Re: [VOTE] Merge dev lists

[Torsten Curdt]

-1

As explained before.

> Only subscribed to 1 project. If consolidation on dev happens then
> there will be no good way to filter out emails I don't care about.

Same here.

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

Re: [PROPOSAL] One development list

[Torsten Curdt]

> Care to elaborate a bit?

I'd argue that for the people who care it's no big deal to subscribe
to the various lists. So tuning in is no problem, tuning out once
consolidated indeed is. It's an all-or-nothing. How is oversight
better when everyone (or at least all PMC members) are subscribed to
all dev list as opposed to just the one?

While the one dev list might work OK for Commons I have the feeling
that these projects are just too different. In Commons one could
imagine to step up for another component. I don't see that with the
rest that is left at jakarta.

If there are many cross-posting (that would be a sign for
cross-concerns) I haven't noticed any.

At least I don't see the benefits of a consolidation. I would rather
ask all PMC members to subscribe to all dev list than merging the
lists. Not to talk about the mailing list archive confusions and
hassle this might create.

but well - that's just my opinion and the -1 should not be
blocking. Just saying that I don't like the idea.

cheers
--
Torsten

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

Re: [PROPOSAL] One development list

[Torsten Curdt]

>>  >>We currently have 8 active development lists at Jakarta, each devoted
>>  >>to a subproject. I've been subscribed to all for a while and based on
>>  >>my observations and the overall benefits of doing so, I think its time
>>  >>to consolidate them into a single development list at Jakarta.
>>  > ...
>>  >>Thoughts?

Not at fan.

-1

cheers
--
Torsten

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

Re: Commons is TLP

[Torsten Curdt]

On 21.06.2007, at 00:57, Martin van den Bemt wrote:


Hi everyone,

The new Commons TLP was established today, with Torsten Curdt as  
Vice President.


...so where do we start with the TLP move is the question :)

cheers
--
Torsten



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [PROPOSAL] The future of Jakarta

[Torsten Curdt]

so this thread died again without a conclusion or resulution.

My take with as few words as possible:

* push for active project to go TLP
* jakarta.apache.org - the portal to all java projects at apache.  
Just a shell - but let's keep the brand. Not necessarily a PMC  
required. (Although a non-code project trying to improve  
collaboration between java projects would be an idea to discuss)
* ${commons}.apache.org - as people have concerns about the name (as  
there is more than java) let's find a new one.

  - commonsj
  - jcommons
  - ...
* dormant.apache.org - maybe a place where we put not just old stuff  
from jakarta


cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [VOTE] Commons moving to TLP

[Torsten Curdt]

+1

On 08.05.2007, at 19:20, Henri Yandell wrote:


Sadly a bit too late to make the next board meeting I suspect.

However, here's a vote for Commons to officially request that it  
move to TLP.


   http://wiki.apache.org/jakarta-commons/TLPResolution

Please add your name if you're a Commons developer and haven't added
your name yet.

[ ] +1 I support the proposal
[ ] +0 I don't care
[ ] -1  I'm opposed to the proposal because...

Voting will close in one week.

Hen

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: bugzilla administration

[Torsten Curdt]

On 10.03.2007, at 16:13, Torsten Curdt wrote:


http://issues.apache.org/bugzilla/show_bug.cgi?id=40577


ups ...wrong bug :) but anyway!

How can I add new version to bugzilla? Who has admin rights on  
bugzilla?


...thanks to Martin Cooper all done now :)

cheers
--
Torsten




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

bugzilla administration

[Torsten Curdt]

http://issues.apache.org/bugzilla/show_bug.cgi?id=40577

How can I add new version to bugzilla? Who has admin rights on bugzilla?

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Three votes of PMC members required

[Torsten Curdt]

+1
--
Torsten

On 10/16/06, Davanum Srinivas <[EMAIL PROTECTED]> wrote:

+1 from me.

-- dims

On 10/15/06, Jochen Wiedmann <[EMAIL PROTECTED]> wrote:
> Hi,
>
> AFAIK the policy is still that three votes of PMC members are
> required. In other words, may I point you to
>
> http://marc.theaimsgroup.com/?t=11606631873
>
> and ask kindly for positive votes? (Unless you have reason for
> negative votes, of course.)
>
> Jochen
>
> --
> My wife Mary and I have been married for forty-seven years and not
> once have we had an argument serious enough to consider divorce;
> murder, yes, but divorce, never.
> (Jack Benny)
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>


--
Davanum Srinivas : http://www.wso2.net (Oxygen for Web Service Developers)

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [VOTE] Move Velocity to TLP

[Torsten Curdt]

+1

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Re: Jakarta BOF @ apachecon USA

[Torsten Curdt]

Would it be possible to have it on Monday or Tuesday or is there another
firm date for BOFs ?


