Re: English Dictionaries update cycle

[Keith N. McKenna]

Marcus wrote:
> Am 05/16/2016 11:46 PM, schrieb Marco A.G.Pinto:
>> The other day someone complained in the extension page that he had
>> updates all the time.
>>
>> I replied to that person saying that I do monthly updates of the British
>> speller adding 600+ words.
> 
> Interesting, I wouldn't expect such a complaint for a spell checker. ;-)
> 
[knmc]

I suspect the reason for the complaint is because the extension includes
multiple variants of the English language, but the monthly updates are
only to the British variant which is a fairly small part of the total
downloads of the English language. That leaves those that use other
variants having to upgrade the extension for little gain or to put up
with the nag every time AOO is started about there being extension
upgrades. For example I use the en-US dictionaries and the last time it
was updated was January of 2016. Since that time there have been 4
updates to the oxt file that I have had no need for.


> My 2ct.
> 
> Marcus




signature.asc
Description: OpenPGP digital signature

Re: English Dictionaries update cycle

[Marcus]

Am 05/16/2016 11:46 PM, schrieb Marco A.G.Pinto:

The other day someone complained in the extension page that he had
updates all the time.

I replied to that person saying that I do monthly updates of the British
speller adding 600+ words.


Interesting, I wouldn't expect such a complaint for a spell checker. ;-)


I talked with JZA about the update frequency and I would like to know
your opinion about the update cycle (JZA suggested a 6 month cycle but I
need more opinions).

For Mozilla (Firefox + Thunderbird) the monthly is okay since it is a
small file (400-500 kB) with auto-update and only the British. Also,
every six weeks  a Mozilla upgrade comes out  and people want the
dictionary to be compatible with the latest alphas.

What is your opinion?


I would ask the complainer what he would suggest. Then add the opinions 
from others. At the end you should take this into account and decide for 
yourself what the best update frequence is.


As an extra you can write a short note at the extension webpage about 
the reason for your update frequence. Then others know your motivation.


My 2ct.

Marcus


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

English Dictionaries update cycle

[Marco A.G.Pinto]

Hello!

The other day someone complained in the extension page that he had
updates all the time.

I replied to that person saying that I do monthly updates of the British
speller adding 600+ words.

I talked with JZA about the update frequency and I would like to know
your opinion about the update cycle (JZA suggested a 6 month cycle but I
need more opinions).

For Mozilla (Firefox + Thunderbird) the monthly is okay since it is a
small file (400-500 kB) with auto-update and only the British. Also,
every six weeks  a Mozilla upgrade comes out  and people want the
dictionary to be compatible with the latest alphas.

What is your opinion?

Thanks!

Kind regards,
   >Marco A.G.Pinto
 

-- 


signature.asc
Description: OpenPGP digital signature

Re: [PROPOSAL] Removing all tracking from AOO sites

[Kay Schenk]

On Mon, May 16, 2016 at 9:56 AM, Pedro Giffuni  wrote:

> +1
>
> I should note, I used to have access to the Google Analytics stuff
> when I was in the PMC. It was set up by Rob Weir under consensus.
>
> It is certainly a form of tracking and, at least I, have no idea what
> purpose it has served. From some time I have had no access it as it
> mysteriously disappeared from my list of GA sites.
>
> Thanks so much for bringing up this subject, I always meant to bring
> it up but forgot.
>
> Pedro.

​
Really, this is aggregated data. The setup for GA  was agreed to  in 2011
for some pages -- the download one -- and later ported to the whole web
site.

See first discussion:

​http://markmail.org/thread/7koytfcgpdeiqzn6

( a number of discussions on other threads followed)

Some of "us" have access to the AOO analytics data. As I recall, this was
via request to Rob Weir. Right now, I can not find who the individuals are.
I am one of them. It's likely that Rob is still the only one with admin
rights to this, and this should be changed.

I have not logged into GA in quite some time. If it IS possible to provide
public access to some of this data, I would much prefer that then removing
GA entirely.  I will investigate this and get back to the list later in the
week.


-- 
--
MzK

"Time spent with cats is never wasted."
-- Sigmund Freud

RE: [PROPOSAL] Removing all tracking from AOO sites

[Pedro Giffuni]

+1

I should note, I used to have access to the Google Analytics stuff
when I was in the PMC. It was set up by Rob Weir under consensus.

