, janI wrote:

 Hi.

 Based on a discussion today on IRC, I would like to draw your attention
 to the following challenge. henkp is cc because he is the infra person
 doing the rsync magic.

 ASF has a policy that incubator/xxx should be removed when the project
 graduates. We still have our 3.4.1 (and 3.3.0) release stored under
 incubator.

 As part of cleaning rsync, infra want to enforce the policy, but of
 course respect and understand our need to have 3.4.1 available to users
 (especially because 3.4.1 contains languages not released in 4.0).

 I see the following possibilities:

 1) remove openoffice from incubator, but leave version 3.3.0 and 3.4.1
 (with language packs) on the SF mirror. This is preferred by infra.


 +1 to keep it on SF


 2) remove 3.3.0 since its a legacy version, and move 3.4.1 parallel to
 4.0. This is however an expensive operation for all mirror, and should
 only be done if we anticipate patches for 3.4.1


 Removing 3.3 is ok to me but I see demand for keeping 3.4.1. Having it
 besides 4.0 would be natural but needs some work in the download scripts
 ...



 Just for me, do you prefer 1) or 2) ... I personally dont see a big need
 to
 keep 3.4.1 on our servers.


 INHO it is fine to keep 3.4.1 and early on archive.apache.org and SF
 only, provided Infra is OK with us pointing our website links for the
 hashes and signatures of 3.4.1 and earlier to archive and not to the
 dist.  I don't think the bandwidth will be significant for these.


 I was told that these files must be linked from www.apache.org/dist/ only
 and always. If we can get now an exception for archive.apache.org/dist/ -
 hey, it's hosted by Apache, too :-) - this would be good.

www.apache.org/dist/ - current release(s)
archive.apache.org/dist/ - archived releases

See:
http://www.apache.org/dev/mirrors.html#location
http://www.apache.org/dev/release.html#when-to-archive

 Marcus




 3) persuade infra to keep incubator for 3.4.1, but limit the footprint
 as much as possible, remove 3.3.0 and put a timelimit up.


 I believe we can manage to move it out of the incubator in some way.

 We are also adviced, that if/when we change our layout infra need to be
 adviced well in advance. In my opinion we should consider not using
 externaldist, but have the total release in one folder with subfolders.


 Well externaldist was not our idea and I copied the files in this
 structure of advice from infra. We should first clarify what's preferred
 here. externaldist caused some confusion and extra work on our side as
 well.


 We can move the discussion to dev@ is nobody objects.


 yes let's move to dev

 Juergen


 rgds
 jan I.


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [PROPOSAL] New other.html for all possible download links

2013-07-30 Thread sebb

On 31 July 2013 01:41, Dave Fisher dave2w...@comcast.net wrote:

 On Jul 30, 2013, at 3:43 PM, Marcus (OOo) wrote:

 Am 07/30/2013 12:25 AM, schrieb Kay Schenk:
 On Mon, Jul 29, 2013 at 3:21 PM, janIj...@apache.org  wrote:

 On 30 July 2013 00:16, Marcus (OOo)marcus.m...@wtnet.de  wrote:

 Am 07/29/2013 10:39 PM, schrieb sebb:

 On 29 July 2013 21:12, Marcus (OOo)marcus.m...@wtnet.de   wrote:

 Am 07/29/2013 09:45 PM, schrieb sebb:

  On 29 July 2013 19:27, Marcus (OOo)marcus.m...@wtnet.dewrote:


 Am 07/26/2013 11:10 PM, schrieb Marcus (OOo):



 Am 07/26/2013 10:44 PM, schrieb Andrea Pescetti:



 On 25/07/2013 Marcus (OOo) wrote:



 I've created a new webpage to offer all possible download links
 for
 a
 release version:
 http://ooo-site.staging.**apache.org/download/test/**
 other_tables.html
 http://ooo-site.staging.apache.org/download/test/other_tables.html




 This is really nice, looking forward to seeing it online!




 see below

  It's especially important to have a link to checksums for those who
 want
 a static reference to it or want to verify a package on a
 different
 system than the one used for downloading.




 That was one intension, yes.

