Skip to main content

GLASSFISH IS LAME

54 replies [Last post]
gniontas
Offline
Joined: 2009-07-15

Reply viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Eduardo Pelegri-Llopart

There are many ways of organizing development groups and each approach
has different tradeoffs; I have not been close to the KDE or Linux
communities, but I've seen the Apache model and we have never claimed
that GlassFish works that way.

> Sun has more power in the community than
> others, so there is a skew in the democratic system.

That is correct; GlassFish is explicitly lead by Sun [1]. We try to be
transparent and responsive and to provide value to all members in the
community, from individual contributors to partners to users, and also
to Sun itself [2]. The nice thing about middleware right now is that
people have multiple choices, so if they prefer other communities they
can go there.

[1]http://glassfish.dev.java.net/public/GovernancePolicy.html
[2]http://wiki.glassfish.java.net/Wiki.jsp?page=GlassFishPrinciples

I don't claim our arrangement is perfect. I know it can be improved in
a number of ways and we are open to modifications. Why don't we focus
on concrete problems that you have with the current system? You mention...

> ... must ask a
> possibly external decider to merge a proposed patch
> Exactly that hard is the work
> currently for external contributors.

Anybody with commit rights can contribute code directly, under
coordination of the project engineering lead. This is equally true for
Sun and non-Sun folks and the arrangement is pretty common practice in
software projects.

Sure, the engineering lead is a Sun employee but any disagreements on
why something can or cannot be done will be discussed in public. In my
experience this transparency produces sound decisions and I've been a
very strong proponent of transparency in GlassFish, the JCP and elsewhere.

I'm happy to discuss which parts of this arrangement are not working for
you, if that is the problem you are encountering.

- eduard/o

BTW, regarding your last paragraph, even if one believes that democracy
is the right government for states, it does not follow that it is the
best one for open source projects. Democracy in states has evolved over
the last couple of millenia and there have been many examples exploring
what works and what does not work. Over this time we (humans) have
explored issues like representation, balance of power, constitutions,
bills of rights, etc.

Open Source projects, on the other hand, a _very_ recent invention and
still evolving. At this point I'd even argue that it is likely that we
will end with different types of Open Source project governances that
are optimizing for different goals, but it is hard to predict given how
short lived all these projects have been.

Markus Karg wrote:
>> re: local autonomy -- sure, we have been working on that for many
>> years.
>> Sun has had strong investments in India at least since mid 90s. We
>> have
>> local managers, local directors, etc, etc.
>
> Thank you for playing the ball into my domain: Yes, SUN has local
> engineers, but the community has not. My proposal was to have the
> community elect those local deciders. That means, Sun then must ask a
> possibly external decider to merge a proposed patch. Not vice versa. See
> how hard the work then would be for Sun? Exactly that hard is the work
> currently for external contributors.
>
>> I think we understand how all this works in theory, but in the
>> practice,
>> and in a large organization, there are limits. And the larger and
> more
>> complex, and faster moving, the artifact, the harder it is to make
>> these
>> things work.
>
> Sure, but you mix up the internal structures of Sun with the public
> structures of the community. Sun has more power in the community than
> others, so there is a skew in the democratic system. Also you seem to
> ignore that there are publicly managed projects like Linux and KDE which
> work very well without having a single vendor making decisions behind
> closed doors. And I would not say Linux and KDE is plain theory. From
> you view you could also say, there is no need for public elections as Mr
> Bush already has had embassadors all over the world solving regional
> conflicts without public discussion. But sure we all are glad that the
> president gets elected and that the people can vote for or against
> particular actions. Don't we?
>
> Regards
> Markus
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

cheeser
Offline
Joined: 2004-08-02

> oooooh. I rest my case:)

I really wish you would. You promised once you were leaving.

> Now I know exactly why nothing gets done. Where are
> the Engineers ??

A *lot* gets done. If something you want done isn't, then try writing some code. That's what open source is about. You can't complain if you don't pitch in. You're just an angry little troll. I wish list management would do something about it but sadly that's not up to me.

Eduardo Pelegri-Llopart

Everywhere. US West Coast, US East Coast, Middle of US, Western Europe,
Eastern Europe, India, China, Japan...

- eduard/o

glassfish@javadesktop.org wrote:
>> Sun has had strong investments in India at least since mid 90s.
>> We have local managers, local directors, etc, etc.
>
> oooooh. I rest my case:)
> Now I know exactly why nothing gets done. Where are the Engineers ??
> [Message sent by forum member 'gniontas' (gniontas)]
>
> http://forums.java.net/jive/thread.jspa?messageID=357511
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

> re: local autonomy -- sure, we have been working on that for many
> years.
> Sun has had strong investments in India at least since mid 90s. We
> have
> local managers, local directors, etc, etc.

Thank you for playing the ball into my domain: Yes, SUN has local
engineers, but the community has not. My proposal was to have the
community elect those local deciders. That means, Sun then must ask a
possibly external decider to merge a proposed patch. Not vice versa. See
how hard the work then would be for Sun? Exactly that hard is the work
currently for external contributors.

> I think we understand how all this works in theory, but in the
> practice,
> and in a large organization, there are limits. And the larger and
more
> complex, and faster moving, the artifact, the harder it is to make
> these
> things work.

Sure, but you mix up the internal structures of Sun with the public
structures of the community. Sun has more power in the community than
others, so there is a skew in the democratic system. Also you seem to
ignore that there are publicly managed projects like Linux and KDE which
work very well without having a single vendor making decisions behind
closed doors. And I would not say Linux and KDE is plain theory. From
you view you could also say, there is no need for public elections as Mr
Bush already has had embassadors all over the world solving regional
conflicts without public discussion. But sure we all are glad that the
president gets elected and that the people can vote for or against
particular actions. Don't we?

Regards
Markus

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Martin Gainty

mg>hopefully quick comments

> Date: Sat, 25 Jul 2009 10:55:58 +0200
> From: karg@quipsy.de
> To: users@glassfish.dev.java.net
> Subject: RE: Re: GlassFish Contributions (was GLASSFISH IS LAME)
>
> > re: local autonomy -- sure, we have been working on that for many
> > years.
> > Sun has had strong investments in India at least since mid 90s. We
> > have
> > local managers, local directors, etc, etc.
mg>boasting "excluding development/support/feedback from the rest of the planet" is a plus?

> Thank you for playing the ball into my domain: Yes, SUN has local
> engineers, but the community has not. My proposal was to have the
> community elect those local deciders.
mg>and make the whole process truly democratic..good point

> That means, Sun then must ask a
> possibly external decider to merge a proposed patch.
mg>the concept of an independent judge who will not favour one ethnic group over the rest of the community

> Not vice versa. See
> how hard the work then would be for Sun? Exactly that hard is the work
> currently for external contributors.
>
> > I think we understand how all this works in theory, but in the
> > practice,
> > and in a large organization, there are limits.
mg>the poster that wrote this is imposing his countrys limitations

>And the larger and more
> > complex, and faster moving, the artifact, the harder it is to make
> > these things work.
mg>i would solicit advice from monsieur ellison before writing these emails
mg>a competent manager makes things work vs making excuses
>
> Sure, but you mix up the internal structures of Sun with the public
> structures of the community. Sun has more power in the community than
> others, so there is a skew in the democratic system. Also you seem to
> ignore that there are publicly managed projects like Linux and KDE which
> work very well without having a single vendor making decisions behind
> closed doors. And I would not say Linux and KDE is plain theory. From
> you view you could also say, there is no need for public elections as Mr
> Bush already has had embassadors all over the world solving regional
> conflicts without public discussionis .
mg>currently retired and doesnt read emails.. so that point is moot

> But sure we all are glad that the
> president gets elected and that the people can vote for or against
> particular actions. Don't we?
mg>hierarchical models work for organisations with large uneducated pool of labour accomplishing simple repeatable tasks
mg>technological problems dictate a flatter model with participation and discussion by all who wish to participate
mg>attempts to move a current democratic process into a hierarchical model will be met with justifiable resistance

> Regards
> Markus
mg>Regards
mg>Martin
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>

_________________________________________________________________
Bingâ„¢ brings you maps, menus, and reviews organized in one place. Try it now.
http://www.bing.com/search?q=restaurants&form=MLOGEN&publ=WLHMTAG&crea=T...
[att1.html]

Eduardo Pelegri-Llopart

See comments inlined. I've elided some text...

Martin Gainty wrote:
> mg>hopefully quick comments
>
>> From: karg@quipsy.de
>> To: users@glassfish.dev.java.net
>> Subject: RE: Re: GlassFish Contributions (was GLASSFISH IS LAME)
>>
>>> re: local autonomy -- sure, we have been working on that for many
>>> years.
>>> Sun has had strong investments in India at least since mid 90s. We
>>> have
>>> local managers, local directors, etc, etc.
> mg>boasting "excluding development/support/feedback from the rest of the planet" is a plus?

I believe you are misreading my post.

Nobody is excluding anybody. Nowadays all development / product
outreach is global but, even if one assumes a shared language, the TZ
distances are not going away.

I was replying to Markus comment on the benefit of aligning development
tasks along geographic (time-zone) clusters to reduce the communication
/ TZ-differential cost. This is a common, and good, practice in the
industry, and Sun also follows it.

The topic came up because I was pointing to the need of synchronous
meetings to coordinate across efforts in multi-TZs. Markus said one
could reduce the need for these by this grouping; I replied saying we
already do it, but the need remains.

> mg>and make the whole process truly democratic..good point
> mg>the concept of an independent judge who will not favour one ethnic group over the rest of the community

See my previous post. Also, from a pragmatic perspective, an
"independent judge" would be _VERY_ hard to find.

> mg>the poster that wrote this is imposing his countrys limitations

I don't know what you could mean with that statement.

> mg>i would solicit advice from monsieur ellison before writing these emails

I'll ignore this one.

> mg>a competent manager makes things work vs making excuses

No excuses intended. I'm explaining how things work (and work well) for
us. And I'm not a manager, btw.

> mg>hierarchical models work for organisations with large uneducated pool of labour accomplishing simple repeatable tasks
> mg>technological problems dictate a flatter model with participation and discussion by all who wish to participate
> mg>attempts to move a current democratic process into a hierarchical model will be met with justifiable resistance

I don't know what you mean; hopefully we can skip that.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Stephen Connolly

2009/7/24 Eduardo Pelegri-Llopart

> Apache only uses asynchronous - just email afaik - tools. Given their
> goals that seems right. I think our use of both types of tools is the right
> for our goals and our size... and the level of external interest.
>
>
Hmmm not true, for example Apache Maven uses #maven @ irc.codehaus.org (it's
at codehaus.org for hystorical[1] reasons)

-Stephen

[1] a combination of hysterical and historical
[att1.html]

km
Offline
Joined: 2005-10-28

Stephen -- hystorical -- great addition to my vocab ;)

Cheers.

Martin Gainty

i guess that depends on your definition of 'defect'

what marcus is suggesting is that the process of prioritisation of bugs as well as bug assignment should be discussed by the OSS community..a forum similar to a bundestag or american congress where there would be open discussion for a full day..vote the next day on should the bug be addressed..and whether to branch the patch or release affected classes into next point rel
Also there would be a vote on who the bug gets assigned to (approved by majority of committers)
and their should be an overseer so that if the guy assigned decides to go on vaca for a month the overseer
will become responsible for implementing the patch within the promised timeline
i agree this will slow the cycle of OpenDiscussion-BugVote-CodeBug-UAT-integration-acceptance cycle but this would allow the OSS community to have input

i subscribe to another apache group that had a clear security bug assigned..the individual auto-assigned doesnt
have the right skillset to get the job done..with the end result the patch will take upwards of a week or 2 instead of a day/2day
the client requesting the sec patch has specific needs and will probably fail the patch with the end result that a month
will go by before this *not so poor* client will gethis security patch
Lastly Unit tests should be included in the patch distribution to determine validity of functional,unit,regression and integration capabilities of the patch

Mit Freundlichen Grüßen
Martin
______________________________________________
Verzicht und Vertraulichkeitanmerkung

Diese Nachricht ist vertraulich. Sollten Sie nicht der vorgesehene Empfaenger sein, so bitten wir hoeflich um eine Mitteilung. Jede unbefugte Weiterleitung oder Fertigung einer Kopie ist unzulaessig. Diese Nachricht dient lediglich dem Austausch von Informationen und entfaltet keine rechtliche Bindungswirkung. Aufgrund der leichten Manipulierbarkeit von E-Mails koennen wir keine Haftung fuer den Inhalt uebernehmen.