I reckon let's better not make the official BOF on Mon/Tue as probably
most people (users) arrive on Wed. We should give them a chance to
join us as well ...but we can still go for a beer/food on Mon/Tue ;-)

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Re: Jakarta BOF @ apachecon USA

[Torsten Curdt]

Cool :) Let's do this at dinner, since there are only 2 of us ;)


Hehehe :-)

...from experience I am sure a few more people will show up.
So I reckon we should schedule it as an usual BOF ...but doesn't
mean that's gonna exclude food or beers. Let's leave it up to the
"community dynamics" ;-)

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Jakarta BOF @ apachecon USA

[Torsten Curdt]

+1 cool bananas ..will be there

cheers
--
Torsten

On 8/22/06, Martin van den Bemt <[EMAIL PROTECTED]> wrote:

Hi fellow Jakartians,

I welcome everyone going to Apachecon to the Jakarta BOF (which I just 
registered in the Wiki,
assuming that is sufficient to have the BOF take place).
I just put it at a certain date and time, though details are not known yet.

The main reason for the BOF is to discuss the future of Jakarta, with me being 
moderator and giving
the initial data of what the current thoughts are on this, based on discussions 
that happened on the
  mailinglist.

Mvgr,
Martin

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Re: Re: Re: Opening up the PMC

[Torsten Curdt]

> Why that combination? ...feels quite unnatural to me - I am sure so it will
> for the users.

It's our list of small largely inactive subprojects. We know that none of
those are going to go TLP, and their dev lists are quiet by an order of
magnitude compared to an inactive TLP potential project like Slide.


True ...but that grouping is not obvious to the users. Besides conditions
*could* change.


> Combining mailinglist would definitely get a -1 from me. I don't want the
> few BCEL subscribers left to unsubscribe because they get annoyed
> because they are receiving mails they are not interested in.
>
> We might get away with that at commons - but in general I think it's
> quite a bad idea.

We've a bunch of relatively small codebases and subprojects that on their
own are a stretch to maintain oversight over. It's a match for the Commons
approach of having those communities share the same space and help each
other on general issue topics, such as releases.


Well, oversight does not really apply to the users. Besides developers
could subscribe to all sub projects list. We could even ask comitters or
at least PMC members to subscribe to all of them to achieve the same
kind of oversight. ...it would just be more convenient for the users.

IMO rather have it too fine grained and combine than too general and you
are stuck with what you don't want.

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Re: Re: Opening up the PMC

[Torsten Curdt]

The problem with moving Commons up is that when you look at where Jakarta
needs to go, and when you look at where Commons generally is now; they are
the same places - and it's hard to distinguish between the focuses.


Hm... interesting... funnily I have never seen it like that before.
Always had the
impression jakarta is for "products" while commons is for libraries. Sometimes
it's just not that clear where to draw the line though.


Jakarta needs a way to blend community oversight with small numbers of
active committers per component. That's pretty much Commons. For a start,
I would merge BCEL, BSF, ECS, JCS, ORO and Regexp dev lists into either
[EMAIL PROTECTED] or [EMAIL PROTECTED]


Why that combination? ...feels quite unnatural to me - I am sure so it will
for the users.

Combining mailinglist would definitely get a -1 from me. I don't want the
few BCEL subscribers left to unsubscribe because they get annoyed
because they are receiving mails they are not interested in.

We might get away with that at commons - but in general I think it's
quite a bad idea.


1) BCEL, BSF, ECS, JCS, ORO and Regexp dev lists merging
2) Alexandria dies
3) POI to TLP
4) Consider merging user lists from 1)
5) Flatten HttpComponents, Velocity, Taglibs.

Some mad ideas :)


Sorry that I have to agree with the last sentence in some extend ;-)

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Re: Opening up the PMC

[Torsten Curdt]

> But if you argue into that direction -no matter how often this has
> been discussed already- I would rather question the idea of an
> umbrella PMC then... (*ducks*)

Time and energy to express the ideas you have in that direction ?


Let's try :-)

IMO it's quite awkward to have oversight over a project where I don't
even know people nor the codebase. Of course most of people on the PMC
are active in a couple of the different jakarta projects ...but still
not in every project.