It is certainly a form of tracking and, at least I, have no idea what 
purpose it has served. From some time I have had no access it as it

mysteriously disappeared from my list of GA sites.

Thanks so much for bringing up this subject, I always meant to bring
it up but forgot.

Pedro.

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

RE: [PROPOSAL] Removing all tracking from AOO sites

[Dennis E. Hamilton]

A question, below.

> -Original Message-
> From: Marcus [mailto:marcus.m...@wtnet.de]
> Sent: Monday, May 16, 2016 00:08
> To: dev@openoffice.apache.org
> Subject: Re: [PROPOSAL] Removing all tracking from AOO sites
> 
> Am 05/16/2016 12:02 AM, schrieb Andrea Pescetti:
> > Dennis E. Hamilton wrote:
> >> PROPOSAL
> >> Remove all tracking elements from AOO-authored and published web
> pages.
> >
> > As it is, I don't like this proposal. I mean, removing Google
> Analytics
> > for example is quite a strong move since we don't have anything
> equally
> > informative in place. Sure, we could use other systems (and in
> principle
> > I would support this option) or collect nothing, but the "We are not
> > using it" argument is not really strong for analytics where the
> > historical values are important; so what we collect today may be
> useful
> > in future, even if we don't use it today.
[orcmid] 

Where is the information by which anyone here can make use of current Google 
analytics to perform an investigation into historical values?

Who has the information necessary to do that and how can anyone use it?

I may simply be ignorant of an established arrangement.  Where can information 
about it be found?

> 
> simply +1
> 
> >> BACKGROUND
> >> https://bz.apache.org/ooo/show_bug.cgi?id=126959 browsers now report
> >> that there is insecure content being provided on some of the pages
> >
> > This is a much narrower case and here I agree that we should
> > investigate/remove. But so long as something works well I don't see a
> > reason to remove it; I don't like it very much, but others have used
> it
> > in the past and may need to use it in future.
> 
> The proposal from Dennis and the reason for it is not connected - at
> least this is my opinion.
> 
> The reason for browser error messages that HTTP and HTTPS content is
> mixed-up is in no relation with collecting some data about website
> visitors.
> 
> I also think we should keep the analytics stuff like it is and
> concentrate on the problems that just results in a wrong usage of URLs
> in the HTML code on some webpages.
> 
> 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

Concurrency in OO code

[Rory O'Farrell]

I'm not asking for specific details at this stage, as I'm making only a general 
enquiry,  

Can anyone familiar with OO code inform me if there is any concurrency of 
execution of code sections within the current code?  If so, is such concurrency 
under coder control or forced on OO by the compile libraries?  Might there be 
switches whereby this could be disabled?

My reason for asking is that, if so, it might prove helpful to disable such 
concurrency for a customised (one off) debugging version of OO to help isolate 
one of the regular problems 

-- 
Rory O'Farrell 

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

Re: [PROPOSAL] Removing all tracking from AOO sites

[Max Merbald]

Hi there,

there are too many trackers on too many web pages already. I think we 
should make a positive example and have all trackers removed. It appears 
that it's not the Apache OpenOffice project who have an advantage from 
the tracking junk but only others. Remove them!



Max



Am 15.05.2016 um 23:01 schrieb Dennis E. Hamilton:

There are web pages delivered by Apache OpenOffice properties that, when 
browsed, signal their access to tracking sites that collect statistics about 
site usage.  This is done by fetching scripts and/or images from the sites that 
provide the statistics.

This is a form of tracking that is carried out by services not under control of 
the Apache OpenOffice project.

The effort to track usage also causes "insecure content" warnings in browsers, 
depending on the security options the user has set for their browser and the web address 
being used.

PROPOSAL

Remove all tracking elements from AOO-authored and published web pages.

This eliminates an under-used arrangement and also removes the uncertainty that 
is raised when visitors are warned that there is insecure content associated 
with the web page they are viewing.

This proposal is offered for lazy consensus no earlier than 2016-05-23T23:00Z.