> Date: Thu, 23 Jul 2009 11:01:14 -0700
> From: pelegri@sun.com
> To: users@glassfish.dev.java.net
> Subject: Bug Fixing, Support, German Laws, etc... (was: GLASSFISH IS LAME)
>
> Hi Markus.
>
> You and I exchanged emails on this general topic a week ago but others
> may need additional background to understand your perspective on bug
> fixing and support, so, trying to capture that in a couple of paragraphs...
>
> Markus' context is that there are laws in Germany that require consumer
> products to be free of defects. He interprets that to say that a
> customer of a software product must be given free fixes for all
> "critical" bugs in the product. Sun includes that in our commercial,
> for-fee, GlassFish offering, but Markus says it should apply to the free
> distribution also. Sun commits to push all commercial bug fixes to the
> next free/public distribution, but the isolated, sustaining bug fixes
> are not available _separately_ before that.
>
> It is a bit more complicated than that, and IANAL, etc, but I think that
> gives some background.
>
> There is also a separate topic about what is a P1; I think that needs to
> be addressed better.
>
> From our perspective, much of this boils down to different ways to
> balance the needs of Sun, who is contributing many people to deliver
> GlassFish, with the needs of non-paying users, paying customers, and
> developers and partners.
>
> There are other ways of balancing things. RedHat, Covalent, MySQL,
> JasperSoft, etc, etc, they all use different mechanisms to balance the
> needs of all these stakeholders.
>
> So far, I think most people in the GlassFish community have been happy
> with the tradeoffs we have been following: we have been delivering free
> good quality releases in a regular fashion, while building the next
> generation v3 product and providing "support" to users and customers.
> The last few months have been a bit extra tricky with summer, Oracle,
> JavaOne, etc, but hopefully we are getting back to our normal cycles.
>
> Hope this helps,
> - eduard/o
>
>
> Markus Karg wrote:
> >> Do you have any idea what open source stands for?
> >> I am a long time believer in open source. And I have nothing against
> >> paying for support on open source products that actually show promise.
> >
> > I think nobody has a problem with paying good money for good support. But with GlassFish's P1 bugs it is different. Those bug reports have been investigated in long and expensive research by the users (they did not get money for that from Sun or anybody else), and it should be fair that those users get the bug fix for free, too. I do not see bug fixing as a "support".
> >
> > Regards
> > Markus
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>

_________________________________________________________________
Windows Live™ Hotmail®: Search, add, and share the web’s latest sports videos. Check it out.
http://www.windowslive.com/Online/Hotmail/Campaign/QuickAdd?ocid=TXT_TAG...
[att1.html]

Markus Karg

> how many times did you join the weekly engineering meeting ?
> zero.
> I make an announcement every week on the dev@ alias.

First, I did not say that I want to be the one that participates to the
progress, I just said that I want to be able to vote for the community
members doing that. This is a difference.

Please check this page:
https://glassfish.dev.java.net/public/devindex.html

Do you see any note about the weekly engineering meeting or that
architectural questions are discussed on the dev alias?

And do you think a meeting is a good idea, with respect to the fact that
all contributors are spread all over the world and don't like the idea
to stand up in the middle of the night just because someone (who?)
decided that there are meetings instead of a public online discussion
and voting system?

> if you don't try to participate, you cannot complain later that you
> were not allowed to participate. I know you have actively and
> positively participated but if you want to become part of the
> decisions, you should make yourself visible to the meetings where
> things are decided.

Well, then you should add this possibility to the page linked above, and
you should tell everybody in the GlassFish community to discuss their
internal stuff in that meeting. I was a contributor to TopLink and
Jersey and I doubt that their internal decision had been discussed in
that meeting. Or am I wrong here? Or do you translate "GlassFish
contribution" with "GlassFish Application Server Kernel Contribution"?

> As for the code being spaghetti style, It's certainly what I was
> suggesting (in a more subtle way...) that GF code in general is
> intimidating, I truly believe it got better in v3 but it's definitely
> not an easy task for anyone.

Well, I am still (at least formally) an EclipseLink contributor and I
can tell you that it is still Spaghetti Code (and those guys bought
Sun!) and all my attempts to improve them had been responded with "this
is something that we will not discuss with anybody outside of Oracle".
What a nice and heartly invitation to participate in favour of quality!

> so if you want to participate, please join the engineering meeting,
> get the meet the stakeholders of the area you care about and start
> discussions on what's important to you and that's an advice for anyone
> on this mailing list !

As long as all the architectural problems are really discussed on the
mailing list from beginning to end, thank you for the invitation, I will
join and tell you all the design problems I see in GlassFish and I will
expect that they will get discussed seriously (otherwise it makes no
sense to participate). But be assured that it is impossible to attend to
meetings in the middle of the local night. So if you want to have world
wide contributors, you need to understand that meetings are a bad idea.
If you want an open process, you need to understand that the community
needs to find a process management that is working without drawbacks to
some of us while others just go to work at normal office hours (yes I
know how hard that is without touching a phone, I already managed
several earth spanning open source projects in the past 25 years of
coding).

Regards
Markus

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Eduardo Pelegri-Llopart

> And do you think a meeting is a good idea, with respect to the fact that
> all contributors are spread all over the world and don't like the idea
> to stand up in the middle of the night just because someone (who?)
> decided that there are meetings instead of a public online discussion
> and voting system?

It is a tradeoff between making it very easy for people to contribute
and allowing faster decisions with a higher bandwidth.

What we have been doing is combining synchronous (meetings in different
modalities) and asynchronous (mailing lists and blogs) meetings.
Subcomponent groups tend to favor asynchronous mailing lists discussions
because they are smaller and have more participation (and I know the two
are related). The larger integration of all the pieces tends to favor
the synchronous communication mechanisms. But both tools are used.

Apache only uses asynchronous - just email afaik - tools. Given their
goals that seems right. I think our use of both types of tools is the
right for our goals and our size... and the level of external interest.

If we had many external contributors wanting to participate in the
larger architectural decisions, we would lean more on synchronous tools.

- eduard/o

Markus Karg wrote:
>> how many times did you join the weekly engineering meeting ?
>> zero.
>> I make an announcement every week on the dev@ alias.
>
> First, I did not say that I want to be the one that participates to the
> progress, I just said that I want to be able to vote for the community
> members doing that. This is a difference.
>
> Please check this page:
> https://glassfish.dev.java.net/public/devindex.html
>
> Do you see any note about the weekly engineering meeting or that
> architectural questions are discussed on the dev alias?
>
> And do you think a meeting is a good idea, with respect to the fact that
> all contributors are spread all over the world and don't like the idea
> to stand up in the middle of the night just because someone (who?)
> decided that there are meetings instead of a public online discussion
> and voting system?
>
>> if you don't try to participate, you cannot complain later that you
>> were not allowed to participate. I know you have actively and
>> positively participated but if you want to become part of the
>> decisions, you should make yourself visible to the meetings where
>> things are decided.
>
> Well, then you should add this possibility to the page linked above, and
> you should tell everybody in the GlassFish community to discuss their
> internal stuff in that meeting. I was a contributor to TopLink and
> Jersey and I doubt that their internal decision had been discussed in
> that meeting. Or am I wrong here? Or do you translate "GlassFish
> contribution" with "GlassFish Application Server Kernel Contribution"?
>
>> As for the code being spaghetti style, It's certainly what I was
>> suggesting (in a more subtle way...) that GF code in general is
>> intimidating, I truly believe it got better in v3 but it's definitely
>> not an easy task for anyone.
>
> Well, I am still (at least formally) an EclipseLink contributor and I
> can tell you that it is still Spaghetti Code (and those guys bought
> Sun!) and all my attempts to improve them had been responded with "this
> is something that we will not discuss with anybody outside of Oracle".
> What a nice and heartly invitation to participate in favour of quality!
>
>> so if you want to participate, please join the engineering meeting,
>> get the meet the stakeholders of the area you care about and start
>> discussions on what's important to you and that's an advice for anyone
>> on this mailing list !
>
> As long as all the architectural problems are really discussed on the
> mailing list from beginning to end, thank you for the invitation, I will
> join and tell you all the design problems I see in GlassFish and I will
> expect that they will get discussed seriously (otherwise it makes no
> sense to participate). But be assured that it is impossible to attend to
> meetings in the middle of the local night. So if you want to have world
> wide contributors, you need to understand that meetings are a bad idea.
> If you want an open process, you need to understand that the community
> needs to find a process management that is working without drawbacks to
> some of us while others just go to work at normal office hours (yes I
> know how hard that is without touching a phone, I already managed
> several earth spanning open source projects in the past 25 years of
> coding).
>
> Regards
> Markus
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

> Apache only uses asynchronous - just email afaik - tools. Given their
> goals that seems right. I think our use of both types of tools is the
> right for our goals and our size... and the level of external
interest.
>
> If we had many external contributors wanting to participate in the
> larger architectural decisions, we would lean more on synchronous
> tools.

Eduardo, ever thought about the fact that external participation is low,
DUE TO the synchronous tools...?

Regards
Markus

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Eduardo Pelegri-Llopart

Sure, I know there is a feedback loop, but this is all about
(engineering) tradeoffs.

Two large concentrations of GF developers are in the West Coast of the
US and in India. Asynchronous mail-based coordination between those two
sites is very hard. A common symptom is the "nothing happens while I'm
awake; I went to sleep and whole conversations started, flamed and
ended". Or the "two-email exchanges, 10hrs pause, two email exchanges...".

It's different when everybody is in the US or even US+Europe (& geos in
close TZs, see [1]); then you can have a real conversation by email.
Blogs help with the TZ and synchronous/asynchonous issues, see [2], but
there is a limit to what can be done that way.

Jerome would know, but I think a key part of the engineering meeting is
to address these problems.

[1]http://weblogs.java.net/blog/pelegri/archive/2003/08/distance_in_the.html
[2]http://weblogs.java.net/blog/pelegri/archive/2006/02/time_zones_dont_1.html

Processes evolve to track communities needs. Over time we have added
wikis, twitters, ustream, etc. If you want to contribute we can figure
out how to make it easier for you to participate.

- eduard/o

Markus Karg wrote:
>> Apache only uses asynchronous - just email afaik - tools. Given their
>> goals that seems right. I think our use of both types of tools is the
>> right for our goals and our size... and the level of external
> interest.
>> If we had many external contributors wanting to participate in the
>> larger architectural decisions, we would lean more on synchronous
>> tools.
>
> Eduardo, ever thought about the fact that external participation is low,
> DUE TO the synchronous tools...?
>
> Regards
> Markus
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

Eduardo,

I thought you are in holidays, is the weather so bad that you prefere
discussions with me? ;-)

> Two large concentrations of GF developers are in the West Coast of the
> US and in India. Asynchronous mail-based coordination between those
> two
> sites is very hard. A common symptom is the "nothing happens while
I'm
> awake; I went to sleep and whole conversations started, flamed and
> ended". Or the "two-email exchanges, 10hrs pause, two email
> exchanges...".

This is just a symptom of an architectural design of the overall system.
If the line has a long delay (like internet to the moon), the solution
is to put more power in the nodes and reduce the communication
roundtrips by increasing the semantic information per package. Isn't it
what we all tell our students? :-)

What I like to say is: If you split the system into small bricks, an
have one elected chief for each brick, then there shouldn't be a need
for speedy discussions. If you want to change the brick, develop a
patch, send it to the chief, and wait for his detailed answer. No need
for lots of emails going back and forth. The patch will be containing
work for several days, and the answer will contain details that last for
several days.