  - all possible downloads for a respective language and OS in a
 single
 place




 I foresee another interesting way to use that page, i.e., getting
 more
 localization volunteers. Let's get this version online first, but
 maybe
 we could then add another table with something like The following
 languages are released only as source code:, and then a list of
 each
 of
 the 90+ remaining languages with the link to help us release it (in
 most
 cases, it will be a link to http://openoffice.apache.org/**
 translate.htmlhttp://openoffice.apache.org/translate.html
 but in some cases it might be different).




 Good idea. However, I think I've to add more additions than one can
 thought. But this is no obstacle.

  @Andrea:
 I've already considered your smaller font size wish for the
 checksum
 links.




 Thanks, looks great.




 Thanks. :-)

 Even when I've missed to state it from the beginning but I expect to
 use
 lazy consensus here. If there are no objections I plan to make it
 Live
 at ~Sunday evening Hamburg time.




 As I haven't seen any objections I'll create the new other.html in
 a
 the
 next time.



 Sorry, but I find the page hard to use.

 Most people will not need any language packs, so why clutter the table
 with them?

 Also if a user does want to add multiple language packs, it's hard
 work finding them amongst all the full installations.

 I think it would be a lot clearer for the page to be laid out
 something like the following:



 This was the old system and the goal was to integrate all files that
 belong
 to a specific language and platform.


 I'm not sure that goal is particularly useful to the end-user.

  If it's hard to read due to a small font size, this could be changed.


 It's not the font size.

  Otherwise I don't thing that it's too confusing.


 Well, you are a developer working on OOo.


 Only for the website, not for the source code. But maybe this is no
 longer
 relevant as I do this already for years.

  I am trying to look at it as a non-developer who wants to download the
 software.

   [ cut here ] ---


 ...

  [ cut here ] ---


 If it is possible to provide a dynamic page, then it might be nice to
 determine the platform first (user selected; perhaps with
 auto-detected default), and then use the platform to display only the
 installation sets and language packs for that platform.



 The dynamic thing is not to continue the data guessing from the main
 download webpage but to simplify the modification for new releases.

 The other.html is a kind of fallback when:

 a) the user is not able to use the green box on the previous main
 download
 webpage.

 b) or when he is searching for a build different from the browser's
 language
 / platform.


 AFAICT it's also used when the user wants to add a new language, in
 which case they already have the base installation.


 Yes, and there are maybe some more possibilities.

  As we don't know the reason(s) for a) there shouldn't be any limitations
 to
 give the user the full control to find what he needs.


 Yes, but that's not relevant to the issue of the page design.

  And if b) it will help him as well.


 Which is where the page design is very important.

  The difference of full installations and language packs is described
 directly above the table by your suggestion.


 Yes, but I'm afraid I don't find it easy to read.
 There's quite a lot of information there which is not particularly
 relevant to the end user.

 Also the most common use case - downloading a single base installation
 and no languag packs - is not actually described.


 If you think that the text can be better, then please tell me. Based on
 Apache's famous slogan: Patches are welcome. :-)

  That would avoid problems

Re: [PROPOSAL] New other.html for all possible download links

2013-07-29 Thread sebb

On 29 July 2013 19:27, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/26/2013 11:10 PM, schrieb Marcus (OOo):

 Am 07/26/2013 10:44 PM, schrieb Andrea Pescetti:

 On 25/07/2013 Marcus (OOo) wrote:

 I've created a new webpage to offer all possible download links for a
 release version:
 http://ooo-site.staging.apache.org/download/test/other_tables.html


 This is really nice, looking forward to seeing it online!


 see below

 It's especially important to have a link to checksums for those who want
 a static reference to it or want to verify a package on a different
 system than the one used for downloading.


 That was one intension, yes.

 - all possible downloads for a respective language and OS in a single
 place


 I foresee another interesting way to use that page, i.e., getting more
 localization volunteers. Let's get this version online first, but maybe
 we could then add another table with something like The following
 languages are released only as source code:, and then a list of each of
 the 90+ remaining languages with the link to help us release it (in most
 cases, it will be a link to http://openoffice.apache.org/translate.html
 but in some cases it might be different).


 Good idea. However, I think I've to add more additions than one can
 thought. But this is no obstacle.

 @Andrea:
 I've already considered your smaller font size wish for the checksum
 links.


 Thanks, looks great.


 Thanks. :-)

 Even when I've missed to state it from the beginning but I expect to use
 lazy consensus here. If there are no objections I plan to make it Live
 at ~Sunday evening Hamburg time.


 As I haven't seen any objections I'll create the new other.html in a the
 next time.