Of course issues can be brought up and discussed once they are
explained to the people that haven't been aware of it. But still it's
quite strange that these people have a full counting PMC vote on the
affairs of that very project. It's just makes things a bit harder.
(Maybe that's why the PMC memeber selection process is currently
needed)

On the other hand I do understand that this can be a problem. Hell, we
hardly have enough *committers* on BCEL ...how should this project's
PMC be functional?

I am not really sure how to solve this. I am just ranting. For a few
projects I think they should go toplevel. For the ones I am involved
in at least jakarta commons surely deserves it (not looking into the
naming problem for now). Having a few more toplevel projects could
help to strip down the "general" jakarta PMC a bit ...and maybe then
the committer == PMC idea is more suitable ...maybe let's get a
smaller umbrella ;-)

*shrug* ...just some RTs - but surely no new ones...

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Re: Opening up the PMC

[Torsten Curdt]

Just a few comments (as I am a bit late)

At Cocoon every committer can join the PMC by just asking for it. The
idea is that committer do care and shape the project anyway. If you
don't care enough about the project - why would you be a committer? We
are quite open and were working out most of the things on the dev list
anyway. This seems to work pretty well for the Cocoon community.

...well, now with Jakarta this is a little different situation - as
stated before it's an umbrella project. Which actually makes the
position of the PMC a bit more complicated ...why should a committer
to commons have an impact on the direction of e.g. hivemind. So maybe
we need to be a bit more selective for the PMC members (as we are).
Maybe.

But if you argue into that direction -no matter how often this has
been discussed already- I would rather question the idea of an
umbrella PMC then... (*ducks*)

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Re: Re: State of Slide project

[Torsten Curdt]

More worrying, it sounds like while ecs/oro/regexp have someone holding a
janitor role and looking after things, Slide might not.


However I do think that there are quite some projects relying on it.
Maybe worth checking who is using slide (gump?)

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[ApacheCon] jakarta (commons) beer chitchat

[Torsten Curdt]

Guys,

There are rumours people are interested in a beer chitchat BOF kind of thing :)
Say tonight 8pm at the bar?

cheers
--
Torsten

PS: Please respond on [EMAIL PROTECTED]

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Need feedback on BCEL 5.2 RC2

[Torsten Curdt]

...if you are using BCEL it would be great if you could give the
latest RC a test run.
We would like to release BCEL 5.2 ASAP but we are still lacking some feedback.

http://people.apache.org/~tcurdt/bcel/rc2/
http://vafer.org/blog/20060429164701

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [VOTE] Jakarta Sandbox

[Torsten Curdt]

>  However Jakarta-sandbox is
> SCOPELESS.  Go have a scopeless sandbox on sourceforge IMO.  If you want
> to start a whole NEW project then do that in the incubator IMO.

Why on sourceforge - why not on our infrastructure?
What the difference for you?

You want every tiny (commons) library go through the incubator?
...or do you just don't want full projects sneak in through that sandbox?

So far I don't understand why you are seeing this so problematic.

cheers
--
Torsten

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Managing communities and emails (was Re: [PROPOSAL] Jakarta Language Components)

[Torsten Curdt]

On 15.03.2006, at 10:10, Thomas Dudziak wrote:


On 3/14/06, Torsten Curdt <[EMAIL PROTECTED]> wrote:


...regarding the forums - na. What does that help?


Judging from myself, users don't like to have to subscribe to mailing
lists, especially when they don't need the list on a daily basis. E.g.
I would hate to get every question and answer in the Spring forums as
mail. I'd much rather check the forums every once in a while.


Use Gmane ;)

I am too lazy to always go and check forums.
With the tagging approach you would only have to
really subscribe once. ...and as I said - having feeds would be awesome.



I hate the fact I always have to subscribe to forums and
never liked the interfaces.


You misunderstood: Patrick told me that the 'special' feature of the
system that they use at OpenQA is that it can be used both as a
mailing list (e.g. for the developers) and a forum (for the users if
you want). Therefore, you'd register at a mailing list just as before
and have nothing to do at all with the forum-view, but users could
instead use the forum (and thus won't get all the - for them - noise).
And the system automatically maps between the mailing list and the
forum.


I think what you are basically saying is you want a poll
not a push service. IMO feeds are much better for that
than a forum. But of course you could combine all these
things.

cheers
--
Torsten



smime.p7s
Description: S/MIME cryptographic signature

Re: Managing communities and emails (was Re: [PROPOSAL] Jakarta Language Components)

[Torsten Curdt]

i like the idea of tagging emails better: a single list with cool  
server
side filtering and metrics. we don't have the technology for this  
yet so

i'm willing to see the mailing lists split so long as people would be
willing to consider coming back if it every arrives...