Heavy discussions with lots of small emails going back and forth is the
symptom of a hierarchical system where lots of idiots work for a
stressed manager. This is NOT what the community shall be like. If the
chief and the contributor both know their area of work very well (what
is given if the brick is small enough) then they do not need to send
lots of emails.

At least this is my experience with both company architectures. We tried
both of them and recognized that it makes no sense to have masses of
unexperienced hackers far away and try to manage them centrally.
Instead, have few highly experienced architects working in a losely
coupled way. The communicate less mails, but more facts per email.

For example, in our company, we just exchange information about once a
week by email. Each engineer can work when and where he wants, since he
knows his brick so well. I just have to publish a tasks lists and people
pick their jobs. No need for more communication or daily scrum meetings
etc. It's just about knowledge and responsibility. Not about time zones
and reply delays.

Regards
Markus

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Eduardo Pelegri-Llopart

re: holidays -- yep, but I have break times while I wait for other events.

re: small bricks -- we have been working on that for several years...

re: local autonomy -- sure, we have been working on that for many years.
Sun has had strong investments in India at least since mid 90s. We have
local managers, local directors, etc, etc.

I think we understand how all this works in theory, but in the practice,
and in a large organization, there are limits. And the larger and more
complex, and faster moving, the artifact, the harder it is to make these
things work.

- eduard/o

Markus Karg wrote:
> Eduardo,
>
> I thought you are in holidays, is the weather so bad that you prefere
> discussions with me? ;-)
>
>> Two large concentrations of GF developers are in the West Coast of the
>> US and in India. Asynchronous mail-based coordination between those
>> two
>> sites is very hard. A common symptom is the "nothing happens while
> I'm
>> awake; I went to sleep and whole conversations started, flamed and
>> ended". Or the "two-email exchanges, 10hrs pause, two email
>> exchanges...".
>
> This is just a symptom of an architectural design of the overall system.
> If the line has a long delay (like internet to the moon), the solution
> is to put more power in the nodes and reduce the communication
> roundtrips by increasing the semantic information per package. Isn't it
> what we all tell our students? :-)
>
> What I like to say is: If you split the system into small bricks, an
> have one elected chief for each brick, then there shouldn't be a need
> for speedy discussions. If you want to change the brick, develop a
> patch, send it to the chief, and wait for his detailed answer. No need
> for lots of emails going back and forth. The patch will be containing
> work for several days, and the answer will contain details that last for
> several days.
>
> Heavy discussions with lots of small emails going back and forth is the
> symptom of a hierarchical system where lots of idiots work for a
> stressed manager. This is NOT what the community shall be like. If the
> chief and the contributor both know their area of work very well (what
> is given if the brick is small enough) then they do not need to send
> lots of emails.
>
> At least this is my experience with both company architectures. We tried
> both of them and recognized that it makes no sense to have masses of
> unexperienced hackers far away and try to manage them centrally.
> Instead, have few highly experienced architects working in a losely
> coupled way. The communicate less mails, but more facts per email.
>
> For example, in our company, we just exchange information about once a
> week by email. Each engineer can work when and where he wants, since he
> knows his brick so well. I just have to publish a tasks lists and people
> pick their jobs. No need for more communication or daily scrum meetings
> etc. It's just about knowledge and responsibility. Not about time zones
> and reply delays.
>
> Regards
> Markus
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

gniontas
Offline
Joined: 2009-07-15

> Sun has had strong investments in India at least since mid 90s.
> We have local managers, local directors, etc, etc.

oooooh. I rest my case:)
Now I know exactly why nothing gets done. Where are the Engineers ??

cdonaldson
Offline
Joined: 2007-11-13

The nature of your posts are staggering. After multiple inflammatory posts, including

"C'mon glassfish team. It's a fucking brand new installation. This just tells me that glassfish is just lame and the glassfish developers and maintainers a bunch of losers."

your major install hurdle was linked to an issue that was resolved by following the release notes. Further issues were being addressed by multiple people in the forums. And now the latest tirade.

Sun is a global company with campuses spread throughout the world, including India, and with engineers spread through the world, including India. Some of whom I believe were answering your questions, despite your antics. I'm done, and would be surprised to see anyone else continue this or the other threads.

Eduardo Pelegri-Llopart

Bit by bit we are cleaning things up and making it easier to contribute.

Currently most (non-corporate) contributions to GlassFish happen in the
subprojects, places like Grizzly, Jersey, Mojarra/JSF, Metro, etc. Over
time I expect us to have more of these, and the modularity in v3 (once
we clean up all the kinks) will also help.

That said, note that just having a bug fix ready does not necessarily
mean it can be incorporated into the main release right away. It is an
intrinsic tradeoff from "Public releases are fully tested and supported"
and "Getting releases out in scheduled dates".

...

I'd also like to second Jerome and encourage participation on the
engineering meetings. We try to be very transparent and public in all
decisions, but the more individual contributors there, the easier it is
to do a good job.

- eduard/o

Markus Karg wrote:
> Jerome,
>
> still you did not convince me (and remember, I *am* / *was* a
> contributor to Java EE 6 / 5 in projects).
>
> The point is that GlassFish's internal structures are completely Sun /
> Oracle inventions, so nobody outside of this companies has the time and
> knowledge to understand it. If you really want to get volunteers in the
> boat (remember, we are no students, we all are long-experienced software
> architects), you must get the code to a style that allows externals to
> work on small components which can be overseen in the spare time. I
> tried with EclipseLink. It is nearly impossible, since it is Spaghetti
> Code in some areas.
>
> Also, people not only want to do the "dirty work" like doing the hacking
> while others do the decisions. They want to participiate in the
> decisions. Remember, we do not talk about specification decisions
> (everybody can become part of JCP), we talk about *implementation*
> decision. Why? Because this thread started about bugs, and bugs are not
> decided by JCP, obviously.
>
> So in theory you are right, everybody can participate in JCP. But in
> reality nobody cares for JCP, they all care for the implementation,
> since this is what users see and feel later. And THERE nobody outside of
> Sun / Oracle can officially participate. Or did I miss a link where I
> can add myself to the EJB 3.1 Client Container team to actively
> participate in the decisions that where leading to the hangs that I
> still experience...?
>
> Regards
> Markus
>
>> The processes to define what goes in glassfish is much more open than
>> what you seem to think. First of all, even Sun engineers don't have
>> that much to say in what goes into glassfish because it's mostly
>> driven by specifications. Specifications are driven within the JCP and
>> are open to external participants so one good way of driving features
>> into Glassfish and others appservers is actually to participate to the
>> JCP process. Above the Java EE specifications, we have also released
>> all the product specifications by feature and that's also an
>> opportunity to add or criticize.
>>
>> On top of that, we have had a lot of committers willing to participate
>> to the GlassFish project itself. I have personally tried to involve
>> some of these external committers to some task (yes they have to start
>> small so we gain confidence) but in general, the GlassFish project
>> being so big, the source code is intimidating, or the people just end
>> up not having the time, most of them don't follow up. We do have some
>> who participates daily and I don't think we ever kept them in the
>> dark. All meetings are open for anyone to join at least by phone.
>>
>> At the end, I don't remember shutting out anyone willing to
>> participate (granted his level of participation was useful) but the
>> reality is that we cannot expect external folks to be able to spend
>> considerable amount of their personal time on the project which would
>> be necessary to not introduce more risks to a schedule.
>>
>> For such a big project, the overwhelming size of the code, the time
>> involved in understanding enough of GF is probably the biggest barrier
>> to participation but it's certainly not a conscious decision from Sun
>> to hide anything.
>>
>>> Regards
>>> Markus
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
>> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Jerome Dochez

On Jul 23, 2009, at 6:03 AM, Markus Karg wrote:

>> Markus Karg schrieb:
>>> While I still am not a big friend of the agressive style of
>> conversion
>>> you apply, I must confess that I fear that you are right. ;-( This
>>> is
>>> what I try to change, but it is hard as Oracle/Sun certainly do not
>> do
>>> this project for fun, but for money.
>>
>> Well... I guess most of these companies involved in open source
>> projects do
>> right this "for money", be that directly or indirectly. One thing
>> which
>> is
>> different about the Sun open source projects, here, in my opinion is
>> that,
>> ultimately, the source code of these projects is open and available
>> but
>> at
>> the end it seems to be nothing more than "open source Sun
>> repositories"
>> - no
>> real developer community outside Sun (feel free to correct me if I'm
>> all
>> wrong on that), no real community in terms of making decisions
>> regarding
>> project development and management, and so forth. This, ultimately,
>> makes
>> people suspicious as a lot of them (including us, actually) seem /
>> tend
>> to
>> rely upon open source software also for the reason of not being fully
>> dependent upon one special vendor providing this software (a
>> developer
>> friend of mine just recently had to experience this as Oracle decided
>> to
>> discontinue his beloved OAS in favor of WebLogic). So even being
>> "open
>> source", Glassfish won't eventually help preventing such a
>> scenario. :(
>
> Just what I say: The source is open, but the project is not -- but
> these days people expect open processes when reading about open
> source. Actually there is such a community outside of Sun which
> would be happy to participate in the decision making, but Sun is not
> offering an open process to let them participate. This is a
> difference.

I am sorry but I disagree.

The processes to define what goes in glassfish is much more open than
what you seem to think. First of all, even Sun engineers don't have
that much to say in what goes into glassfish because it's mostly
driven by specifications. Specifications are driven within the JCP and
are open to external participants so one good way of driving features
into Glassfish and others appservers is actually to participate to the
JCP process. Above the Java EE specifications, we have also released
all the product specifications by feature and that's also an
opportunity to add or criticize.

On top of that, we have had a lot of committers willing to participate
to the GlassFish project itself. I have personally tried to involve
some of these external committers to some task (yes they have to start
small so we gain confidence) but in general, the GlassFish project
being so big, the source code is intimidating, or the people just end
up not having the time, most of them don't follow up. We do have some
who participates daily and I don't think we ever kept them in the
dark. All meetings are open for anyone to join at least by phone.

At the end, I don't remember shutting out anyone willing to
participate (granted his level of participation was useful) but the
reality is that we cannot expect external folks to be able to spend
considerable amount of their personal time on the project which would
be necessary to not introduce more risks to a schedule.

For such a big project, the overwhelming size of the code, the time
involved in understanding enough of GF is probably the biggest barrier
to participation but it's certainly not a conscious decision from Sun
to hide anything.

>
> Regards
> Markus

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

Jerome,

still you did not convince me (and remember, I *am* / *was* a
contributor to Java EE 6 / 5 in projects).

The point is that GlassFish's internal structures are completely Sun /
Oracle inventions, so nobody outside of this companies has the time and
knowledge to understand it. If you really want to get volunteers in the
boat (remember, we are no students, we all are long-experienced software
architects), you must get the code to a style that allows externals to
work on small components which can be overseen in the spare time. I
tried with EclipseLink. It is nearly impossible, since it is Spaghetti
Code in some areas.

Also, people not only want to do the "dirty work" like doing the hacking
while others do the decisions. They want to participiate in the
decisions. Remember, we do not talk about specification decisions
(everybody can become part of JCP), we talk about *implementation*
decision. Why? Because this thread started about bugs, and bugs are not
decided by JCP, obviously.

So in theory you are right, everybody can participate in JCP. But in
reality nobody cares for JCP, they all care for the implementation,
since this is what users see and feel later. And THERE nobody outside of
Sun / Oracle can officially participate. Or did I miss a link where I
can add myself to the EJB 3.1 Client Container team to actively
participate in the decisions that where leading to the hangs that I
still experience...?