Sorry, but I find the page hard to use.

Most people will not need any language packs, so why clutter the table
with them?

Also if a user does want to add multiple language packs, it's hard
work finding them amongst all the full installations.

I think it would be a lot clearer for the page to be laid out
something like the following:

 [ cut here ] ---

Apache OpenOffice 4.0.0 - Base installation


Please choose one of the following downloads according to your primary
language and the platform on which it will be installed.

If you want to use multiple languages, you can download additional language
packs - see below for details.

TABLE OF FULL INSTALLs
...

Apache OpenOffice 4.0.0 - Additional Language packs
===

The base installation includes a single language (e.g. perhaps you
downloaded the Spanish base installation).

If you want to use multiple languages (e.g. you also want Galician), you can
download the appropriate language pack(s) and add them to the existing
installation.

The language can be changed via the menu Tools - Options - Language Settings -
Languages - User Interface.

Note that language packs are only suitable for a specific version of OpenOffice
(same version and platform) so make sure you choose the correct
platform (and version).

You can add as many additional language packs as you want to the base
installation.

TABLE OF ADDITIONAL LANGUAGE PACKS
...

 [ cut here ] ---


If it is possible to provide a dynamic page, then it might be nice to
determine the platform first (user selected; perhaps with
auto-detected default), and then use the platform to display only the
installation sets and language packs for that platform.

That would avoid problems with people downloading the language pack
for the wrong platform.

I think it would make for a better end-user experience.

 Marcus


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [PROPOSAL] New other.html for all possible download links

2013-07-29 Thread sebb

On 29 July 2013 21:12, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/29/2013 09:45 PM, schrieb sebb:

 On 29 July 2013 19:27, Marcus (OOo)marcus.m...@wtnet.de  wrote:

 Am 07/26/2013 11:10 PM, schrieb Marcus (OOo):


 Am 07/26/2013 10:44 PM, schrieb Andrea Pescetti:


 On 25/07/2013 Marcus (OOo) wrote:


 I've created a new webpage to offer all possible download links for a
 release version:
 http://ooo-site.staging.apache.org/download/test/other_tables.html



 This is really nice, looking forward to seeing it online!



 see below

 It's especially important to have a link to checksums for those who
 want
 a static reference to it or want to verify a package on a different
 system than the one used for downloading.



 That was one intension, yes.

 - all possible downloads for a respective language and OS in a single
 place



 I foresee another interesting way to use that page, i.e., getting more
 localization volunteers. Let's get this version online first, but maybe
 we could then add another table with something like The following
 languages are released only as source code:, and then a list of each
 of
 the 90+ remaining languages with the link to help us release it (in
 most
 cases, it will be a link to http://openoffice.apache.org/translate.html
 but in some cases it might be different).



 Good idea. However, I think I've to add more additions than one can
 thought. But this is no obstacle.

 @Andrea:
 I've already considered your smaller font size wish for the checksum
 links.



 Thanks, looks great.



 Thanks. :-)

 Even when I've missed to state it from the beginning but I expect to use
 lazy consensus here. If there are no objections I plan to make it Live
 at ~Sunday evening Hamburg time.



 As I haven't seen any objections I'll create the new other.html in a
 the
 next time.


 Sorry, but I find the page hard to use.

 Most people will not need any language packs, so why clutter the table
 with them?

 Also if a user does want to add multiple language packs, it's hard
 work finding them amongst all the full installations.

 I think it would be a lot clearer for the page to be laid out
 something like the following:


 This was the old system and the goal was to integrate all files that belong
 to a specific language and platform.

I'm not sure that goal is particularly useful to the end-user.

 If it's hard to read due to a small font size, this could be changed.

It's not the font size.

 Otherwise I don't thing that it's too confusing.

Well, you are a developer working on OOo.

I am trying to look at it as a non-developer who wants to download the software.

  [ cut here ] ---


 ...

  [ cut here ] ---


 If it is possible to provide a dynamic page, then it might be nice to
 determine the platform first (user selected; perhaps with
 auto-detected default), and then use the platform to display only the
 installation sets and language packs for that platform.


 The dynamic thing is not to continue the data guessing from the main
 download webpage but to simplify the modification for new releases.

 The other.html is a kind of fallback when:

 a) the user is not able to use the green box on the previous main download
 webpage.

 b) or when he is searching for a build different from the browser's language
 / platform.