I was just considering proposing exactly this!


A joke turns into something serious ...but I am all with you guys.
As I said: the more I think about it - the more I like the idea!

The another really cool feature with this tagging approach
would be that cross-posting would no longer need to be banned
- it would just not exist.

A set of checkboxes would allow a user to "subscribe" to various  
lists,
or to virtual groupings such as "jakarta commons" which would  
implicitly

subscribe to the list for every project that is tagged as being a
jakarta-commons project. Of course this implies fine-grained email  
lists

(ie one for each project); the problems of partitioning the subscriber
base too much is avoided by the existence of the groupings.


Yepp!


This system would allow overlapping groups to occur; for example
commons-digester can be filed under both "commons" and "xml" virtual
groups; someone subscribing to *either* group would receive
digester-related emails. It also allows projects to move from one PMC
to another without destroying the existing community (which *is*  
the set

of people receiving emails).


Great possibilites

Groups also allow new projects to be created and added to the  
group; all

people subscribed to the group would then automatically get emails
related to that new project.

Any list which has less than 3 subscribers would automatically forward
its emails to the PMC list (or similar) for purposes of oversight.


interesting


Any person subscribed to 3 or more projects associated with "commons"
would automatically be subscribed to the whole commons group (or maybe
just sent a weekly nag email recommending they do so). That hopefully
allows casual commons developers to get just postings for one or two
projects, without destroying the useful commons-wide community that
exists now.

Having a single point for managing subscriptions would also help  
greatly

with something that regularly frustrates me: suspending subscription
when I'm away on holiday. Currently, I need to unsubscribe to
half-a-dozen lists then resubscribe on return.


Yepp!

Another thing that would be awesome is to integrate it with a
powerful archive system ...and then provide feeds for all the
different tag combinations - and even individual threads!
Man - that would be awesome!


This sort of functionality probably already exists in one of the
open-source mailing list management packages; it isn't anything  
radical

as far as I can see.


Now we only have to find such a project and then convince infrastructure

...regarding the forums - na. What does that help?
I hate the fact I always have to subscribe to forums and
never liked the interfaces.

cheers
--
Torsten

smime.p7s
Description: S/MIME cryptographic signature

Re: Representing project inactivity on the site

[Torsten Curdt]

BCEL - In need of a bugfix release, design-wise ASM is preferrable.


Almost there :) ...still a few bugs to close

cheers
--
Torsten

smime.p7s
Description: S/MIME cryptographic signature

Re: [PROPOSAL] Two community proposals

[Torsten Curdt]

On 05.03.2006, at 20:21, Henri Yandell wrote:



I started to write a long email on the problems in Jakarta, on  
umbrellas, on the lack of a Jakarta community and existence only of  
subcommunities and on how it should be "there is no Jakarta Xxxx,  
you are members of Jakarta - not a subproject"; but you've heard it  
all before.


So, proposal:

-
Given that we are one project and that we should be acting as one  
community - I propose that we:


1) Remove SVN restrictions, all Jakarta committers can commit  
anywhere in Jakarta, with the exception of the Commons-Sandbox as  
it allows Apache committers in general to commit.


So a Commons committer can commit to e.g. BCEL and Hivemind without  
knowing the code bases? H

That doesn't sound right to me :-/

TBH Jakarta feels less as one community ...but more like an umbrella.  
Do you want to change that?


cheers
--
Torsten

smime.p7s
Description: S/MIME cryptographic signature

Re: [site] Kill the vendor support page

[Torsten Curdt]

+1 let's kill it ...wiki would be alright though

cheers
--
Torsten


PGP.sig
Description: This is a digitally signed message part

Re: [VOTE] Jakarta Http Components

[Torsten Curdt]

On 21.10.2005, at 00:03, Henri Yandell wrote:



So, period of comments now over, let's go ahead and vote on the  
creation of a Jakarta Subproject named Http Components with the  
following initial charter:


http://wiki.apache.org/jakarta-httpclient/NewProjectCharter


[x] +1
[ ] -1


cheers
--
Torsten



PGP.sig
Description: This is a digitally signed message part

Re: List moderation

[Torsten Curdt]

I help manage tomcat-users and tomcat-dev, both high volume lists.  
We use the following policy for both lists:


- anyone can subscribe
- subscriptions are not moderated
- subscribers can post
- posts from non-subscribers are rejected

This works pretty well, the only things tending to go wrong are:
- genuine users not subscribing and not reading/understanding the  
reject message - less than 10 a week and easy to deal with
- auto-responders managing to subscribe themselves to the list -  
again less than 10 a week and easy to deal with


The only time we have had real problems in the last 2 years has  
been when an auto-responder managed to get subscribed to the list  
and was replying to its own replies. This killed the list until my  
unsubscribe and deny-subscribe messages got through.