Regards
Markus

> The processes to define what goes in glassfish is much more open than
> what you seem to think. First of all, even Sun engineers don't have
> that much to say in what goes into glassfish because it's mostly
> driven by specifications. Specifications are driven within the JCP and
> are open to external participants so one good way of driving features
> into Glassfish and others appservers is actually to participate to the
> JCP process. Above the Java EE specifications, we have also released
> all the product specifications by feature and that's also an
> opportunity to add or criticize.
>
> On top of that, we have had a lot of committers willing to participate
> to the GlassFish project itself. I have personally tried to involve
> some of these external committers to some task (yes they have to start
> small so we gain confidence) but in general, the GlassFish project
> being so big, the source code is intimidating, or the people just end
> up not having the time, most of them don't follow up. We do have some
> who participates daily and I don't think we ever kept them in the
> dark. All meetings are open for anyone to join at least by phone.
>
> At the end, I don't remember shutting out anyone willing to
> participate (granted his level of participation was useful) but the
> reality is that we cannot expect external folks to be able to spend
> considerable amount of their personal time on the project which would
> be necessary to not introduce more risks to a schedule.
>
> For such a big project, the overwhelming size of the code, the time
> involved in understanding enough of GF is probably the biggest barrier
> to participation but it's certainly not a conscious decision from Sun
> to hide anything.
>
> >
> > Regards
> > Markus
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Jerome Dochez

how many times did you join the weekly engineering meeting ?
zero.

I make an announcement every week on the dev@ alias.
if you don't try to participate, you cannot complain later that you
were not allowed to participate. I know you have actively and
positively participated but if you want to become part of the
decisions, you should make yourself visible to the meetings where
things are decided.

As for the code being spaghetti style, It's certainly what I was
suggesting (in a more subtle way...) that GF code in general is
intimidating, I truly believe it got better in v3 but it's definitely
not an easy task for anyone.

so if you want to participate, please join the engineering meeting,
get the meet the stakeholders of the area you care about and start
discussions on what's important to you and that's an advice for anyone
on this mailing list !

jerome

On Jul 23, 2009, at 7:15 AM, Markus Karg wrote:

> Jerome,
>
> still you did not convince me (and remember, I *am* / *was* a
> contributor to Java EE 6 / 5 in projects).
>
> The point is that GlassFish's internal structures are completely Sun /
> Oracle inventions, so nobody outside of this companies has the time
> and
> knowledge to understand it. If you really want to get volunteers in
> the
> boat (remember, we are no students, we all are long-experienced
> software
> architects), you must get the code to a style that allows externals to
> work on small components which can be overseen in the spare time. I
> tried with EclipseLink. It is nearly impossible, since it is Spaghetti
> Code in some areas.
>
> Also, people not only want to do the "dirty work" like doing the
> hacking
> while others do the decisions. They want to participiate in the
> decisions. Remember, we do not talk about specification decisions
> (everybody can become part of JCP), we talk about *implementation*
> decision. Why? Because this thread started about bugs, and bugs are
> not
> decided by JCP, obviously.
>
> So in theory you are right, everybody can participate in JCP. But in
> reality nobody cares for JCP, they all care for the implementation,
> since this is what users see and feel later. And THERE nobody
> outside of
> Sun / Oracle can officially participate. Or did I miss a link where I
> can add myself to the EJB 3.1 Client Container team to actively
> participate in the decisions that where leading to the hangs that I
> still experience...?
>
> Regards
> Markus
>
>> The processes to define what goes in glassfish is much more open than
>> what you seem to think. First of all, even Sun engineers don't have
>> that much to say in what goes into glassfish because it's mostly
>> driven by specifications. Specifications are driven within the JCP
>> and
>> are open to external participants so one good way of driving features
>> into Glassfish and others appservers is actually to participate to
>> the
>> JCP process. Above the Java EE specifications, we have also released
>> all the product specifications by feature and that's also an
>> opportunity to add or criticize.
>>
>> On top of that, we have had a lot of committers willing to
>> participate
>> to the GlassFish project itself. I have personally tried to involve
>> some of these external committers to some task (yes they have to
>> start
>> small so we gain confidence) but in general, the GlassFish project
>> being so big, the source code is intimidating, or the people just end
>> up not having the time, most of them don't follow up. We do have some
>> who participates daily and I don't think we ever kept them in the
>> dark. All meetings are open for anyone to join at least by phone.
>>
>> At the end, I don't remember shutting out anyone willing to
>> participate (granted his level of participation was useful) but the
>> reality is that we cannot expect external folks to be able to spend
>> considerable amount of their personal time on the project which would
>> be necessary to not introduce more risks to a schedule.
>>
>> For such a big project, the overwhelming size of the code, the time
>> involved in understanding enough of GF is probably the biggest
>> barrier
>> to participation but it's certainly not a conscious decision from Sun
>> to hide anything.
>>
>>>
>>> Regards
>>> Markus
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
>> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Martin, Ray

Mr Dochez,

You are wasting your time. These people are not interested in helping.
They are whiners. They are chest thumpers - making noise for the sake
of making noise. They waste not only your time but everyone's time.
They are just rude. I empathize with you and your fellow workers. Why
some people believe they should abuse other people on a community forum
I do not understand. Anyone that uses abusive language - and then
receives the advocacy of Mr. Karg - who do these people think they are -
it is folks like these who ruin an excellent forum. It is a sad day
when the whiners and chest thumpers take over a forum.

Mr. Dochez, there are many on this forum who appreciate all the hard
work that you SUN guys are doing. I am sorry that you have to put up
with abuse just to do your job. I wish you lots of success.

Thanx for your hard work,
Ray

-----Original Message-----
From: Jerome.Dochez@Sun.COM [mailto:Jerome.Dochez@Sun.COM]
Sent: Thursday, July 23, 2009 11:09 AM
To: users@glassfish.dev.java.net
Subject: Re: GLASSFISH IS LAME

how many times did you join the weekly engineering meeting ?
zero.

I make an announcement every week on the dev@ alias.
if you don't try to participate, you cannot complain later that you were
not allowed to participate. I know you have actively and positively
participated but if you want to become part of the decisions, you should
make yourself visible to the meetings where things are decided.

As for the code being spaghetti style, It's certainly what I was
suggesting (in a more subtle way...) that GF code in general is
intimidating, I truly believe it got better in v3 but it's definitely
not an easy task for anyone.

so if you want to participate, please join the engineering meeting, get
the meet the stakeholders of the area you care about and start
discussions on what's important to you and that's an advice for anyone
on this mailing list !

jerome

On Jul 23, 2009, at 7:15 AM, Markus Karg wrote:

> Jerome,
>
> still you did not convince me (and remember, I *am* / *was* a
> contributor to Java EE 6 / 5 in projects).
>
> The point is that GlassFish's internal structures are completely Sun /

> Oracle inventions, so nobody outside of this companies has the time
> and knowledge to understand it. If you really want to get volunteers
> in the boat (remember, we are no students, we all are long-experienced

> software architects), you must get the code to a style that allows
> externals to work on small components which can be overseen in the
> spare time. I tried with EclipseLink. It is nearly impossible, since
> it is Spaghetti Code in some areas.
>
> Also, people not only want to do the "dirty work" like doing the
> hacking while others do the decisions. They want to participiate in
> the decisions. Remember, we do not talk about specification decisions
> (everybody can become part of JCP), we talk about *implementation*
> decision. Why? Because this thread started about bugs, and bugs are
> not decided by JCP, obviously.
>
> So in theory you are right, everybody can participate in JCP. But in
> reality nobody cares for JCP, they all care for the implementation,
> since this is what users see and feel later. And THERE nobody outside
> of Sun / Oracle can officially participate. Or did I miss a link where

> I can add myself to the EJB 3.1 Client Container team to actively
> participate in the decisions that where leading to the hangs that I
> still experience...?
>
> Regards
> Markus
>
>> The processes to define what goes in glassfish is much more open than

>> what you seem to think. First of all, even Sun engineers don't have
>> that much to say in what goes into glassfish because it's mostly
>> driven by specifications. Specifications are driven within the JCP
>> and are open to external participants so one good way of driving
>> features into Glassfish and others appservers is actually to
>> participate to the JCP process. Above the Java EE specifications, we
>> have also released all the product specifications by feature and
>> that's also an opportunity to add or criticize.
>>
>> On top of that, we have had a lot of committers willing to
>> participate to the GlassFish project itself. I have personally tried
>> to involve some of these external committers to some task (yes they
>> have to start small so we gain confidence) but in general, the
>> GlassFish project being so big, the source code is intimidating, or
>> the people just end up not having the time, most of them don't follow

>> up. We do have some who participates daily and I don't think we ever
>> kept them in the dark. All meetings are open for anyone to join at
>> least by phone.
>>
>> At the end, I don't remember shutting out anyone willing to
>> participate (granted his level of participation was useful) but the
>> reality is that we cannot expect external folks to be able to spend
>> considerable amount of their personal time on the project which would

>> be necessary to not introduce more risks to a schedule.
>>
>> For such a big project, the overwhelming size of the code, the time
>> involved in understanding enough of GF is probably the biggest
>> barrier to participation but it's certainly not a conscious decision
>> from Sun to hide anything.
>>
>>>
>>> Regards
>>> Markus
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
>> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

Martin,

I do not understand why you say this. Speaking for my own, I am / was a
contributor to a lot of open source projects, including TopLink, Jersey,
GlassFish and others and had lead several global open source projects in
the past. We are discussing seriously the management problems that we
see in the GlassFish community. You do not know me or "us", nor do you
know our real intension. While I find the words to express the problems
that I see, others are not as eloquent or patient like me and use stupid
words like "F* You all". Certainly this is rude, but we should ever
think about WHY they express their situation in such a way. If GlassFish
would be so great as we all think, that "idiot" would not have written
"F* You all", obviously. Frankly, I am using GlassFish for years, and
there had been lots of situations when I had the exact same three words
in my mind (and I call myself a contributor to the GlassFish eco
system). I really appreciate Sun's and Oracle's work, but there is much
potential to improve (especially if one comes from "real" open processes
and now sees that lots of stuff is going on behind "closed" doors) and
we must take all critics seriously to find the actual cause and fix it.
If you just talk to friendly people, you'll oversee a lot of real
problems worth to get fixed.