AFAICT it's also used when the user wants to add a new language, in
which case they already have the base installation.

 As we don't know the reason(s) for a) there shouldn't be any limitations to
 give the user the full control to find what he needs.

Yes, but that's not relevant to the issue of the page design.

 And if b) it will help him as well.

Which is where the page design is very important.

 The difference of full installations and language packs is described
 directly above the table by your suggestion.

Yes, but I'm afraid I don't find it easy to read.
There's quite a lot of information there which is not particularly
relevant to the end user.

Also the most common use case - downloading a single base installation
and no languag packs - is not actually described.


 That would avoid problems with people downloading the language pack
 for the wrong platform.


 Sorry, but this can also happen in the current other.html.


Of course; I was just making a suggestion to improve the page further.


 I think it would make for a better end-user experience.


 I would say: Let the users decide. :-) If we get a reasonable amount of
 complains then we can go back to different tables.


 Marcus


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

[Website] http://www.openoffice.org/download/legacy/index.html

2013-07-27 Thread sebb

I hope this is the right place to report this. There does not seem to
be a Bugzilla category for website content, and Contact Us only
refers to website technical issues.

http://www.openoffice.org/download/legacy/index.html

has the following sentences:

1) It's highly recommended not to use them but only when there is no
better choice.

2) OpenOffice.org is licensed under the GNU Lesser General Public
License v3 (LGPL v3).

==

1) does not read well, it should be something like:

It's highly recommended not to use them *except* when there is no
better choice.

2) is very ambiguous. To what does OpenOffice.org refer? The
website? The phrase? Some legacy software?

Please can it be clarified?

Thanks.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

[website] http://www.openoffice.org/qa/issue_handling/submission_gateway.html

2013-07-27 Thread sebb

There is a broken link on

http://www.openoffice.org/qa/issue_handling/submission_gateway.html

The link entitled

... the OpenOffice.org infrastructure(including the website, the wiki,
mailing lists, etc.)

leads to

https://issues.apache.org/ooo/enter_bug.cgi?product=wwwversion=current

which responds with

Sorry, either the product www does not exist or you aren't authorized
to enter a bug into it.

[I was logged in at the time]

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [Website] http://www.openoffice.org/download/legacy/index.html

2013-07-27 Thread sebb

On 27 July 2013 11:08, sebb seb...@gmail.com wrote:
 I hope this is the right place to report this. There does not seem to
 be a Bugzilla category for website content, and Contact Us only
 refers to website technical issues.

 http://www.openoffice.org/download/legacy/index.html

 has the following sentences:

 1) It's highly recommended not to use them but only when there is no
 better choice.

The above text also appears on http://www.openoffice.org/download/archive.html

 2) OpenOffice.org is licensed under the GNU Lesser General Public
 License v3 (LGPL v3).

 ==

 1) does not read well, it should be something like:

 It's highly recommended not to use them *except* when there is no
 better choice.

 2) is very ambiguous. To what does OpenOffice.org refer? The
 website? The phrase? Some legacy software?

 Please can it be clarified?

 Thanks.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: unassign BZ issues.

2013-07-27 Thread sebb

On 27 July 2013 11:47, janI j...@apache.org wrote:
 Hi

 @rob, you are the BZ expert, is it possible to automate a BZ process, that
 unassigns BZ issues, assigned to a person but untouched for a year ?

 see
 https://issues.apache.org/ooo/show_bug.cgi?id=122035

 I think its important, that issues not being worked on are kept in the
 global pool so it is obvious that anyone can work on it.

In the main instance, it's not possible to assign issues, because once
they have been assigned, Bugzilla only sends change notifications to
the new assignee.

Or maybe that particular Bugzilla 'feature' has been fixed in your Bugzilla?

Easy enough to check - create a dummy issue and assign it; make some
changes and see where the mails get sent.

 rgds
 jan I.

-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: unassign BZ issues.

2013-07-27 Thread sebb

On 27 July 2013 12:10, janI j...@apache.org wrote:
 On 27 July 2013 12:54, sebb seb...@gmail.com wrote:

 On 27 July 2013 11:47, janI j...@apache.org wrote:
  Hi
 
  @rob, you are the BZ expert, is it possible to automate a BZ process,
 that
  unassigns BZ issues, assigned to a person but untouched for a year ?
 
  see
  https://issues.apache.org/ooo/show_bug.cgi?id=122035
 
  I think its important, that issues not being worked on are kept in the
  global pool so it is obvious that anyone can work on it.