Whilst no system is perfect this works well for us and on most days  
I spend between 0 and 5 minutes dealing with list issues.




So this can be decided on a per-project basis already?

cheers
--
Torsten



PGP.sig
Description: This is a digitally signed message part

Re: List moderation

[Torsten Curdt]

given that there's a rising tide of demand to make all lists  
subscriber

only, i think that this would be better than simply sending all
unsubscribed posts to /dev/null.


+1


i have personal itches about some of the required technologies and i
know some of the spam assassin guys are keen on some of the others.
maybe i'll try to ambush some people on irc...


that would be great ...any luck yet? :)

cheers
--
Torsten




PGP.sig
Description: This is a digitally signed message part

Re: last modified

[Torsten Curdt]

This might have to do something with the timestamp resolution of  
the filesystem. Not every filesystem have a resolution of  
milliseconds but some higher factor.


Yeah, Brett already pointed offlist me on that.

It gets ever worse e.g 10ms  for  FAT on create but 2sec for write  
(http://msdn.microsoft.com/library/default.asp?url=/library/en-us/ 
fileio/fs/setting_and_getting_the_timestamp_of_a_file.asp)

To be cross platform dont expect more than 2sec.

Bad news, isnt it?


Ouch! Hmm... need to think about
what implications this might have
for the jci monitor...


Mario

PS: After some googling I found it might be worth to collect those  
precision and create a web page with them.


Sounds like a good idea

Thanks
--
Torsten


PGP.sig
Description: This is a digitally signed message part

last modified

[Torsten Curdt]

Guys,

I hope this is not too OT for this list but I
am running out of ideas. What could be possibly
be wrong with the following piece of code:

import java.io.File;

public class test {
  public static void main(String[] args) throws Exception {
 File dir = new File("dir");
 dir.mkdir();
 long m = dir.lastModified();
 File subdir = new File(dir, "subdir");
 subdir.mkdir();
 while(dir.lastModified() == m) {
   Thread.sleep(1000);
   System.out.print('.');
 }
 subdir.delete();
 dir.delete();
  }
}

To me it seems like the last modified
is not being updated (even if I re-create
the File object) unless the directory
gets touched *outside* the jvm.

Anyone aware of this problem?

cheers
--
Torsten


PGP.sig
Description: This is a digitally signed message part

Re: List moderation

[Torsten Curdt]

we're moving towards insisting on pgp keys so we could check and
moderate through all signed apache.org email without danger.



You mean forcing all subscribers to use gpg
Although I love the idea of forcing people to
sign their mails

http://vafer.org/blog/tcurdt/archives/93.html
http://vafer.org/blog/tcurdt/archives/000174.html

I fear there is no way we can get away with that.




This would leave open for moderation only a few broad lists (for
instance [EMAIL PROTECTED], which I have the "pleasure" to
moderate :-( ), killing a lot of wasted hours.

Also, FYI, joes2 is doing some info gathering on moderation  
request to

aid us in filtering incoming moderation spam. See
irc://irc.freenode.net/#asfinfra for more info.


Language is english for all lists. Filtering out
mails with all kinds of weird characters would be
a first step. Still wondering why we don't have
that already.


i've been wondering for a while whether it'd be better to move towards
bots and some sort of web app for smarter collaborative moderation.  
(one

moderators marks a message as spam and all identical ones are
automatically rejected from all pending lists.)


That would be much better ...but it's also quite
an effort to set this up. I just don't know whether
it's worth it for those few cross postings

cheers
--
Torsten




PGP.sig
Description: This is a digitally signed message part

Re: List moderation

[Torsten Curdt]

Hi,

I've been a moderator for the bcel-dev@jakarta.apache.org and
bcel-user@jakarta.apache.org for a few months now. In that time there
have been 2 valid emails from people who weren't subscribed, and a few
thousand spam mails.

I'm rather tired of being a human spam filter for the mailing list. As
the ratio of bad to good emails is so high, I suggest that all mails
from unsubscribed users simply be discarded making moderation
unnecessary.

What's the general opinion about this?


The question is why do we use moderation at all?

I can see only the reason of cross-posting
which is discouraged anyway ...and still could
be done by people who are subscribed to both lists.