Regards
Markus

> -----Original Message-----
> From: Martin, Ray [mailto:armart3@tycho.ncsc.mil]
> Sent: Donnerstag, 23. Juli 2009 19:00
> To: users@glassfish.dev.java.net
> Subject: RE: Re: GLASSFISH IS LAME
>
> Mr Dochez,
>
> You are wasting your time. These people are not interested in
helping.
> They are whiners. They are chest thumpers - making noise for the sake
> of making noise. They waste not only your time but everyone's time.
> They are just rude. I empathize with you and your fellow workers.
Why
> some people believe they should abuse other people on a community
forum
> I do not understand. Anyone that uses abusive language - and then
> receives the advocacy of Mr. Karg - who do these people think they are
> -
> it is folks like these who ruin an excellent forum. It is a sad day
> when the whiners and chest thumpers take over a forum.
>
> Mr. Dochez, there are many on this forum who appreciate all the hard
> work that you SUN guys are doing. I am sorry that you have to put up
> with abuse just to do your job. I wish you lots of success.
>
> Thanx for your hard work,
> Ray
>
> -----Original Message-----
> From: Jerome.Dochez@Sun.COM [mailto:Jerome.Dochez@Sun.COM]
> Sent: Thursday, July 23, 2009 11:09 AM
> To: users@glassfish.dev.java.net
> Subject: Re: GLASSFISH IS LAME
>
> how many times did you join the weekly engineering meeting ?
> zero.
>
> I make an announcement every week on the dev@ alias.
> if you don't try to participate, you cannot complain later that you
> were
> not allowed to participate. I know you have actively and positively
> participated but if you want to become part of the decisions, you
> should
> make yourself visible to the meetings where things are decided.
>
> As for the code being spaghetti style, It's certainly what I was
> suggesting (in a more subtle way...) that GF code in general is
> intimidating, I truly believe it got better in v3 but it's definitely
> not an easy task for anyone.
>
> so if you want to participate, please join the engineering meeting,
get
> the meet the stakeholders of the area you care about and start
> discussions on what's important to you and that's an advice for anyone
> on this mailing list !
>
> jerome
>
> On Jul 23, 2009, at 7:15 AM, Markus Karg wrote:
>
> > Jerome,
> >
> > still you did not convince me (and remember, I *am* / *was* a
> > contributor to Java EE 6 / 5 in projects).
> >
> > The point is that GlassFish's internal structures are completely Sun
> /
>
> > Oracle inventions, so nobody outside of this companies has the time
> > and knowledge to understand it. If you really want to get volunteers
> > in the boat (remember, we are no students, we all are long-
> experienced
>
> > software architects), you must get the code to a style that allows
> > externals to work on small components which can be overseen in the
> > spare time. I tried with EclipseLink. It is nearly impossible, since
> > it is Spaghetti Code in some areas.
> >
> > Also, people not only want to do the "dirty work" like doing the
> > hacking while others do the decisions. They want to participiate in
> > the decisions. Remember, we do not talk about specification
decisions
> > (everybody can become part of JCP), we talk about *implementation*
> > decision. Why? Because this thread started about bugs, and bugs are
> > not decided by JCP, obviously.
> >
> > So in theory you are right, everybody can participate in JCP. But in
> > reality nobody cares for JCP, they all care for the implementation,
> > since this is what users see and feel later. And THERE nobody
outside
> > of Sun / Oracle can officially participate. Or did I miss a link
> where
>
> > I can add myself to the EJB 3.1 Client Container team to actively
> > participate in the decisions that where leading to the hangs that I
> > still experience...?
> >
> > Regards
> > Markus
> >
> >> The processes to define what goes in glassfish is much more open
> than
>
> >> what you seem to think. First of all, even Sun engineers don't have
> >> that much to say in what goes into glassfish because it's mostly
> >> driven by specifications. Specifications are driven within the JCP
> >> and are open to external participants so one good way of driving
> >> features into Glassfish and others appservers is actually to
> >> participate to the JCP process. Above the Java EE specifications,
we
> >> have also released all the product specifications by feature and
> >> that's also an opportunity to add or criticize.
> >>
> >> On top of that, we have had a lot of committers willing to
> >> participate to the GlassFish project itself. I have personally
tried
> >> to involve some of these external committers to some task (yes they
> >> have to start small so we gain confidence) but in general, the
> >> GlassFish project being so big, the source code is intimidating, or
> >> the people just end up not having the time, most of them don't
> follow
>
> >> up. We do have some who participates daily and I don't think we
ever
> >> kept them in the dark. All meetings are open for anyone to join at
> >> least by phone.
> >>
> >> At the end, I don't remember shutting out anyone willing to
> >> participate (granted his level of participation was useful) but the
> >> reality is that we cannot expect external folks to be able to spend
> >> considerable amount of their personal time on the project which
> would
>
> >> be necessary to not introduce more risks to a schedule.
> >>
> >> For such a big project, the overwhelming size of the code, the time
> >> involved in understanding enough of GF is probably the biggest
> >> barrier to participation but it's certainly not a conscious
decision
> >> from Sun to hide anything.
> >>
> >>>
> >>> Regards
> >>> Markus
> >>
> >>
> >>
--------------------------------------------------------------------
> -
> >> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> >> For additional commands, e-mail: users-help@glassfish.dev.java.net
> >
> >
> >
---------------------------------------------------------------------
> > To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> > For additional commands, e-mail: users-help@glassfish.dev.java.net
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Paul Sandoz

On Jul 22, 2009, at 10:04 PM, Mark Mielke wrote:
>
> My only complaint with Sun is how slow things move, and especially
> of late. I'd like Java EE 6 to be out as a production quality
> release tomorrow. :-)
>

Or yesterday ? :-)

EE 6 involves multiple parties and multiple specifications that need
to align, some of these specifications are not yet final and while
close to being so there are still some ongoing changes/discussions,
and of course the relevant implementations need to change as well.

Getting all these technical "ducks" aligned is hard work (the "herding
cats" analogy also applies) so it can take time. In my personal
opinion i think many people from many companies or as individuals are
working quite hard to ensure that EE 6 gets finalized.

I think where the community can help is when an EE 6 feature gets put
into the trunk that people can give it is test drive in a nightly
build. For example many EJB 3.1 features are now available, IIRC there
is some JSF stuff, and there is partial integration of 299.

I work on JAX-RS/Jersey and EJB 3.1 integration is available. I will
be implementing Servlet 3.0 and 299/330 integration at some point. I
will ping this list when such implementations become available. The
Aquarium blog is a good source of information for when features are
available.

Paul.
[att1.html]

Markus Karg

The community appreciates all that work I it is clear that Mark was
joking.

But actually, the community wants to do more than just nightly test
drives. They want contribute to the features, and they want to
contribute to the specifications and project management. And that is a
point where community ends in the Sun / Oracle world currently.

Regards

Markus

From: Paul.Sandoz@Sun.COM [mailto:Paul.Sandoz@Sun.COM]
Sent: Donnerstag, 23. Juli 2009 15:32
To: users@glassfish.dev.java.net
Subject: Java EE 6

On Jul 22, 2009, at 10:04 PM, Mark Mielke wrote:

My only complaint with Sun is how slow things move, and
especially of late. I'd like Java EE 6 to be out as a production quality
release tomorrow. :-)

Or yesterday ? :-)

EE 6 involves multiple parties and multiple specifications that need to
align, some of these specifications are not yet final and while close to
being so there are still some ongoing changes/discussions, and of course
the relevant implementations need to change as well.

Getting all these technical "ducks" aligned is hard work (the "herding
cats" analogy also applies) so it can take time. In my personal opinion
i think many people from many companies or as individuals are working
quite hard to ensure that EE 6 gets finalized.

I think where the community can help is when an EE 6 feature gets put
into the trunk that people can give it is test drive in a nightly build.
For example many EJB 3.1 features are now available, IIRC there is some
JSF stuff, and there is partial integration of 299.

I work on JAX-RS/Jersey and EJB 3.1 integration is available. I will be
implementing Servlet 3.0 and 299/330 integration at some point. I will
ping this list when such implementations become available. The Aquarium
blog is a good source of information for when features are available.

Paul.

[att1.html]

vkoniecz
Offline
Joined: 2008-11-14

Hello,

There are some interesting points made in this thread.
I hope that everyone will not get too hot so that something good could happen from this.

In a way, I agree with Markus. Not on all points but on the main points, I agree.

I am also an experienced software architect so I also think that the entry ticket may be high for some people. I am not working for a software editor but for a manufacturing company.
The discussions are often very technical (could be expected) and sometimes I have the impression that the functionalities are a bit forgotten behind the technical side.

JEE specifications exist so that higher level functionalities can be built atop all the core functionalities defined in those specs.
Unfortunately those higher level functionalities are developed leading to constraints that may not penalize a small company with a few adaministrators or developers. But it is very painful to deploy GlassFish when you deal with 1000s projects / 1000s developers / 100s administrators, ... .
An example: we can not use LDAP authentication because it requires a asadmin group.
We have a LDAP, a very huge LDAP ... so we can not define one asadmin group or as many asadmin subgroup as would be required. Tt is difficult to understand that this kind of constraints still exist ... But it exists and we are trying to deploy GlassFish despite that maybe the company I work for may be just too big for GlassFish ...
On the other side, its size may not make it representative of the average company using GlassFish. But there may be a lot of companies that would be blocked by such a constraint.

Some other expected higher level functionalities are cumbersome to use at the present time. (expected from an Enterprise level server)
And GlassFish may end being used only as a .... web server ... (Servlet + JSP)

Well implementing JEE specs is one thing, providing those functionalities is another one.
All this to say that the community may well contribute positively with the specification of those higher level functionalities so that they can be used by as many people as possible by providing relevant feedbacks.
And well, we can also say there is some chance that the implementation of those higher level functionalities may have impacts on the implementation of the core functionalities ...
For contributing with code, it is more difficult. It depends on every company policy.

The other point could be that my company is interested in a stable and mature application server for production usage. So JEE6 and GlassFish v3 may be out ot question for several years.

So what about the engineering meetings with JEE5 and GF v2.1 ?

My 25 cents ... I hope ...

Bye

John Clingan

Good to hear from you again, vkoniecz. We appreciate all of the
feedback on this thread. More below.

On Jul 23, 2009, at 8:52 AM, glassfish@javadesktop.org wrote:

> Hello,
>
>
> There are some interesting points made in this thread.
> I hope that everyone will not get too hot so that something good
> could happen from this.
>
> In a way, I agree with Markus. Not on all points but on the main
> points, I agree.
>
> I am also an experienced software architect so I also think that the
> entry ticket may be high for some people. I am not working for a
> software editor but for a manufacturing company.
> The discussions are often very technical (could be expected) and
> sometimes I have the impression that the functionalities are a bit
> forgotten behind the technical side.
>
> JEE specifications exist so that higher level functionalities can be
> built atop all the core functionalities defined in those specs.
> Unfortunately those higher level functionalities are developed
> leading to constraints that may not penalize a small company with a
> few adaministrators or developers. But it is very painful to deploy
> GlassFish when you deal with 1000s projects / 1000s developers /
> 100s administrators, ... .
> An example: we can not use LDAP authentication because it requires a
> asadmin group.

In a prior life as a consultant, I ran into this limitation. Now as
product manager, I am working with engineering to address this
particular issue, although in GlassFish v3. Based on prior (and
ongoing) conversations, I do understand your particular environment.

>
> We have a LDAP, a very huge LDAP ... so we can not define one
> asadmin group or as many asadmin subgroup as would be required. Tt
> is difficult to understand that this kind of constraints still
> exist ... But it exists and we are trying to deploy GlassFish
> despite that maybe the company I work for may be just too big for
> GlassFish ...
> On the other side, its size may not make it representative of the
> average company using GlassFish. But there may be a lot of companies
> that would be blocked by such a constraint.
>
> Some other expected higher level functionalities are cumbersome to
> use at the present time. (expected from an Enterprise level server)

Please keep the feedback coming. It's not hitting the dead queue :-)

>
> And GlassFish may end being used only as a .... web server ...
> (Servlet + JSP)
>
> Well implementing JEE specs is one thing, providing those
> functionalities is another one.
> All this to say that the community may well contribute positively
> with the specification of those higher level functionalities so that
> they can be used by as many people as possible by providing relevant
> feedbacks.
> And well, we can also say there is some chance that the
> implementation of those higher level functionalities may have
> impacts on the implementation of the core functionalities ...
> For contributing with code, it is more difficult. It depends on
> every company policy.
>
> The other point could be that my company is interested in a stable
> and mature application server for production usage. So JEE6 and
> GlassFish v3 may be out ot question for several years.
>
> So what about the engineering meetings with JEE5 and GF v2.1 ?
>

We are happy to have these discussions, but keep in mind that some
features require quite a bit of product modifications (such as Role
Based Access Control) since it is a cross-cutting concern. This may
impact the very stability you are looking for. As a result, we tend to
reserve some of these features for major product releases.

I would like to say THANK YOU (again) for all of the feedback you have
been giving.

>
> My 25 cents ... I hope ...
>
>
> Bye
> [Message sent by forum member 'vkoniecz' (vkoniecz)]
>
> http://forums.java.net/jive/thread.jspa?messageID=357253
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

vkoniecz
Offline
Joined: 2008-11-14

Hello John

Good to hear from you too !

Just to clarify.

I have said that the company I work for is interested in a stable and mature application server for production usage.
But I did not say that we will not be able to take some risks on stability or maturity if a very interesting feature has been integrated.
We would gain so much from the removal of the asadmin constraints or from the integration of Role Based Access Control in the administration console that we would probably be in position to take some risks (as long as we can keep them under control).
Unfortunately the risks do not include GF v3 that we can not even consider until it exists.
It is just about the balance between costs and gains for my company, as well as for Sun when making its decisions.