BACKGROUND

We have been enabling https: access to Apache OpenOffice web properties.  This 
is in line with the desire to use secure connections.  By secure is meant that 
the traffic on the connection itself is encrypted and there is protection 
against man-in-the-middle and spoof sites that high-jack the traffic in some 
manner.  This is not complete privacy.

As pointed out in a couple of Bugzilla issues, such as 
, browsers now report that 
there is insecure content being provided on some of the pages that are visited.  This 
is because browsing the page fetches some content without using https.  These sites 
are under Apache OpenOffice control and we cannot change to using a secure 
connection.  Even if we did, it would not change the tracking that is achieved.  It 
would simply not call attention to it.

These images, and also some scripts, are used for gathering access and usage 
statistics from various services.

Since we are not routinely making the available statistics public, this is one 
of those we-collect-it-because-we-can and not because-we-need-it.  There is, as 
far as can be determined on these lists, no active use of the information that 
is collected.

  -- Dennis E. Hamilton
 orc...@apache.org
 dennis.hamil...@acm.org+1-206-779-9430 
https://bz.apache.org/ooo/show_bug.cgi?id=126959
 https://keybase.io/orcmid  PGP F96E 89FF D456 628A
 X.509 certs used and requested for signed e-mail



-
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] Removing all tracking from AOO sites

[Roberto Galoppini]

Sent from my iPhone

> On 16 mag 2016, at 09:07, Marcus  wrote:
> 
> Am 05/16/2016 12:02 AM, schrieb Andrea Pescetti:
>> Dennis E. Hamilton wrote:
>>> PROPOSAL
>>> Remove all tracking elements from AOO-authored and published web pages.
>> 
>> As it is, I don't like this proposal. I mean, removing Google Analytics
>> for example is quite a strong move since we don't have anything equally
>> informative in place. Sure, we could use other systems (and in principle
>> I would support this option) or collect nothing, but the "We are not
>> using it" argument is not really strong for analytics where the
>> historical values are important; so what we collect today may be useful
>> in future, even if we don't use it today.
> 
> simply +1

+1

> 
>>> BACKGROUND
>>> https://bz.apache.org/ooo/show_bug.cgi?id=126959 browsers now report
>>> that there is insecure content being provided on some of the pages
>> 
>> This is a much narrower case and here I agree that we should
>> investigate/remove. But so long as something works well I don't see a
>> reason to remove it; I don't like it very much, but others have used it
>> in the past and may need to use it in future.
> 
> The proposal from Dennis and the reason for it is not connected - at least 
> this is my opinion.
> 
> The reason for browser error messages that HTTP and HTTPS content is mixed-up 
> is in no relation with collecting some data about website visitors.
> 
> I also think we should keep the analytics stuff like it is and concentrate on 
> the problems that just results in a wrong usage of URLs in the HTML code on 
> some webpages.

Fully agree. Only issue maybe to comunicate clearly and properly we do use 
cookies for such a purpose, as recommended by few national regulations.

Roberto

> 
> 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] Removing all tracking from AOO sites

[Marcus]

Am 05/16/2016 12:02 AM, schrieb Andrea Pescetti:

Dennis E. Hamilton wrote:

PROPOSAL
Remove all tracking elements from AOO-authored and published web pages.


As it is, I don't like this proposal. I mean, removing Google Analytics
for example is quite a strong move since we don't have anything equally
informative in place. Sure, we could use other systems (and in principle
I would support this option) or collect nothing, but the "We are not
using it" argument is not really strong for analytics where the
historical values are important; so what we collect today may be useful
in future, even if we don't use it today.


simply +1


BACKGROUND
https://bz.apache.org/ooo/show_bug.cgi?id=126959 browsers now report
that there is insecure content being provided on some of the pages


This is a much narrower case and here I agree that we should
investigate/remove. But so long as something works well I don't see a
reason to remove it; I don't like it very much, but others have used it
in the past and may need to use it in future.


The proposal from Dennis and the reason for it is not connected - at 
least this is my opinion.


The reason for browser error messages that HTTP and HTTPS content is 
mixed-up is in no relation with collecting some data about website visitors.


I also think we should keep the analytics stuff like it is and 
concentrate on the problems that just results in a wrong usage of URLs 
in the HTML code on some webpages.


Marcus


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