...so I am with you on this

cheers
--
Torsten


PGP.sig
Description: This is a digitally signed message part

Re: Jakarta - Cleaning house

[Torsten Curdt]

Finish SVN migrations (Turbine-3, POI, JMeter, Cactus)



Turbine-3 should go from CVS to the graveyard. No detour through SVN
please.


AFAIU (not totally sure) all projects need to migrated to CVS
by the end of the year because CVS is meant to go away. If we
want it to be still available we need to migrate it, too.

cheers
--
Torsten



PGP.sig
Description: This is a digitally signed message part

Re: Tracking commons component liveliness

[Torsten Curdt]

I have a big problem with putting people's names beside projects  
and components on a public web page. Besides being yet another  
thing that needs to be kept up to date, it will only encourage  
people to contact the developers directly, instead of using the  
mailing lists. From my own perspective, this is a huge problem  
already, and I'd be -1 to anything that's going to further  
exacerbate it.


...in general I find the idea not too bad.
But Martin is right - putting names to projects
is bad in general.

What about a slight modification - what if
every jakarta sandbox committer is required
to report back to the *PMC* what sandbox he
is working on? (lazy feedback of course.
no feedback not working on anything)

...the PMC will sort out the statistic
internally. Since project oversight is
one of the duties of the PMC ...I think
it would make sense.

WDYT?

cheers
--
Torsten


PGP.sig
Description: This is a digitally signed message part

Re: Name for commons-like area for web

[Torsten Curdt]

>>Probably needs to be discussed.
> 
> 
> please, please no sub-sub-projects!

...just a theoretical option.
But I guess you are right ;)

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: Name for commons-like area for web

[Torsten Curdt]

> Pardon my ignorance, but if Web were a subproject under Jakarta
> Commons, could Web itself have subprojects?

AFAIK there is no project that has such subproject.
...but that does not mean it's completely impossible.
Probably needs to be discussed.

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: Name for commons-like area for web

[Torsten Curdt]

> What's wrong with Webapp Commons?

Commons = Jakarta Commons

...for a whole bunch of people.

As long as it is not under the
Jakarta Commons umbrella like
"Jakarta Commons Web" I would
be clearly against using the
word "commons".

Actually if it is meant to be
java only it could even fit well
into the jakarta commons project.

...IMO

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: Name for commons-like area for web

[Torsten Curdt]

> The most obvious would be CommonsWeb or WebCommons, as the general user
> community could link the concept to commons easily enough. However,
> there is a danger that it could be confusing precisely because of that.

As long as it is not under the
umbrella of the Jakarta Commons
project I would avaoid the term
"Commons".

I think I like WebLibs the best.

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: Jakarta Apache Tomcat as a TLP ?

[Torsten Curdt]

> Please, lets calm the things down.
> Henri will write an email to SD magazine, and the earth
> will still spin tomorrow.

...that's an excellent ending for this discussion :)

Thanks
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: [draft-2] SD Magazine: request for change

[Torsten Curdt]

> Hi Kate,
> 
> I'd like to thank SD/Jolt on behalf of the Jakarta community for the
> JOLT "Productivity Winner" award for Tomcat 5.0. Media recognition of
> the work the Tomcat community puts in is always very welcome.
> 
> I'm also writing to let you know about a [serious] error on your JOLT
> product excellence awards press release because I am concerned it might
> be reproduced in your forthcoming June 2005 issue:

I would drop "serious"

> http://www.sdmagazine.com/pressroom/jolt_winners_2005.pdf
> 
> The release [incorrectly] attributes Apache's Tomcat 5.0 product to "The

Also the "incorrectly"

> Apache Jakarta Project and leading Tomcat contributor JBoss".
> 
> There is one big problem with this attribution for us, Apache does not
> have a concept of leading contributors. It is completely out of sync
> with the very philosophies that lie at the heart of the Apache Software
> Foundation (ASF), as there are 70 committers to the Tomcat codebase,
> many of whom are employed by other companies or contribute individually.
> 
> We would like to request that this be changed to:

"request" sounds a bit harsh ...what about "see"

> "Tomcat 5.0 (The Apache Software Foundation)"
> 
> in both the press release (pdf url above) and the forthcoming June 2005
> issue.
> 
> Officially the name of the product is "Apache Tomcat 5.0" and not just
> "Tomcat 5.0", but I will leave it to your discretion as to whether you'd
> like to prefix Tomcat with Apache or not, as the subsequent mention of
> the ASF is fine.

...always play nice on the first contact.