Those are indeed very specific needs expressed by a customer that paid a subscription.
So this mainly interests Sun as a software editor and there we are on a commercial discussion basis.
We are talking about support in that case : we will expect from Sun not to be blocked by every P1 bug that still exists or
any feature too difficult to implement.
(With the number of applications under development here, the probability that one application is impacted by a P1 bug is high !!)

Nonetheless, just take the point from an open source community point of view.
This allows to consider options or possibilities that Sun employees may not have thought ...
This is also a good way to consolidate GlassFish assets and increase its potential.
A community based discussion is much more about communication, durability, credibility and quality.
I am still not even talking about making decisions or workflows.
Good ideas or relevant needs do not need heavy decision process to be taken into account.
I talked with you before about customers groups organized by software editors around their own technologies.
Those groups are interesting for two main reasons:
1) it allows the customers to discuss about their achievements and their needs, and to share these with other customers
2) it allows the editors to learn about its customers needs, and to introduce its own ideas for collecting their opinions
Just replace customers by users and editors by contributors, if you prefer.
This may not be the right thing to do with an open source community, maybe more interesting for Sun for its own customers.
Nevertheless I think that GlassFish needs something to achieve the same objectives: JEE vX implementation may not be an achievement in itself.
This is not what finally will make GlassFish used or subscriptions paid. For early adopters maybe, but I am convinced that these really are a minority.
I completely understand the need for Sun to complete a "GlassFish v3, JEE6 reference implementation".
I would just put the stress on the fact that GlassFish v2.x, JEE5 reference implementation, will probably be of interest for a lot of people for a large period of time.
(At least for GF v3 or JEE6 to be very well tested)

Anyway I hope we could keep discussing about GlassFish for a long time.

Markus Karg

> "Welcome to GlassFish v3 Preview, a highly productive platform for
> building enterprise applications with dynamic languages and early
> access to Java EE 6 technologies. It is an early access, preview
> release of Java EE 6 technologies with community-only support. "

Typically people are scared by lots of text and only search for the download button. Maybe it would be good if we reduce the amount of text on the page and replace it by a really fat, agressively coloured and highly visible warning like "This is an unfinished product which is still under heavy construction. It is likely to fail. Do not use it in production use. There is no commercial support.". And alle the other information found behind explicit links. I think that is foolproof.

Remember, if you once think you met the biggest idiot of the world, that every other morning an even bigger idiot will get up.

Regards
Markus

Markus Karg

While I still am not a big friend of the agressive style of conversion you apply, I must confess that I fear that you are right. ;-( This is what I try to change, but it is hard as Oracle/Sun certainly do not do this project for fun, but for money.

> -----Original Message-----
> From: glassfish@javadesktop.org [mailto:glassfish@javadesktop.org]
> Sent: Mittwoch, 22. Juli 2009 20:07
> To: users@glassfish.dev.java.net
> Subject: Re: GLASSFISH IS LAME
>
> Sure cheeser. I only contribute to promising projects. You and your
> lame project will outphase itself.
>
> cheese
> [Message sent by forum member 'gniontas' (gniontas)]
>
> http://forums.java.net/jive/thread.jspa?messageID=357034
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net

Kristian Rink

Markus Karg schrieb:
> While I still am not a big friend of the agressive style of conversion
> you apply, I must confess that I fear that you are right. ;-( This is
> what I try to change, but it is hard as Oracle/Sun certainly do not do
> this project for fun, but for money.

Well... I guess most of these companies involved in open source projects do
right this "for money", be that directly or indirectly. One thing which is
different about the Sun open source projects, here, in my opinion is that,
ultimately, the source code of these projects is open and available but at
the end it seems to be nothing more than "open source Sun repositories" - no
real developer community outside Sun (feel free to correct me if I'm all
wrong on that), no real community in terms of making decisions regarding
project development and management, and so forth. This, ultimately, makes
people suspicious as a lot of them (including us, actually) seem / tend to
rely upon open source software also for the reason of not being fully
dependent upon one special vendor providing this software (a developer
friend of mine just recently had to experience this as Oracle decided to
discontinue his beloved OAS in favor of WebLogic). So even being "open
source", Glassfish won't eventually help preventing such a scenario. :(

K.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

> Markus Karg schrieb:
> > While I still am not a big friend of the agressive style of
> conversion
> > you apply, I must confess that I fear that you are right. ;-( This is
> > what I try to change, but it is hard as Oracle/Sun certainly do not
> do
> > this project for fun, but for money.
>
> Well... I guess most of these companies involved in open source
> projects do
> right this "for money", be that directly or indirectly. One thing which
> is
> different about the Sun open source projects, here, in my opinion is
> that,
> ultimately, the source code of these projects is open and available but
> at
> the end it seems to be nothing more than "open source Sun repositories"
> - no
> real developer community outside Sun (feel free to correct me if I'm
> all
> wrong on that), no real community in terms of making decisions
> regarding
> project development and management, and so forth. This, ultimately,
> makes
> people suspicious as a lot of them (including us, actually) seem / tend
> to
> rely upon open source software also for the reason of not being fully
> dependent upon one special vendor providing this software (a developer
> friend of mine just recently had to experience this as Oracle decided
> to
> discontinue his beloved OAS in favor of WebLogic). So even being "open
> source", Glassfish won't eventually help preventing such a scenario. :(

Just what I say: The source is open, but the project is not -- but these days people expect open processes when reading about open source. Actually there is such a community outside of Sun which would be happy to participate in the decision making, but Sun is not offering an open process to let them participate. This is a difference.

Regards
Markus

Paul Sandoz

On Jul 23, 2009, at 3:03 PM, Markus Karg wrote:
> Just what I say: The source is open, but the project is not -- but
> these days people expect open processes when reading about open
> source. Actually there is such a community outside of Sun which
> would be happy to participate in the decision making, but Sun is not
> offering an open process to let them participate. This is a
> difference.
>

Have you attending a GF v3 engineering meeting?

http://glassfish.markmail.org/search/?q=%09V3+Engineering
+meeting#query:V3%20Engineering%20meeting%20order%3Adate-backward+page:
1+mid:z2bt7tcb2547g374+state:results

Paul.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Martin Gainty

that link was the July 21 meeting
would you have a link for the date of the next meeting?

Thanks,
Martin Gainty
______________________________________________
Verzicht und Vertraulichkeitanmerkung/Note de déni et de confidentialité

Diese Nachricht ist vertraulich. Sollten Sie nicht der vorgesehene Empfaenger sein, so bitten wir hoeflich um eine Mitteilung. Jede unbefugte Weiterleitung oder Fertigung einer Kopie ist unzulaessig. Diese Nachricht dient lediglich dem Austausch von Informationen und entfaltet keine rechtliche Bindungswirkung. Aufgrund der leichten Manipulierbarkeit von E-Mails koennen wir keine Haftung fuer den Inhalt uebernehmen.
Ce message est confidentiel et peut être privilégié. Si vous n'êtes pas le destinataire prévu, nous te demandons avec bonté que pour satisfaire informez l'expéditeur. N'importe quelle diffusion non autorisée ou la copie de ceci est interdite. Ce message sert à l'information seulement et n'aura pas n'importe quel effet légalement obligatoire. Étant donné que les email peuvent facilement être sujets à la manipulation, nous ne pouvons accepter aucune responsabilité pour le contenu fourni.

> Date: Thu, 23 Jul 2009 15:16:05 +0200
> From: Paul.Sandoz@Sun.COM
> To: users@glassfish.dev.java.net
> Subject: Attending GF v3 engineering meetings?
>
> On Jul 23, 2009, at 3:03 PM, Markus Karg wrote:
> > Just what I say: The source is open, but the project is not -- but
> > these days people expect open processes when reading about open
> > source. Actually there is such a community outside of Sun which
> > would be happy to participate in the decision making, but Sun is not
> > offering an open process to let them participate. This is a
> > difference.
> >
>
> Have you attending a GF v3 engineering meeting?
>
> http://glassfish.markmail.org/search/?q=%09V3+Engineering
> +meeting#query:V3%20Engineering%20meeting%20order%3Adate-backward+page:
> 1+mid:z2bt7tcb2547g374+state:results
>
> Paul.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net
>

_________________________________________________________________
Windows Live™ Hotmail®: Search, add, and share the web’s latest sports videos. Check it out.
http://www.windowslive.com/Online/Hotmail/Campaign/QuickAdd?ocid=TXT_TAG...
[att1.html]

Markus Karg

> Do you have any idea what open source stands for?
> I am a long time believer in open source. And I have nothing against
> paying for support on open source products that actually show promise.

I think nobody has a problem with paying good money for good support. But with GlassFish's P1 bugs it is different. Those bug reports have been investigated in long and expensive research by the users (they did not get money for that from Sun or anybody else), and it should be fair that those users get the bug fix for free, too. I do not see bug fixing as a "support".

Regards
Markus

Eduardo Pelegri-Llopart

Hi Markus.

You and I exchanged emails on this general topic a week ago but others
may need additional background to understand your perspective on bug
fixing and support, so, trying to capture that in a couple of paragraphs...

Markus' context is that there are laws in Germany that require consumer
products to be free of defects. He interprets that to say that a
customer of a software product must be given free fixes for all
"critical" bugs in the product. Sun includes that in our commercial,
for-fee, GlassFish offering, but Markus says it should apply to the free
distribution also. Sun commits to push all commercial bug fixes to the
next free/public distribution, but the isolated, sustaining bug fixes
are not available _separately_ before that.

It is a bit more complicated than that, and IANAL, etc, but I think that
gives some background.

There is also a separate topic about what is a P1; I think that needs to
be addressed better.

From our perspective, much of this boils down to different ways to
balance the needs of Sun, who is contributing many people to deliver
GlassFish, with the needs of non-paying users, paying customers, and
developers and partners.

There are other ways of balancing things. RedHat, Covalent, MySQL,
JasperSoft, etc, etc, they all use different mechanisms to balance the
needs of all these stakeholders.

So far, I think most people in the GlassFish community have been happy
with the tradeoffs we have been following: we have been delivering free
good quality releases in a regular fashion, while building the next
generation v3 product and providing "support" to users and customers.
The last few months have been a bit extra tricky with summer, Oracle,
JavaOne, etc, but hopefully we are getting back to our normal cycles.

Hope this helps,
- eduard/o

Markus Karg wrote:
>> Do you have any idea what open source stands for?
>> I am a long time believer in open source. And I have nothing against
>> paying for support on open source products that actually show promise.
>
> I think nobody has a problem with paying good money for good support. But with GlassFish's P1 bugs it is different. Those bug reports have been investigated in long and expensive research by the users (they did not get money for that from Sun or anybody else), and it should be fair that those users get the bug fix for free, too. I do not see bug fixing as a "support".
>
> Regards
> Markus
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

Eduardo,

thank you for clarifying the situation and explaining my position (which
you excellently summarised).

As I am doing open source for longer time than GlassFish exists, I think
it is not overbearing if I explain a bit the situation out of the view
of an open source campaigner.

Open Source campaigners have a vision of open source that has nothing to
do with source code. It has to do with democracy. For "us", open source
means an enterprise (not necessarily a software project), best described
by the following "rules":

* Open Source is about community value, not about company turnover.

(According to that rule, many contributors spend lots of valueable time
and money into their projects without getting anything back but just the
good feeling of having done a great work.)

(According to that rule, it is Sun's problem that they like or need to
finance their contributions in some way. The other contributors also
must eat every day, but they do not necessarily buy bread from working
on GlassFish. Resulting is the idea that Sun must pay us to test drive
their contributions and bug reports *IF* we would have to pay for Sun to
fix those and promptly publish the fixes, as they also pay their
internal test team to do the same job. "We" also have to make an income
somehow, and our time is also worth getting paid. We are no hobbyists.)

* A contribution is a donation and belongs to all contributors.

(According to that rule, as soon as Sun publishes code, it is no more
Sun's code and alls contributors have the same rights and
responsibilities. GlassFish is no more Sun's project or code, it is the
project and code of the community, as a contribution is a donation. A
donation is not a donation if you expect to get anything back for it.).