 In the main instance, it's not possible to assign issues, because once
 they have been assigned, Bugzilla only sends change notifications to
 the new assignee.


 I thought it was possible to make a report containing all issues not
 touched a year and not assigned to iss...@openoffice.apache.org, and then
 use this report as basis for a script (or what its called in BZ) that
 assigns the issues to iss...@openoffice.apache.org

 It was not a question of where the mails get sent to, but more to have an
 updated list of who works on what.

Yes, I realise that.

But should you be allowing issues to be assigned away from the
devloper list at all?

Does it not matter that changes to such issues are no longer reported
to the developer list?

 rgds
 jan I.



 Or maybe that particular Bugzilla 'feature' has been fixed in your
 Bugzilla?

 Easy enough to check - create a dummy issue and assign it; make some
 changes and see where the mails get sent.




  rgds
  jan I.

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org



-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: unassign BZ issues.

2013-07-27 Thread sebb

On 27 July 2013 13:56, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/27/2013 02:16 PM, schrieb sebb:

 On 27 July 2013 12:10, janIj...@apache.org  wrote:

 On 27 July 2013 12:54, sebbseb...@gmail.com  wrote:

 On 27 July 2013 11:47, janIj...@apache.org  wrote:

 Hi

 @rob, you are the BZ expert, is it possible to automate a BZ process,

 that

 unassigns BZ issues, assigned to a person but untouched for a year ?

 see
 https://issues.apache.org/ooo/show_bug.cgi?id=122035

 I think its important, that issues not being worked on are kept in the
 global pool so it is obvious that anyone can work on it.


 In the main instance, it's not possible to assign issues, because once
 they have been assigned, Bugzilla only sends change notifications to
 the new assignee.


 I thought it was possible to make a report containing all issues not
 touched a year and not assigned to iss...@openoffice.apache.org, and
 then
 use this report as basis for a script (or what its called in BZ) that
 assigns the issues to iss...@openoffice.apache.org

 It was not a question of where the mails get sent to, but more to have an
 updated list of who works on what.


 Yes, I realise that.

 But should you be allowing issues to be assigned away from the
 devloper list at all?


 We have a dedicated mailing list only for BZ issues. That should be enough.
 Everybody who is interested in these mails can subscribe there.

Yes.


 Does it not matter that changes to such issues are no longer reported
 to the developer list?


 They were never reported to dev@. And, to be honest, I would not prefer it.
 We have sometimes more than 200 mails from issues per day. Do you want to
 have them all on a dev@ mailing list? ;-)

Yes - by that I mean a mailing list that devs can subscribe to; does
not have to be the dev discussion list.

 At least here in this dev@ not everybody is a developer and works on code.
 I'm sure that is different with other Apache projects.

Sorry, I was a bit lax: when I wrote the devloper list, I meant the
developer list used for BZ mails

 Marcus




 Or maybe that particular Bugzilla 'feature' has been fixed in your
 Bugzilla?

 Easy enough to check - create a dummy issue and assign it; make some
 changes and see where the mails get sent.




 rgds
 jan I.


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [website] http://www.openoffice.org/qa/issue_handling/submission_gateway.html

2013-07-27 Thread sebb

On 27 July 2013 15:25, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/27/2013 12:11 PM, schrieb sebb:

 There is a broken link on

 http://www.openoffice.org/qa/issue_handling/submission_gateway.html

 The link entitled

 ... the OpenOffice.org infrastructure(including the website, the wiki,
 mailing lists, etc.)

 leads to

 https://issues.apache.org/ooo/enter_bug.cgi?product=wwwversion=current

 which responds with

 Sorry, either the product www does not exist or you aren't authorized
 to enter a bug into it.

 [I was logged in at the time]


 There were some recent changes with the associations of products and
 categories. I've corrected the link.

 Thanks for the hint.

Thanks, that works now.

 Marcus


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [Website] http://www.openoffice.org/download/legacy/index.html

2013-07-27 Thread sebb

Thanks for the very prompt response, all looks good now.

On 27 July 2013 15:16, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/27/2013 12:15 PM, schrieb sebb:

 On 27 July 2013 11:08, sebbseb...@gmail.com  wrote:

 I hope this is the right place to report this. There does not seem to
 be a Bugzilla category for website content, and Contact Us only
 refers to website technical issues.

 http://www.openoffice.org/download/legacy/index.html

 has the following sentences:

 1) It's highly recommended not to use them but only when there is no
 better choice.


 The above text also appears on
 http://www.openoffice.org/download/archive.html


 Thanks, also here it's corrected.

 Marcus




 2) OpenOffice.org is licensed under the GNU Lesser General Public
 License v3 (LGPL v3).

 ==

 1) does not read well, it should be something like:

 It's highly recommended not to use them *except* when there is no
 better choice.

 2) is very ambiguous. To what does OpenOffice.org refer? The
 website? The phrase? Some legacy software?

 Please can it be clarified?

 Thanks.


 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: Checksum doesn't match

2013-07-27 Thread sebb

On 27 July 2013 20:48, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Am 07/27/2013 08:10 PM, schrieb Maritta Coppieters:

 Hi, trying to download open office 4.0.0, can't get it to install.  It
 looks like the md5 checksum doesn't match, any known issues?

 I'm on a Toshiba Satellite L855D 64 bit, running Windows 7, downloading
 with Firefox 22.  Download file is
 Apache_OpenOffice_4.0.0_Win_x86_install_en-US, 27,601KB.


 The download seems to be incomplete because it has to be ~143.4 MB. So, your
 file size is not correct.

I've seen this before with FIrefox - the download appears to have
completed successfully but is in fact incomplete. No error is
reported.

 Make sure you download from the official website:

 http://www.openoffice.org/download/other.html

 And you should compare the MD5 checksum of the downloaded file with the hash
 value from this file:

 http://www.apache.org/dist/openoffice/4.0.0/binaries/en-US/Apache_OpenOffice_4.0.0_Win_x86_install_en-US.exe.md5

 If needed, here is a howto description for using checksums:

 http://www.openoffice.org/download/checksums/3.4.1_checksums.html#howto

 I hope this will help you.

 Marcus

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: [Request for help][Download] Download AOO via mobile devices?

2013-07-27 Thread sebb

On 27 July 2013 21:19, Marcus (OOo) marcus.m...@wtnet.de wrote:
 Sometimes we get comments from users like I cannot find AOO for my iPhone
 or similar.

 Instead of leaving them alone I've started to update the DL scripting to
 recognize platforms and operating systems that AOO does not support. But
 instead of give an error back or just do nothing, the scripting shows that
 the release for the (hopefully currectly) recognized platform / OS is not
 available and points to the other.html webpage.

 Now I would like to ask you to test this with devices and OSes you have
 available to see how it works and where it has to be improved:

 http://ooo-site.staging.apache.org/download/test/index.html

 a) Is the device or OS recognized and stated correctly?
 b) Which combinations are not recognized and instead the term unknown OS
 is shown?

 Please remember:
 The analyze webpage will help to see what is recognized and actually shown:

 http://ooo-site.staging.apache.org/download/test/analyze.html

 Please open BZ issues for these problems:

 https://issues.apache.org/ooo/

Maybe a single BZ issue would be better than one per reporter?

There could be a lot of issues created; it's likely there will be many
duplicates.

Perhaps create a generic issue and link to it from this email thread.

 Thank you very much for you help.

 Marcus

 -
 To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
 For additional commands, e-mail: dev-h...@openoffice.apache.org


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Re: dev@openoffice.apache.org

2013-07-27 Thread sebb

Maybe the link to the user documentation should be a lot more obvious
on the download page ... or could the download process take the user
to the documentation area while (or after) the download completes?

On 27 July 2013 21:56, Clarence M Weaver cmweav...@gmail.com wrote:
 You can find the manual and other documentation at 
 http://wiki.openoffice.org/wiki/Documentation

 Gook luck
 Clarence

 On Jul 27, 2013, at 4:48 PM, Judith Hartman judithhart...@yahoo.com wrote:

 My husband and I downloaded the above software as our Mac never came with 
 any type of word processor. Amazon offered this software: 
 dev@openoffice.apache.org
 but we need a manual so we can use its features. We have tried to figure out 
 the different things to do but no idea how to use columns, tabs, etc. So any 
  help would be very appreciated.
 Thank you,
 Judith Hartman


-
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

88 matches

Mail list logo