My 2 cents

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: [draft] SD Magazine: request for change

[Torsten Curdt]

>>As a matter of fact we have quite some committers in our 
>>community that are sponsored by the companies they are 
>>working for. Who is able to define whether who is the 
>>"leading" or "main" contributor? I would not want to risk 
>>picking the wrong one and pissing off other contributors. So 
>>either name them all or drop this classification. Terms like 
>>"main" or "leading"
>>are a problem.
>>
> 
> 
> Again, anyone outside of the ASF is allowed to make any observations
> they feel are warranted - regardless of ASF corporate policy.  It might
> fly in the face of what the ASF wants, but it doesn't make the statement
> valid or invalid.

Sure

>  Whether the PRC agrees with "a leading", "the
> leading", "a main", "the main", makes little difference.  SD can say
> what they say; the real issue is the relationship between the ASF and
> JBoss.

Well, I don't know much about the ASF and JBoss
story. (And I think I also don't want to know ;)

But for me it's simple and not about JBoss.
If it was some ACME company - it would still
not be ok. It's just something you don't say
or do.

And what we should only point the magazine
to is the fact that we don't like it.

...what they do is up to them.

But just keeping our mouths shut and only rail
about the fact probably does not help the
JBoss relationship at all.

> People at the ASF are so "worked up" over JBoss in particular (and vice
> versa).  Every time someone outside of both ASF and JBoss just makes an
> observation that JBoss is a leading contributor, we set up the PR
> machine and launch off some emails, maybe a few people will go over to
> the TSS forums and blow up at Jboss people, and I could expect a few
> nasty blog rants from both sides.

Well, if you write something like that you
have to cope with criticism. But I would aim
at the magazine - not JBoss.

...maybe that's the difference.

> I just don't think it's constructive,
> that's all.  Believe me now or hear me later - there are good things to
> be gained from calming this situation down a little.

TBH ...I think writing a polite letter
is constructive. And as a homework we
should try to realize it's not JBoss
who wrote that line.

That should calm everyone down.
...at least it should

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: [draft] SD Magazine: request for change

[Torsten Curdt]

> I think continuing with the current attitude would only lead my company
> to reevaluate its involvement in ASF projects, and I could not really
> blame them if they did. Of course, this may be what some people here
> seek (hopefully, it is not and it's just my paranoia at work).

I am sure the community is thankful for the individual
contributions and therefor is also thankful for the
companies letting their employees spend time for working
on Apache projects. But...

Usually companies don't let their employees spend
so much time (and therefor money) just for the good
of mankind. Usually they have an interest in fixing
certain things because they want to benefit from a
product being developed and maintained by a large
number of developers they *don't* have to pay.

As a matter of fact we have quite some committers
in our community that are sponsored by the companies
they are working for. Who is able to define whether
who is the "leading" or "main" contributor? I would
not want to risk picking the wrong one and pissing
off other contributors. So either name them all or
drop this classification. Terms like "main" or "leading"
are a problem.

Naming them all for such an award is very inappropriate
IMO. No problem listing all the companies that contribute
to a certain project somewhere. A contributors file ...or
even on the website. But refering to JBoss as the main
contributor is not in the spirit of the community IMO.
So that's why we should ask for the change.

Assuming JBoss will respect the community this
should be no problem at all ...and be no reason
to "reevaluate the involvement".

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: [draft] SD Magazine: request for change

[Torsten Curdt]

Sounds good +1

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: Leading Contributor?

[Torsten Curdt]

Davanum Srinivas wrote:
> Tim,
> 
> This is a *VERY* slippery slope. All of us are committers at Apache on
> a personal basis and our employment is no bearing on our committer
> status. Remember, they (or anyone for that matter) else can fork
> Tomcat (and call it by any name they want and do whatever they want)
> if they crave publicity. IF they want to do joint publicity, they can
> of course contact the prc and make as much press releases to their
> heart's content. This is not the right way to do it.

Totally agree! ...there are other people working full-time
on Apache projects. How do we decide which companies are
worth being mentioned? ...let's better not go down that road

cheers
--
Torsten


signature.asc
Description: OpenPGP digital signature

Re: Dormancy worries

[Torsten Curdt]