* All stakeholders are equal. No one has more rights than others. All
role owners are democratically elected.

(According to that rule, many open source projects have public
discussion forums for even the most technical details and do elections
of all project roles. From this point of view it is not acceptable that
Sun dictates the project leaders. Often "open source companies" hire
people that already are elected, instead of placing their current
employees into the role since they think it is "their" product, while
actually it already was donated to the community. This does not mean
that I think that anybody does a bad job at Sun, it is just a
description of the point of view.)

* All decisions are public and found by consensus.

(According to that rule, any architectural or design question has to be
discussed in the mailing list and is not to be decided by a single
stakeholder. Explicitly it is not Oracle's sole decision whether to
change EclipseLink's architecture or not, as long as there are
contributors willing to provide this).

* The Global Village

(According to that rule, management must be done in a way that allows
everybody to participate without overly stressing him. Due to that,
there often are no meetings and schedules, but agile work plans and the
using of public discussion forums and voting systems. Yes, this means
that many things are more expensive to reach and need longer time to
market. But open source is not about making money, it is about a
community reaching a target.)

I hope that this view makes it more clear why people try to get more
influence on the project management. It is not about being displeased
with Sun's people, but with Sun's and Oracle's policy of "being the boss
by definition" from a more fundamental position of open source in
general. It is not about starting a revolution and kicking out anybody
of his role or position, it is about getting ASKED about our opinion to
a lot of architectural, technical and management questions (including
things like marketing, release schedules, when to publish bug fixes, and
so on). One example: From my point of view, Sun's decision to delay
publication of bug fixes to a months later release, is incompatible with
the idea (not the pure text) of the GPL. The discussion about this
should be not inside of Sun, but inside of the community, since the bug
fix is basing on the community's (not Sun's) code (Sun did donate it,
from the view of open source people, so they gave up right of
ownership); as a result, it is the community's decision when a derived
work (which is a bug fix) is to be published (since it is the community
that provides the licence to create such a fix basing on the GlassFish
code, it is not Sun). I think you got the point.

>From this view the problems arise, and there are people coming to
GlassFish that live in THAT world (which might be strange and foreign to
Sun and Oracle). Those people, including "the idiot", see that Sun
announces "open source" but not provide what the above axioms do
pretent. I know that Sun is working hard to make it happen. But that is
not the point. The point is, the fact that it is not yet reached, makes
the open source campaigners angry, since it looks like "stealing" the
"brand" of open source for a commercial use. An some of them then are
not willing to write explanations like this one (explaining mostly while
I am advocating "the idiot"), they just write "F* You All", just as
Greenpeave writes a slogan on a nuclear power plant like "You will kill
us all" and gets called "Eco Terrorists" for it.

Regards
Markus

> -----Original Message-----
> From: Eduardo.Pelegrillopart@sun.com
> [mailto:Eduardo.Pelegrillopart@sun.com] On Behalf Of Eduardo Pelegri-
> Llopart
> Sent: Donnerstag, 23. Juli 2009 20:01
> To: users@glassfish.dev.java.net
> Subject: Bug Fixing, Support, German Laws, etc... (was: GLASSFISH IS
> LAME)
>
> Hi Markus.
>
> You and I exchanged emails on this general topic a week ago but others
> may need additional background to understand your perspective on bug
> fixing and support, so, trying to capture that in a couple of
> paragraphs...
>
> Markus' context is that there are laws in Germany that require
consumer
> products to be free of defects. He interprets that to say that a
> customer of a software product must be given free fixes for all
> "critical" bugs in the product. Sun includes that in our commercial,
> for-fee, GlassFish offering, but Markus says it should apply to the
> free
> distribution also. Sun commits to push all commercial bug fixes to
the
> next free/public distribution, but the isolated, sustaining bug fixes
> are not available _separately_ before that.
>
> It is a bit more complicated than that, and IANAL, etc, but I think
> that
> gives some background.
>
> There is also a separate topic about what is a P1; I think that needs
> to
> be addressed better.
>
> From our perspective, much of this boils down to different ways to
> balance the needs of Sun, who is contributing many people to deliver
> GlassFish, with the needs of non-paying users, paying customers, and
> developers and partners.
>
> There are other ways of balancing things. RedHat, Covalent, MySQL,
> JasperSoft, etc, etc, they all use different mechanisms to balance the
> needs of all these stakeholders.
>
> So far, I think most people in the GlassFish community have been happy
> with the tradeoffs we have been following: we have been delivering
free
> good quality releases in a regular fashion, while building the next
> generation v3 product and providing "support" to users and customers.
> The last few months have been a bit extra tricky with summer, Oracle,
> JavaOne, etc, but hopefully we are getting back to our normal cycles.
>
> Hope this helps,
> - eduard/o
>
>
> Markus Karg wrote:
> >> Do you have any idea what open source stands for?
> >> I am a long time believer in open source. And I have nothing
against
> >> paying for support on open source products that actually show
> promise.
> >
> > I think nobody has a problem with paying good money for good
support.
> But with GlassFish's P1 bugs it is different. Those bug reports have
> been investigated in long and expensive research by the users (they
did
> not get money for that from Sun or anybody else), and it should be
fair
> that those users get the bug fix for free, too. I do not see bug
fixing
> as a "support".
> >
> > Regards
> > Markus
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Mark Mielke

On 07/22/2009 01:37 PM, glassfish@javadesktop.org wrote:
> Do you have any idea what open source stands for?
>

I have to wonder if you do. :-)

Open source means very little - except that the software provider gives
you access to their source code. It doesn't tell us whether the software
costs money to license or support or not. It doesn't say what quality
the software is. It doesn't say what level of community contributions
there are, or what level of corporate sponsorship is provided. None of
this is related to "open source". Glassfish is "open source", as the
source code is provided to the public. It's as simple as that, and as
irrelevant as that. Great, it is open source - so what?

This is where we get to other expectations:

> I am a long time believer in open source. And I have nothing against paying for support on open source products that actually show promise.
>
> But my problem is with projects like glassfish who call themselves open source but really is a sluggish community bringing bad name to all the open source projects.

"Sluggish community". This comes to expectations and investment. The
community is you and me. A sluggish community means contributions are
low from the entire community. Sun does not provide a community - Sun
provides sponsorship. It seems that you and others believe that Sun's
sponsorship level seems especially low of late. I tend to agree. But,
where is the agreement that says that Sun will provide the sponsorship
level you expect throughout, even while it is busy in the "due
diligence" process of being acquired by Oracle? This comes to your
expectations and your investment. What have you done for this community?
What investment have you made to prevent this community from being
"sluggish?" If the community is sluggish - it means you and I are sluggish.

Does Sun have a sponsorship responsibility to provide leadership in this
community? Sure. If they want to claim that this is an activity
community, then they should be enabling the community to be active. They
should be removing obstacles that prevent people from contributing. I've
seen some evidence that they acknowledge this. But - the contributions
still come from you and me. Once they've removed all obstacles - it is
you and me that would be providing "community" contributions.

> Another victim here is JAVA itself. JAVA was designed to be simple and platform independent. You guys have made JAVA based systems so complex and cumbersome that no new developer wants to use it anymore.
>

This is highly off topic and highly subjective. Yes, Java is complex and
cumbersome - but it solves complex and cumbersome problems. "You guys" =
SUN and the JCP, which has representation from several major companies.
I do not agree with your conclusion - as I think Java EE 1.4 and earlier
was much harder to use than Java EE 5, and upcoming Java EE 6. I think
things are improving significantly, albeit slowly. Where I dismissed
Java EE 1.4 as a platform I wanted to be based on before due to its
complexity, I as a "new developer", am looking to take advantage of Java
EE 5, and the upcoming Java EE 6.

> Yes, we need a better system than JBOSS. And that's why I have been looking at Glassfish. The least I expect from the installer is that it works as per the QUICK START GUIDE. The f***ing quick start guide tells me to go to http://localhost:4848 for the admin console. As it turns out there is no admin console. I then post a question on the forum. The forum sends me on a wild goose chase saying try installer X and then installer Y. I try about 5 installers and one of them from months ago works. So, I decide to look at the help page and see how the admin console works.... hmmm.. no help page. I go back to the forum and ask how to make the help page work.......SILENCE.
>

Does Glassfish v2.1 work for you? Does the "supported" Sun App Server
9.1(?) based on Glassfish v3 work for you? Or is the only image you
tried a latest preview build? Did you notice the "preview" stamps all
over the place?

Glassfish v3 preview has worked for me, but I know that Glassfish v3 is
not yet stable, and I have chosen to stay with Glassfish v2.1.

I think I recall one Glassfish v3 preview build failing to install for
me - and I could not figure it out. I rebooted my machine (coincidence)
and then it started to work. This lead me to the conclusion that it was
something in my system (not Glassfish) that was causing the problem.

For people trying to offer you help, but not having access to your
system, and sending you on a "wild goose chase" - I have to assume you
have not been part of an open source community before if you find this
surprising. It happens all the time. It is what happens when you have
varied level of experience and skill community members trying to help
out. Sometimes they get it wrong.

> So, after 2 days of banging my head on my desk, I decide to send out some blasting emails. At this point, it's not a matter of whether I can fix the problem myself and get on with my project. At this point, the question is whether I want to spend anymore days frustrating over this product and a sluggish community. The question is whether I trust y job and the future of my company over this product a product like this. The question is whether I can save other people's time and money by writing some real reviews about Glassfish.
>
> I think I will do more service to open source community by exposing glassfish community than by helping you guys fix the problem.

I suggest you start with the officially released version of Glassfish if
you want to provide a real review of the product.

If you review is going to be based off a development milestone, your
review is likely to be poor, and this is to be expected.

Cheers,
mark

--
Mark Mielke

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

> > Do you have any idea what open source stands for?
> I have to wonder if you do. :-)
> Open source means very little - except that the software provider gives
> you access to their source code. It doesn't tell us whether the
> software
> costs money to license or support or not. It doesn't say what quality
> the software is. It doesn't say what level of community contributions
> there are, or what level of corporate sponsorship is provided. None of
> this is related to "open source". Glassfish is "open source", as the
> source code is provided to the public. It's as simple as that, and as
> irrelevant as that. Great, it is open source - so what?

This is true, but the overall experience with "NON-LAME" Open Source projects like Apache, MySQL, Linux, KDE etc. is that all those assumptions *there* are true. So people expect Sun's Open Source projects to fulfill the same assumptions, just due to the big stakeholder backing it, the size of the community, and the well known project name.

Certainly you are right from the pure word "open source", but the "idiot" is also right from his overall experience with other big open source projects. So we (the community) should understand what people expect, and try to fulfil it, or make clear, that in GlassFish things are a bit different (so nobody later gets disappointed after downloading).

Regards
Markus

cdonaldson
Offline
Joined: 2007-11-13

Hi Markus,

Sun branding, like many corporate brandings, can be confusing. In this case I am not sure how we can improve. The GFv3 page:

https://glassfish.dev.java.net/downloads/v3-preview.html

starts with:

"Welcome to GlassFish v3 Preview, a highly productive platform for building enterprise applications with dynamic languages and early access to Java EE 6 technologies. It is an early access, preview release of Java EE 6 technologies with community-only support. "

and ends with:

"How do I get help with GlassFish v3 Preview?

The community edition of GlassFish Server v3 Preview is supported by the community. Commercial support is not available. Contact the GlassFish user mailing list or the GlassFish Forum with any issues, comments, or questions. Also, see the GlassFish Community wiki documentation. The GlassFish Community is standing by to help."

The Sun GF pages:

http://www.sun.com/software/products/appsrvr/get_it.jsp

seem to demonstrate a difference between GFv2:

"Sun GlassFish Enterprise Server v2.1 is a high-performance, enterprise-ready Java EE 5 server with support for clustering and in-memory based high availability. It is free for development, deployment and re-distribution. Sun offers highly competitive support subscription pricing for lower deployment costs. "

and GFv3:

"An early access release of GlassFish v3, featuring the enhanced productivity of Java EE 6 and the Java EE 6 Web Profile. Additional enhancements include a microkernel architecture based on OSGi for improved startup time and reduced resource utilization, and fine-grained monitoring and DTrace probes for improved observability. Developers should consider the Java EE 6 SDK Preview."

The GFv3 release notes also continually mention targeting developers.

The Sun Java EE pages:

java.sun.com/javaee/downloads/index.jsp

also delineate between the "Latest Release" (GFv2) and "Next Release (Preview)".

Standard and Premium Subscriptions do include support for GlassFish Enterprise Server v3 Prelude (when downloaded as the Sun branded product vs the GF community release), but I think we've done a reasonable job demonstrating that this is targeted more for developers to begin using JEE6 features than for Enterprise production environments. There are use cases for deploying prelude in production, but the caveats of JEE6 not being finalized and this being a preview release seem to be in place. The prevailing Sun philosophy has been to get early access releases to developers for years (they used to be referred to, in appserver world, as the "Platform Edition"), and this is another instance of that.

Now, this may be more apparent to me simply because I've waded through this many times, and it is quite possible that I

1.) have blinders on and/or
2.) just know where to look to find the above links.

Should we be altering the:

https://glassfish.dev.java.net/downloads/v3-preview.html

page further? Are links leading users down the wrong path? Can you specify pain/confusion points?

Regards,
Chris

Mark Mielke

On 07/22/2009 03:29 PM, glassfish@javadesktop.org wrote:
> Now, this may be more apparent to me simply because I've waded through this many times, and it is quite possible that I
>
> 1.) have blinders on and/or
> 2.) just know where to look to find the above links.
>
> Should we be altering the:
>
> https://glassfish.dev.java.net/downloads/v3-preview.html
>
> page further? Are links leading users down the wrong path? Can you specify pain/confusion points?
>

You have done a fine job of differentiating it from my perspective. I
never had the confusion that the original poster has had, and I would
consider myself a "new user" still. Thanks for putting in the effort to
illustrate how well the releases are differentiated.

My only complaint with Sun is how slow things move, and especially of
late. I'd like Java EE 6 to be out as a production quality release
tomorrow. :-)

Cheers,
mark

--
Mark Mielke

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

cdonaldson
Offline
Joined: 2007-11-13

Hi Mark,

Well, I wish I had the sway to make things a bit more agile. I do believe that is one of the long term intents with v3; by making it modular, it should encourage both faster updates in more specific areas, as well as an easier approach for people outside of Sun to contribute (coming full circle to this thread's discussion of "community"). We're eagerly working on and looking forward to v3's enterprise release, despite any publicity related slowdowns.

That said, I am probably going to stop responding to this particular thread and let "GLASSFISH IS LAME" start a (likely) slow death. :)

Regards,
Chris

Mark Mielke

On 07/22/2009 08:44 AM, Markus Karg wrote:
>
> Mark,
>
> as I said, I dispraise his wording. But I take his concerns serious,
> because Sun actually is selling v3 support already and for newcomers
> it looks like it is a finished product. There should be a big fat
> warning at the download site that v3 is only a preview. BTW, I do not
> agree that anybody shall go away and use JBoss, because actually is
> what is happening at the moment, and what is weakening the community.
> Instead, we (the community) must work hard to convince those
> unsatisfied people that GlassFish is great, and we must convince
> Sun/Oracle to invest into the product so it actually becomes that
> great product that it is already in our minds.
>
>

I'd rather have thoughtful, intelligent, contributing members in any
community I am part of.

The community is better off without people the like of the original
poster. His childish contribution causes more damage than good, no
matter how forgiving, sympathetic and considerate you are to true
underlying problems. If he has a legitimate problem, he can present it
as a professional. As it is, I feel no desire to aid him or justify him.

I take your concerns serious, and if I could help you, I would be right
on it.

Cheers,
mark

--
Mark Mielke

[att1.html]

Markus Karg

Mark,

maye he just freaked out due to the severe problems and time pressure, who knows. For me, I think a community is better the bigger it is. And as bigger a community is, as more idiots you will find, statistically spoken. Since there had been no idiots before, I welcome one solely to the sake of statistics. ;-)

Regards

Markus

From: Mark Mielke [mailto:mark@mark.mielke.cc]
Sent: Mittwoch, 22. Juli 2009 19:47
To: users@glassfish.dev.java.net
Cc: Eduardo Pelegri-Llopart
Subject: Re: GLASSFISH IS LAME

On 07/22/2009 08:44 AM, Markus Karg wrote:

Mark,

as I said, I dispraise his wording. But I take his concerns serious, because Sun actually is selling v3 support already and for newcomers it looks like it is a finished product. There should be a big fat warning at the download site that v3 is only a preview. BTW, I do not agree that anybody shall go away and use JBoss, because actually is what is happening at the moment, and what is weakening the community. Instead, we (the community) must work hard to convince those unsatisfied people that GlassFish is great, and we must convince Sun/Oracle to invest into the product so it actually becomes that great product that it is already in our minds.

I'd rather have thoughtful, intelligent, contributing members in any community I am part of.

The community is better off without people the like of the original poster. His childish contribution causes more damage than good, no matter how forgiving, sympathetic and considerate you are to true underlying problems. If he has a legitimate problem, he can present it as a professional. As it is, I feel no desire to aid him or justify him.

I take your concerns serious, and if I could help you, I would be right on it.

Cheers,
mark

--
Mark Mielke
[att1.html]

gniontas
Offline
Joined: 2009-07-15

Do you have any idea what open source stands for?
I am a long time believer in open source. And I have nothing against paying for support on open source products that actually show promise.

But my problem is with projects like glassfish who call themselves open source but really is a sluggish community bringing bad name to all the open source projects. Another victim here is JAVA itself. JAVA was designed to be simple and platform independent. You guys have made JAVA based systems so complex and cumbersome that no new developer wants to use it anymore.

Yes, we need a better system than JBOSS. And that's why I have been looking at Glassfish. The least I expect from the installer is that it works as per the QUICK START GUIDE. The f***ing quick start guide tells me to go to http://localhost:4848 for the admin console. As it turns out there is no admin console. I then post a question on the forum. The forum sends me on a wild goose chase saying try installer X and then installer Y. I try about 5 installers and one of them from months ago works. So, I decide to look at the help page and see how the admin console works.... hmmm.. no help page. I go back to the forum and ask how to make the help page work.......SILENCE.

So, after 2 days of banging my head on my desk, I decide to send out some blasting emails. At this point, it's not a matter of whether I can fix the problem myself and get on with my project. At this point, the question is whether I want to spend anymore days frustrating over this product and a sluggish community. The question is whether I trust y job and the future of my company over this product a product like this. The question is whether I can save other people's time and money by writing some real reviews about Glassfish.

I think I will do more service to open source community by exposing glassfish community than by helping you guys fix the problem.

GOODBYE GLASSFISH.

cheeser
Offline
Joined: 2004-08-02

Yes. But you clearly don't. Open source means that anyone can contribute and can fix bugs. So if you're seeing a bug and not either 1) paying someone to fix it for you or 2) doing it yourself, you can climb down off that high horse. Open source means putting up or shutting up and you're not currently contributing so I guess that narrows your options. Your abusive, atagonistic attitude will get you exactly nowhere. So you may say goodbye glassfish but i'm sure the rest of the community will have no problems saying "goodbye, troll." I'm sorry things aren't working out for you like they are the countless other working systems. I'm sure there's some quantum/metaphysical anomaly in your server and you're certainly not to blame for *that* but your attitude is entirely unacceptable. So good luck with whatever you choose. I'm sure we won't miss you here.

> Do you have any idea what open source stands for?
> I am a long time believer in open source. And I have
> nothing against paying for support on open source
> products that actually show promise.
>
> But my problem is with projects like glassfish who
> call themselves open source but really is a sluggish
> community bringing bad name to all the open source
> projects. Another victim here is JAVA itself. JAVA
> was designed to be simple and platform independent.
> You guys have made JAVA based systems so complex and
> cumbersome that no new developer wants to use it
> anymore.
>
> Yes, we need a better system than JBOSS. And that's
> why I have been looking at Glassfish. The least I
> expect from the installer is that it works as per the
> QUICK START GUIDE. The f***ing quick start guide
> tells me to go to http://localhost:4848 for the admin
> console. As it turns out there is no admin console. I
> then post a question on the forum. The forum sends me
> on a wild goose chase saying try installer X and then
> installer Y. I try about 5 installers and one of them
> from months ago works. So, I decide to look at the
> help page and see how the admin console works....
> hmmm.. no help page. I go back to the forum and ask
> how to make the help page work.......SILENCE.
>
> So, after 2 days of banging my head on my desk, I
> decide to send out some blasting emails. At this
> point, it's not a matter of whether I can fix the
> problem myself and get on with my project. At this
> point, the question is whether I want to spend
> anymore days frustrating over this product and a
> sluggish community. The question is whether I trust
> y job and the future of my company over this product
> a product like this. The question is whether I can
> save other people's time and money by writing some
> real reviews about Glassfish.
>
> I think I will do more service to open source
> community by exposing glassfish community than by
> helping you guys fix the problem.
>
>
> GOODBYE GLASSFISH.

gniontas
Offline
Joined: 2009-07-15

Sure cheeser. I only contribute to promising projects. You and your lame project will outphase itself.

cheese

cheeser
Offline
Joined: 2004-08-02

Those lucky projects!

> Sure cheeser. I only contribute to promising
> projects. You and your lame project will outphase
> itself.
>
> cheese

Glenn Holmer

On Tue, 2009-07-21 at 19:58 -0700, glassfish@javadesktop.org wrote:
> See here why!

Drop the attitude.

--
____________________________________________________________
Glenn Holmer gholmer@weycogroup.com
Software Engineer phone: 414-908-1809
Weyco Group, Inc. fax: 414-908-1601

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
For additional commands, e-mail: users-help@glassfish.dev.java.net

Markus Karg

Eduardo,

while I absolutely dispraise the wording of this user, I must tell you that this case once more proofs that the current bug handling policy (not fixing bugs reported by the community in favour of doing things asked for by paying users; not fixing P1 bugs for months; not publishing bug fixed releases more often and separate from feature releases; etc.) is just not working in the open source world, and in the end leads to a deep scratches in GlassFish's image (for both, the open and the commercial one). One more company that will not use GlassFish. One more company that will not recommend GlassFish to others. One more company that will never buy a support contract. One more company that most likely will never look at Sun's products (including hardware) anymore in near future. Your policy has negative effects. Please change it. Just as we discussed last week, and the week before... If Sun needs more money to fix more bugs or publish bug fixed releases more open, then provide "per incident cases" and "support subsciprions" for higher prices. You know that companies are willing to pay for real help, just like mine wants to, and you know the reasons why we have not yet signed a contract. Your move next.

If I were the community leader, my policy would be to fix any bug first before adding any new functionality. But I am not, so the community is waiting for Sun to act now. Please act.

To gniontas: I really understand your aggression and what you want to express, but I bet if you would have used a less aggressive wording, more people would have been interested in helping you. "Fuck you all" is not the kind of motivation that open source volunteers like to hear. I wonder if you really thout that you actually get any help using the F word in a request for help.

Regards
Markus

> -----Original Message-----
> From: glassfish@javadesktop.org [mailto:glassfish@javadesktop.org]
> Sent: Mittwoch, 22. Juli 2009 04:59
> To: users@glassfish.dev.java.net
> Subject: GLASSFISH IS LAME
>
> See here why!
>
> http://forums.java.net/jive/thread.jspa?threadID=64406&tstart=0
> [Message sent by forum member 'gniontas' (gniontas)]
>
> http://forums.java.net/jive/thread.jspa?messageID=356890
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@glassfish.dev.java.net
> For additional commands, e-mail: users-help@glassfish.dev.java.net