Noel J. Bergman wrote:
I share your concerns.
Same here! ...there was a poll about how many
active committers are around on bcel-dev.
IIRC *one* responded. Other than that there
are just users around waiting for patches being
applied or just lurking.
http://marc.theaimsgroup.com/?l=bcel-dev&m=110294999712414&w=2
The general question is what to do with such
projects, and I think we should be open to creative ideas.
The question is whether those projects
maybe missed to nominate some more committers?
...maybe it would be worth looking into the
patch queue? Not applying patches is also
one way to make people go away.
There has been
some suggestion that they be cleaned up by migrating them to some other
domain to be mothballed.  The code, web site, and mailing list archives
would be preserved, and could be restored at such time in the future when a
community might arise.
Hmm... the question is whether we really *can* mothballing them.
IMHO too many projects rely on them. It's an important piece of
technology. At least bugfixes need to be applied. FWICS migration
costs (e.g. to asm) are fairly high.
There *are* people that have patches in the queue.
BCEL and BSF both seem to be in danger of having no actual Apache
>>community
I only know about BCEL ...but that's for sure a danger I see too!
cheers
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: new sandbox projects

[Torsten Curdt]

Any voting needed for the sandbox components?

To get in to the Sanbox, no. To get out again, to Commons Proper, yes. ;-)
ok :-)
Otherwise it would be great if I could get
access to the sandbox so I can move the stuff
over.

I'll make a request to have you added to the jakarta group.
great! thanks, Martin
cheers
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: new sandbox projects

[Torsten Curdt]

Any Apache committer can have sandbox karma just for the asking.

The only complication is that the committer will need to get the jakarta 
unix group, so it'll take us a little bit longer to add karma.
Any voting needed for the sandbox components?
Otherwise it would be great if I could get
access to the sandbox so I can move the stuff
over.
cheers
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

new sandbox projects

[Torsten Curdt]

Hey, folks!
Over at cocoon we have some code that might be worth
sharing on jakarta commons. So I was wondering
if the sandbox is open to any committer or only to
jakarta committers? (which I am not) I heard
different stories...
I factored out our javaflow (java continuations)
implementation and a java compiler interface (jci)
featuring a compiling classloader.
Any opinions on hosting that at jakarta commons?
Or should we keep that stuff under our umbrella?
cheers
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: support for native java continuations

[Torsten Curdt]

I am probably ignorant, but what are continuations?
Hope these links help to get the picture...
http://cocoon.apache.org/2.1/userdocs/flow/continuations.html
http://cocoon.apache.org/2.1/userdocs/flow/
http://vafer.org/blog/tcurdt/archives/48.html
http://vafer.org/blog/tcurdt/archives/50.html
http://www.ai.mit.edu/~gregs/ll1-discuss-archive-html/msg00501.html
http://rifers.org/downloads/rife/rife_fosdem.pdf
cheers
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: support for native java continuations

[Torsten Curdt]

Just curious:
Why is this not developed under the Apache Incubator?
Is there a reason for the codehaus move?
We already have a first implementation
based on java byte code transformation
(via BCEL) in Cocoon. But this initiative
is aiming for a JSR that hopefully give us
native support - one day...
Actually there is no particular reason
for codehaus. They were just very quick
setting everything up :) Besides it's
only for collaboration on the JSR. So
there should be no code involved.
cheers
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

support for native java continuations

[Torsten Curdt]

We'd like to invite everyone who is interested
to join our initiative on codehaus.org. We are
aiming to write and submit a JSR for native java
continuations support inside the JVM.
We are currently looking for people that have
any kind of expertise in the continuations field
or just like to support the effort by any means.
Please subscribe by sending a mail to
 [EMAIL PROTECTED]
all post will have to go to
 [EMAIL PROTECTED]
Please spread the word,
Thanks
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

jelly releases

[Torsten Curdt]

Is there a particular reason why the jelly release is
hosted on ibiblio.org? (No nagging - just curious)
--
Torsten


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: optimizeit

[Torsten Curdt]

>>Does anyone know if there is an Apache wide license for
>>optimzeit? IIRC I heard someone say something like that
>
There isn't one.. I forwarded the mail exchange I had with a Borland
manager to jakarta pmc about a year ago or something, don't know if it
got picked up.
Hm... maybe JProbe then?
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

optimizeit

[Torsten Curdt]

Does anyone know if there is an Apache wide license for
optimzeit? IIRC I heard someone say something like that
--
Torsten
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Forrest (a.k.a. xml.apache.org 2.0)

[Torsten Curdt]

> > What about NS4 and IE4 and earlier? Well ... read
> > http://www.alistapart.com/stories/netscape/ then come back to talk
> > about
> > it (maybe off these lists if this is considered OT).
>
> h, what do others think on this?

I have to admit that this article is quite true.
And we don't have a customer telling us "but it
needs to look good on old browsers, too"

http://www.alistapart.com/stories/tohell/

Maybe we could analyse the logs and see what percentages
we are talking about...
--
Torsten


--
To unsubscribe, e-mail:   
For additional commands, e-mail: