hudson.java.net is alive

[Susan Duncan - Oracle]

I'm pleased to announce that http://hudson.java.net is alive and
running on the new infrastructure. The migration is not fully complete
(see details below) but the team has worked to get this far prior to
the beginning of the holiday in order to minimize the disruption

- The existing http://hudson.dev.java.net will be redirected to
http://hudson.java.net. The new page doesn't have any customizations
applied to it yet so what you will see is the default L&F. We'll be
addressing this in the next week or so.

- Only the source code repository and mailing lists are enabled. The
mailing lists only have project admin subscribed. The archives and
the subscriptions will be added in the next few days. The existing
mailing list aliases will be redirected to the new mailing list
archives

Until mailing is fully enabled you will not receive commit
notifications. However, you can browse the commits archive

- JIRA and downloads will be enabled as we continue through the
migration of these projects. The scheduled end date is 12/3

Please let us know if you have any questions or comments.

Happy Thanksgiving Holiday (for those of you in the USA)

Susan
Oracle Product Management

[Jason Dillon]

Almost every link under "Project Features" goes to a page stating The file <page> appears to be missing. To add content to your website, use webdavs.

"Project Links" at least go somewhere... otherwise the hudson.java.net site is mostly useless.

--jason

[Maxim Veksler]

I'm getting 

The file /web/projects/hudson/index.html appears to be missing.

To add content to your website, use webdavs.

[Susan Duncan - Oracle]

Seems that we had a gremlin over night - but all is well now. Bear in
mind that some areas (as specified in my original post) will not be
fully functional until next week. Our first aim was to get the
repository up again as soon as possible

[Mirko Friedenhagen]

Hello Susan,

I do not see any messages in the list archives except of one test message.

Regards

[Kohsuke Kawaguchi]

The archives are available at
http://hudson.361315.n4.nabble.com/Hudson-f361315.subapps.html

2010/11/25 Mirko Friedenhagen <mfried...@gmail.com>:

--
Kohsuke Kawaguchi

[Susan Duncan - Oracle]

Mailing is not fully migrated yet - that will come early next week

rgds

Susan
Oracle Product Management

[Jesse Farinacci]

Community,

Elephant in the room: Are we going to deprecate the java.net site and
mailing lists and all the other features? What services are still
utilized on java.net, now? I think none, as the "important" pieces of
our infrastructure are donated to the community by Atlassian, GitHub,
and Google Groups.

I can only see confusion for our users by having these competing resources.

-Jesse

--
There are 10 types of people in this world, those
that can read binary and those that can not.

[Andrew Bayer]

Given the overwhelming support for moving to GitHub, I think we're at the point where the java.net infrastructure doesn't play much of a role - it still powers the backend authentication for JIRA, and it's currently still our Maven repository, but I've done some preliminary work on switching authentication over to our own LDAP server, and setting up our own Maven repository. But before those changes can be made, we do need to resolve the current situation - the community consensus definitely is to move to GitHub, and the Google Groups mailing lists are functional and getting real traffic. I believe we should deprecate the java.net site/infrastructure and make that clear ASAP.

A.

[Andrew Bayer]

(adding the java.net mailing lists to make sure we hit all audiences with this)

[Nigel Magnay]

If you've got the h/w to self-host, that'd be the best route. It's not like it's the 1st time java.net has been unstable..

[Rob Petti]

I would also feel much better with the project drivers having more
control over
the infrastructure, since it seems like we were yanked offline without
any warning or
consultation. There was a plan in the works for dealing with the
migration, but Oracle
jumped the gun, it seems.

With git, at least we can continue work relatively unimpeded in the
unlikely event that
github goes down, and google groups seems like a vast improvement over
the old mailing
list.

I also vote for deprecating java.net.

On Nov 26, 11:34 am, Andrew Bayer <andrew.ba...@gmail.com> wrote:
> (adding the java.net mailing lists to make sure we hit all audiences with
> this)
>

[Kohsuke Kawaguchi]

And I really like the aspect of Git that enables other people to
contribute changes with very low overhead.
We are already seeing some of those happening in the context of
localization in just within the past few days.

For example, many contributers in the Japanese community hesitate to
ask for a commit access, for one reason or another,
but they can fork and push changes and send me e-mail all right.

2010/11/26 Rob Petti <rob....@gmail.com>:

--
Kohsuke Kawaguchi

[R. Tyler Croy]

On Fri, 26 Nov 2010, Nigel Magnay wrote:

> If you've got the h/w to self-host, that'd be the best route. It's not like
> it's the 1st time java.net has been unstable..

Just to throw this out there, we do have some hardware colocated with Contegix
[0] right now that hosts the hudson-labs infrastructure which has enough
headroom to host JIRA, Gerrit, Nexus, LDAP, etc.

The only real downside of it has been the cost (~$60/month for hosting, we own
the machine) which abayer and I have been footing.

[0] http://www.contegix.com

- R. Tyler Croy
--------------------------------------
GitHub: http://github.com/rtyler
Twitter: http://twitter.com/agentdero

[Kohsuke Kawaguchi]

The plan was to solicit a donation, and you can definitely count me in on that.

2010/11/26 R. Tyler Croy <ty...@monkeypox.org>:

--
Kohsuke Kawaguchi

[Jacob Robertson]

In regards to SVN I'm a little confused on either the timeline or the
layout of the infrastructure. This morning, get a I get a 302 for
both

https://hudson.dev.java.net/svn/hudson/trunk/hudson/plugins

and

https://hudson.java.net/svn/hudson/trunk/hudson/plugins

when trying to browse through Eclipse.

I have some code I've been wanting to commit for a number of days now... :(

[ted]

I think this thread is a bit out of control. Oracle has been talking
to a lot of the users of hudson who are not part of this very small
core group on the forums. Susan's job is to represent them until they
feel the urge to reply directly to the mailing lists or forums. One
of their biggest complaints is the lack of formal solicitation of
comments/requirements before changes like this are made, and a lot of
frustration around the inability to get their changed requests into
the core of hudson.

Oracle's goal is to grow the community and make hudson stronger. You
all might not be aware of this, but the actual hudson user base is
very large. Much bigger than what you see on the mailing lists or in
the forums. The unfortunate part of that is how many of these users
do not contribute to the core, and do not participate in these
discussions. They want to do that, but don't feel like they can be
heard. We want them to be heard. We need to make the hudson
community a place that will welcome all the hudson users and encourage
its growth and longevity. We will be announcing some changes in the
upcoming weeks that we believe will foster that.

For now, however, we are going to stay on the java.net
infrastructure. We believe it is important for hudson to stay
connected with the rest of the the java community, as well as take
advantage of some of the cool changes we will have coming to
java.net. Moving to GIT can be done while staying on java.net. It is
not a requirement to move to github. Another big complaint that we
have heard from many of the projects on java.net is that the old/
current java.net infrastructure has been unstable and cannot support
the types of things that some of the projects want to do. This move
onto the kenai infrastructure will allow us to add many more of those
capabilities and have this hosted by one of the biggest IT
organizations in the world who are use to guaranteeing up time and
reliability.

We can't stop some of you from writing the "Oracle is evil" posts, but
I think in the case of java.net, open source and especially hudson
that is an unfair and unsubstantiated characterization. We really
like hudson. We like it being free. We want the community of
contributors to grow to the hundreds. We want the community of users
to grow to the tens of thousands. And we would like it to have a
license where anyone can use it as they like. This is about trying to
create an environment where it can grow as a healthy open source
community.

Because it is open source, we can't stop anybody from forking it. We
do however own the trademark to the name so you cannot use the name
outside of the core community. We acquired that as part of Sun. We
hope that everyone working on hudson today will do as they claim to
want, and work with us to make hudson stronger. Susan, Winston or I
are always willing to have a constructive conversation about how to
make things better. We have done that with dozens of hudson users not
represented on this list, and you will start to see them taking a more
active part in the weeks and months ahead.

I am sorry for the inconvenience caused by our moving the hudson
project to the new java.net. We sent out an email to all of the
projects (including hudson) warning them of this change and all of
them except hudson had planned for it. As KK pointed out in his
email, he somehow missed the heads-up email our IT department sent
him. Our goal, as a community, should be working together to get
things back running normally on the new java.net infrastructure. We
are making good progress on that and Susan will be posted more
information about that later today.

thanks for your support and feedback and I hope it continues.

-ted

Ted Farrell

[ted]

I notice that my full profile was not added to my reply. A lot of you
know who I am, but for those who don't, I am Ted Farrell. I am the
chief architect for tools and middleware at oracle and lead the
development team supporting hudson that susan and winston are part
of. My email is ted.f...@oracle.com. thx.

[Andrew Bayer]

Ted et al -

I'll reply in more detail shortly, but I just wanted to ask why the obvious development community support and enthusiasm for moving to GitHub is not relevant - I understand that there are plenty of Hudson *users* not on the mailing lists, as one would expect, but Hudson's *developers* are on the list, so it's their opinions we've been hearing.

A.

[ted]

Hi Andrew. One of the features of the kenai software that is support
for git. I'd be interested in why we could not decide as a community
to move to git and stay on java.net in doing so. Is there something
specific that you would not be getting? If so, I think we can look at
adding any missing functionality to the platform.

-ted

On Nov 29, 9:32 am, Andrew Bayer <andrew.ba...@gmail.com> wrote:
> Ted et al -
>
> I'll reply in more detail shortly, but I just wanted to ask why the obvious
> development community support and enthusiasm for moving to GitHub is not
> relevant - I understand that there are plenty of Hudson *users* not on the
> mailing lists, as one would expect, but Hudson's *developers* are on the
> list, so it's their opinions we've been hearing.
>
> A.
>
> On Mon, Nov 29, 2010 at 9:28 AM, ted <ted...@gmail.com> wrote:
> > I notice that my full profile was not added to my reply.  A lot of you
> > know who I am, but for those who don't, I am Ted Farrell.  I am the
> > chief architect for tools and middleware at oracle and lead the
> > development team supporting hudson that susan and winston are part

> > of.  My email is ted.farr...@oracle.com.  thx.

[R. Tyler Croy]

On Mon, 29 Nov 2010, ted wrote:

> Hi Andrew. One of the features of the kenai software that is support
> for git. I'd be interested in why we could not decide as a community
> to move to git and stay on java.net in doing so. Is there something
> specific that you would not be getting? If so, I think we can look at
> adding any missing functionality to the platform.

Howdy Ted, one of the primary reasons for selecting GitHu instead of one of the
many Git hosts such as Gitorious (including Kenai) is the very low barrier to
entry for a lot of developers these days.

We had considered "self-hosting" the Git service but poo-pooed that idea in
favor of GitHub since having a GitHub account is almost as common as having a
twitter handle or gmail address.

Being publicly available/forkable on GitHub can be a huge win for a project
like Hudson both in publicity and accessibility, especially if the project
thrives on hundreds (if not thousands) of contributors to the core and plugins.

If Oracle's goal is to grow the community and make Hudson stronger, than I
would think you guys could get behind the development of Hudson moving to
GitHub and continue to support issues/wiki/www.hudson-ci.org (all Oracle
powered).

We're all here for the same reason (presumably), to make Hudson better. I
firmly believe that moving to GitHub will help us further improve Hudson beyond
what we've been capable of doing with SVN on Java.net

[Andrew Bayer]

Above and beyond the reasons for GitHub specifically (which rtyler has addressed - I should add that the social aspects of GitHub combined with the network effect of so much of the open source community being based on GitHub are what sell me), I'm really concerned that the Hudson development community is not being consulted or factoring into Oracle's decisions. We have no desire to fork, not in the least - Oracle has played an absolutely critical role in Hudson since day one, and I for one want to see them continue to do so. But I don't see how dictating to the developers who actually work on Hudson as to where their work goes, how it's managed, etc is the right way for that to remain the case.

A.

[Nigel Magnay]

Just having git support != git support on github. They work full time on providing the best community development tools; I doubt kenai could even catch up, let alone surpass what they're doing.

Because it is open source, we can't stop anybody from forking it.  We
do however own the trademark to the name so you cannot use the name
outside of the core community.  We acquired that as part of Sun.  We
hope that everyone working on hudson today will do as they claim to
want, and work with us to make hudson stronger.  

I'm confused.

What things are you saying you will not let the Hudson developer community do? 

I.E: Are you saying that, as the holders of the Hudson 'name', you are prohibiting the developer community from choosing (for ourselves) to migrate the infrastructure (bug tracking / wiki)? The repositories ?  

So far the response from the developers has been pretty strongly in favour of the migration to google groups for mail, to github for code repositories & collaboration, and to a self-hosted site for bug tracking and information.

Just so we know where the line is where we will have to force a vote for a fork with a new name.

[Jesse Farinacci]

Greetings,

I'm going to restrict my first response to matters of java.net
robustness and general maintenance.

On Mon, Nov 29, 2010 at 12:17 PM, ted <ted...@gmail.com> wrote:
>
> Another big complaint that we
> have heard from many of the projects on java.net is that the old/
> current java.net infrastructure has been unstable and cannot support
> the types of things that some of the projects want to do.

Ok, that's an important admission I'll keep in mind for the next paragraph.

> I think in the case of java.net, open source and especially hudson
> that is an unfair and unsubstantiated characterization.

It isn't unfair or unsubstantiated when you list it as a big complaint
yourself. I also don't think it is unsubstantiated when a company as
reputable as Sonatype starts the 'Java.net Maven Repository Rescue
Mission'[1] on March 5th of 2010.

[1] http://www.sonatype.com/people/2010/02/java-net-maven-repository-rescue-mission-on-march-5th/

> I am sorry for the inconvenience caused by our moving the hudson
> project to the new java.net.  We sent out an email to all of the
> projects (including hudson) warning them of this change and all of
> them except hudson had planned for it.

It doesn't appear that we were moved as expected[2]. Yes, the
community was warned, but to my knowledge we were told to be expecting
to be moved in November. While you state that KK may have missed an
email, frankly, it doesn't seem like a very good engagement of the
community.

[2] http://www.hudson-labs.org/content/javanet-migration-status-update

The past care and feeding of java.net is not something that inspires
any confidence in anyone that has used it. The current migration and
upkeep isn't doing much to change attitudes either..

[Andrew Bayer]

On Mon, Nov 29, 2010 at 10:16 AM, Nigel Magnay <nigel....@gmail.com> wrote:

So far the response from the developers has been pretty strongly in favour of the migration to google groups for mail, to github for code repositories & collaboration, and to a self-hosted site for bug tracking and information.

Actually, the intent is to keep using the existing issues.hudson-ci.org and wiki.hudson-ci.org, which Oracle generously host for us. I don't see any need to change that.

A. 

[jvanzyl]

At least for Apache and Eclipse the canonical repository for which all
IP/legal review has to be hosted on the organizations hardware.
Although I prefer github and do Maven work there (I, in fact, hate
working with SVN at this point) the canonical repository is at Apache
for the official Apache Maven releases. The same goes for Eclipse and
their projects. So this is not really any different for Hudson and
it's not overly surprising that the canonical repository for Hudson be
at Kenai with Oracle.

On Nov 29, 12:32 pm, Andrew Bayer <andrew.ba...@gmail.com> wrote:
> Ted et al -
>
> I'll reply in more detail shortly, but I just wanted to ask why the obvious
> development community support and enthusiasm for moving to GitHub is not
> relevant - I understand that there are plenty of Hudson *users* not on the
> mailing lists, as one would expect, but Hudson's *developers* are on the
> list, so it's their opinions we've been hearing.
>
> A.
>
>
>
> On Mon, Nov 29, 2010 at 9:28 AM, ted <ted...@gmail.com> wrote:
> > I notice that my full profile was not added to my reply.  A lot of you
> > know who I am, but for those who don't, I am Ted Farrell.  I am the
> > chief architect for tools and middleware at oracle and lead the
> > development team supporting hudson that susan and winston are part

> > of.  My email is ted.farr...@oracle.com.  thx.

[jvanzyl]

On Nov 29, 1:21 pm, Jesse Farinacci <jie...@gmail.com> wrote:
> Greetings,
>
> I'm going to restrict my first response to matters of java.net
> robustness and general maintenance.
>
> On Mon, Nov 29, 2010 at 12:17 PM, ted <ted...@gmail.com> wrote:
>
> > Another big complaint that we
> > have heard from many of the projects on java.net is that the old/
> > current java.net infrastructure has been unstable and cannot support
> > the types of things that some of the projects want to do.
>
> Ok, that's an important admission I'll keep in mind for the next paragraph.
>
> > I think in the case of java.net, open source and especially hudson
> > that is an unfair and unsubstantiated characterization.
>
> It isn't unfair or unsubstantiated when you list it as a big complaint
> yourself. I also don't think it is unsubstantiated when a company as
> reputable as Sonatype starts the 'Java.net Maven Repository Rescue
> Mission'[1] on March 5th of 2010.
>

> [1]http://www.sonatype.com/people/2010/02/java-net-maven-repository-resc...
>

This was, in all honestly, a problem at Sun. This has moved pretty
quickly within Oracle, and an agreement was made and work is in
progress to fix the problems with the repositories. The fix is not
simple but pretty soon after I posted that blog someone from Oracle
contacted me and the relationship has been constructive. To that end
Sonatype plans to work with Oracle on Hudson as well. Ted has been
straight forward with us and we've been getting good things done.

> > I am sorry for the inconvenience caused by our moving the hudson
> > project to the new java.net.  We sent out an email to all of the
> > projects (including hudson) warning them of this change and all of
> > them except hudson had planned for it.
>
> It doesn't appear that we were moved as expected[2]. Yes, the
> community was warned, but to my knowledge we were told to be expecting
> to be moved in November. While you state that KK may have missed an
> email, frankly, it doesn't seem like a very good engagement of the
> community.
>
> [2]http://www.hudson-labs.org/content/javanet-migration-status-update
>
> The past care and feeding of java.net is not something that inspires
> any confidence in anyone that has used it. The current migration and
> upkeep isn't doing much to change attitudes either..
>

With even the best of intentions from infrastructure folks shit
happens. SVN has gone out for a week at Apache, we've had snapshot
repos wiped out by mistake but you figure it out and move on. One of
the moves to Kenai is to remove a lot of the instability and general
lack of care and feeding the Java.net infrastructure had. The provider
is slowly being phased out which should result in something more
stable over time.

For the record I think the workflow which will help all communities
flourish with Git is something like the following:

http://www.sonatype.com/~jvanzyl/EmpoweringUserWorkflow.png

It can be done with Github, Gitorious, and Gerrit and I'm sure Oracle
will be amenable to adding this functionality quickly. We have a flow
similar to this at Eclipse using Gerrit and it works pretty well. The
one key feature at Github which is missing is the legal/IP aspects.
There needs to be a managed/canonical repository through which the
CLAs are collected so the license can be asserted. In my experience
technical able and keen developers overlook this. Hudson is a clear
case in point and it's just not an aspect of the project that can be
overlooked. It's just too important.

[Jesse Farinacci]

Hi,

On Mon, Nov 29, 2010 at 1:39 PM, jvanzyl <jason....@gmail.com> wrote:
>
> This was, in all honestly, a problem at Sun. This has moved pretty
> quickly within Oracle, and an agreement was made and work is in
> progress to fix the problems with the repositories.

I agree that it was a problem for Sun, but we're now nearly 9 months
since *outsiders* have issued a call to arms and have we got a good
fix in place? No.

> The fix is not
> simple but pretty soon after I posted that blog someone from Oracle
> contacted me and the relationship has been constructive. To that end
> Sonatype plans to work with Oracle on Hudson as well. Ted has been
> straight forward with us and we've been getting good things done.

I think this highlights one of the core problems going on here. So
much so that I am encouraged to ask the rhetorical question: Why
bother forking the project when it basically already has been? There
are apparently lots of discussions going on OUTSIDE of the established
community that created and nurtured the project for years now. That's
really not a good way to encourage a F/OSS community, as has been a
stated goal, ad nausem.

>> to be moved in November. While you state that KK may have missed an

I should have said NOT moved in November. Eek! Need coffee..

> With even the best of intentions from infrastructure folks shit
> happens. SVN has gone out for a week at Apache, we've had snapshot
> repos wiped out by mistake but you figure it out and move on.

That's fine, I know that shit happens. But I don't see anyone billing
Apache infrastructure, as Ted puts it, "... hosted by one of the
biggest IT organizations in the world who are use [sic] to
guaranteeing up time and reliability." If you want to set a standard,
then you have to meet it.

[Andrew Bayer]

FWIW, the intention is to end up managing contributions to core via Gerrit, hosted at hudson-labs.org probably, and syncing from there to the hudson core repo at Github, but that part of the migration is still in the theoretical stage. For now, access to core is only available to members of the core team in Github, and only those with CLAs on record with Oracle are added to that team.

Regardless, my original concern, more than any particulars about implementation, still remains - does the Hudson development community actually make the decisions regarding the Hudson project's development infrastructure? A key difference between the Apache or Eclipse situations Jason describes and Hudson's situation is that Apache and Eclipse projects have formalized and clearly established governance. Hudson, obviously, does not, and that's why we're were we are right now. I would very much like to work out a formal governance plan/structure with the rest of the Hudson development community, Oracle, and other interested parties.

A.

[bwilson]

> The
> one key feature at Github which is missing is the legal/IP aspects.
> There needs to be a managed/canonical repository through which the
> CLAs are collected so the license can be asserted. In my experience
> technical able and keen developers overlook this. Hudson is a clear
> case in point and it's just not an aspect of the project that can be
> overlooked. It's just too important.
>

Cripes - has someone got Linus' email address? He's evidently been
doing it wrong for the last 15 years.

[ted]

OK, there are more of you than there are of me so I'll consolidate my
replies to try and keep things readable.

Andrew, Tyler, if you have a list of the github features that you guys
use the most that wouldn't be available with the kenai version, that
would be helpful to get. You can email them to me directly or post
them here. This isn't intended to talk you out of it, but I am
curious as to what we are missing for a kenai-based git solution.

Nigel, what I am saying is that I believe the *final* decision of what
to do w.r.t. infrastructure belongs to Oracle and that decision should
be made according to the will of the community as it makes sense. We
want to be able to provide the infrastructure for the community so
that the developers don't have to worry about it. Up until now KK was
making that final decision since he was single-handedly supporting the
machines while he was at sun. We are now supporting that and have/are
moving this platform into a real server farm for better durability/
reliability. We are not prohibiting the developer community from
making decisions. In fact we are encouraging that they help form the
decisions being made. The decisions just need to be checked with the
realities of hosting the community and what is best for the growth of
the hudson ecosystem (eg. syncing with other projects, etc.)

Jesse, I am not clear as to your point. You left out the first part
of my paragraph about "Oracle is evil". That is what I was saying is
unsubstantiated. The old java.net infrastructure which sonatype was
rescuing was not kenai. It was a third-party hosting service that Sun
employed. Part of our move to the new java.net is to enable us to
better serve the java community projects on an infrastructure we are
in control of, instead of waiting/relying on changes from someone
else.

The mailing lists and achieves are just about back and working again
on the new java.net. To avoid any more confusion and fracturing, I
suggest we move back to using the mailing lists, wiki, issues, and svn
on java.net for now, and create a plan on what to do next. I think
the overwhelming opinion of the developers we talk to is the desire to
move to git. Many of them however don't really care if we move to
github or not. Some of you have strong opinions that we do. I would
like to take the opportunity to look into this a bit more before we
make a final decision. I see our options as moving to git and a)
staying only on java.net, b) moving only to github or c) having
java.net host it and mirror it on github, similar to what we are doing
today, only having them both use git instead of svn.

Our goal is to make sure everyone's opinion is heard and that we end
up with a sustainable solution. We value and respect all of the
opinions being voiced here and on email/phone to us. I think if you
can just be patient a bit longer you'll find that you'll have an
environment that the community is behind. This isn't any kind of
attempt to hurt or control the community. We just need to take into
account all of the factors involved, including synergies with other
java projects, the potential hudson user/contributor market, our
hosting capabilities, etc.

I hope this makes sense and that the discussion continues.

-ted

[ted]

I should also clarify that nothing should change until susan announces
things are ready on java.net. thx.

[ted]

Jesse, the reason we needed to reach out to so many of the users
"offline" was that they are not participating in the community. The
goal was get more of them back here. Now that we are making progress,
we are bringing the conversations back to the community. There is no
fork or secrets. We are trying to fix what we were told is broken
with the current hudson community. The simple fact is that there are
some people who are not comfortable voicing their opinions in an
environment which is centered around a small set of very vocal
people. If you read the tone of your replies to me, they don't seem
very willing to have an open conversation. They seem unnecessarily
aggressive. A lot of people will shy away when they see things like
this.

As for my comment about our IT department, I think you missed the
point. We are in the process as we speak to move this infrastructure
to our IT department. It has not been there up until now and I could
go into why it has taken so long, but I need a drink in my hand to go
through it all again and my therapist recommends I don't start
drinking before noon. :)

I think some skepticism is healthy, but I am getting the feeling you
are predisposed to not liking anything Oracle has to say. I hope that
we can move beyond that and start working together. I think we have
common goals w.r.t. the success of hudson and I think we can make it
better and stronger than it is today and really reach a lot of new
users and developers.

-ted

On Nov 29, 10:55 am, Jesse Farinacci <jie...@gmail.com> wrote:
> Hi,
>

[Andrew Bayer]

Ted -

I don't mean to be difficult, but I'm still confused as to who this silent majority is and what we as a community have been doing to drive them off - I'd love to know, so that we can address those problems, as a community. I don't see how it's helpful for anyone for Oracle to declare that the Hudson community is somehow stifling discussion without actually showing what we've done wrong, or bringing the offended parties into the conversation. Assuming that everyone in this discussion wants Hudson to continue to be a vibrant open source project with a strong, diverse and active development community, I think all interested parties need to have a voice in the discussion of any problems in the current Hudson community and its future.

My single biggest concern right now is that I'm not sure what all the connotations of your statement that Oracle has the final decision on Hudson infrastructure means for Hudson as a community-run project. Obviously, Oracle needs to be a key voice in the Hudson decision-making process, but I'm really not comfortable with the idea that any one entity alone has final say. Hudson's strength lies in its ecosystem, its massive array of community-developed plugins. I don't see how that community-centric focus can work if the community itself doesn't have control over its future.

A. 

[Jesse Farinacci]

Greetings,

On Mon, Nov 29, 2010 at 2:31 PM, ted <ted...@gmail.com> wrote:
>
> Jesse, the reason we needed to reach out to so many of the users
> "offline" was that they are not participating in the community.

I guess I don't see any advantage to proactively engaging people who
are so clearly not interested in engaging the community to which they
silently belong. I think you address some of this later, so let's
continue..

>  The
> goal was get more of them back here.  Now that we are making progress,
> we are bringing the conversations back to the community.  There is no
> fork or secrets.

My comment was directed at Jason when he said that nearly 9 months
ago, shortly after the Call to Arms against java.net's inferiority,
that he was in discussion with Oracle back then. Where are the mailing
list archives of that discussion coming back to the community table?
That means for 9 months there have been senior level people discussing
strategy outside of any user level arena. While it may not have been a
deliberate conspiracy, it sure as hell isn't a community process by
any stretch of imagination.

> We are trying to fix what we were told is broken
> with the current hudson community.

Here I believe one constant area for improvement which you've
mentioned, and also bugs me, is getting changes through the system.
Using git, and to a large part GitHub, in conjunction with Gerrit is
something that has been discussed and planned for quite a while now.
Even the most casual glance at GitHub's statistics, and the virtual
Whose Who of technical leadership that swear by it, shows the game
changing power that it provides. And at the risk of being overly
dramatic, once we couple that with Gerrit I expect a new renaissance
for Hudson.

> The simple fact is that there are
> some people who are not comfortable voicing their opinions in an
> environment which is centered around a small set of very vocal
> people.

I am reminded of the adage: the wheel that does the squeaking is the
wheel that gets the grease. In fact, the only time I can recall anyone
complaining about not being able to contributions in is when java.net
had mailing lists go offline for over a week and emails were not
automatically re-driven. To which the community, mostly Kohsuke and
Alan, promptly took action.

Also, this isn't some elementary school code camp where it's important
that everyone gets a turn... the people that actually pick up shovels
and do work don't seem to be doing any complaining so I am left to
wonder where is the real problem?

>  If you read the tone of your replies to me, they don't seem
> very willing to have an open conversation.  They seem unnecessarily
> aggressive.  A lot of people will shy away when they see things like
> this.

Dale Carnegie I'm not, but it's rather irrelevant. I find it extremely
hard to believe that so many users, let alone high profile ones, would
have so much invested in an application like Hudson and yet remain
completely silent on these issues. Who are these people? If you won't
or can't reveal them, then as far as I'm concerned they do not exist.

> As for my comment about our IT department, I think you missed the
> point.  We are in the process as we speak to move this infrastructure
> to our IT department.  It has not been there up until now and I could

So there will be a second migration within Kenai itself?

> go into why it has taken so long, but I need a drink in my hand to go
> through it all again and my therapist recommends I don't start
> drinking before noon. :)

*snickers* I think I can understand that, I just don't think it jives
with the whole concept that there's a large contingent of non-vocal
w/r/t Hudson community yet vocal w/r/t Sonatype/Oracle. I don't think
it would have taken so long to get changes in place if it were true. I
don't think the migration, and another internal one to follow?, would
have been such a low priority if it were true.

I'll take you at your word and suggest I've misunderstood the gravity
of these migrations, or perhaps the relative importance for these
people/groups to be able to have current deployments (and thereby be
relatively unconcerned with infrastructure outages lasting weeks).

> I think some skepticism is healthy, but I am getting the feeling you
> are predisposed to not liking anything Oracle has to say.  I hope that
> we can move beyond that and start working together.  I think we have
> common goals w.r.t. the success of hudson and I think we can make it
> better and stronger than it is today and really reach a lot of new
> users and developers.

There's always room for improvement, but I don't like it that we risk
alienating our existing user base (which is considerable[1]: that's
nearly __ 23,159 __ installs as of 3 months ago if you don't have
access to the Google Doc) to bring in non-vocal people who don't seem
to be contributing.

[1] https://spreadsheets.google.com/ccc?key=0AjOVjPKIuGKFdHVra1RhRkt6MmVDZEd5clREaVd4bUE&hl=en&authkey=CNqwqoUD#gid=0

[ted]

On Nov 29, 11:59 am, Jesse Farinacci <jie...@gmail.com> wrote:
> Greetings,
>
> On Mon, Nov 29, 2010 at 2:31 PM, ted <ted...@gmail.com> wrote:
>
> > Jesse, the reason we needed to reach out to so many of the users
> > "offline" was that they are not participating in the community.
>
> I guess I don't see any advantage to proactively engaging people who
> are so clearly not interested in engaging the community to which they
> silently belong. I think you address some of this later, so let's
> continue..

I'll consolidate my answer with similar items below.

> >  The
> > goal was get more of them back here.  Now that we are making progress,
> > we are bringing the conversations back to the community.  There is no
> > fork or secrets.
>
> My comment was directed at Jason when he said that nearly 9 months
> ago, shortly after the Call to Arms against java.net's inferiority,
> that he was in discussion with Oracle back then. Where are the mailing
> list archives of that discussion coming back to the community table?
> That means for 9 months there have been senior level people discussing
> strategy outside of any user level arena. While it may not have been a
> deliberate conspiracy, it sure as hell isn't a community process by
> any stretch of imagination.

Having some people talk about an initial proposal that you can then
submit for feedback/review/changes to a larger audience is pretty much
how anything real gets done. You can't start with an open canvas to
an unlimited set of people. I wasn't involved in a lot of what is
being referenced with sonatype/oracle, but if 9 months passed I doubt
seriously that the discussions were active that whole time. I see it
more as a statement of how long it takes to get anything done at a
certain level. I connected with Jason a month or so ago when we were
soliciting feedback about how to make hudson better. We were closing
in on being able to host things ourselves and thought it was time to
engage the community. We didn't want to engage everybody and then go
dark for 6+ months while we implemented a hosting platform.

>
> > We are trying to fix what we were told is broken
> > with the current hudson community.
>
> Here I believe one constant area for improvement which you've
> mentioned, and also bugs me, is getting changes through the system.
> Using git, and to a large part GitHub, in conjunction with Gerrit is
> something that has been discussed and planned for quite a while now.
> Even the most casual glance at GitHub's statistics, and the virtual
> Whose Who of technical leadership that swear by it, shows the game
> changing power that it provides. And at the risk of being overly
> dramatic, once we couple that with Gerrit I expect a new renaissance
> for Hudson.

Like I said, we don't have anything against github. I just want to
make sure we make a decision where all parties understand all
consequences. I think we will be in agreement with this by the time
it is settled. The hudson community was talking about git for a while
and it was only escalated because of this migration debacle. My goal
is to get that settled down and back running smoothly on java.net and
then talk about good things to come.

> > The simple fact is that there are
> > some people who are not comfortable voicing their opinions in an
> > environment which is centered around a small set of very vocal
> > people.
>
> I am reminded of the adage: the wheel that does the squeaking is the
> wheel that gets the grease.

I answer this below.

> >  If you read the tone of your replies to me, they don't seem
> > very willing to have an open conversation.  They seem unnecessarily
> > aggressive.  A lot of people will shy away when they see things like
> > this.
>
> Dale Carnegie I'm not, but it's rather irrelevant. I find it extremely
> hard to believe that so many users, let alone high profile ones, would
> have so much invested in an application like Hudson and yet remain
> completely silent on these issues. Who are these people? If you won't
> or can't reveal them, then as far as I'm concerned they do not exist.

What we actually found was a whole bunch of users who took a version
of hudson and never looked back. I wouldn't call these forks since it
was internal use only, but they grabbed the code and made a bunch of
changes for their own usage. Some did so because they thought it
would be easier or didn't care so much about re-contributing or
refreshing. Some did this because they had licensing concerns or
tried to get changes into the core and couldn't and became
frustrated. We found a couple examples of the latter that were about
to fork hudson into new communities because of that. We were able to
stop both of them and will hopefully bring them into the main hudson
community as active members. You can appreciate these kinds of
conversations don't lend themselves to a public forum.

So I think it isn't about community members sitting quietly as it is
about people needing to get things done and leaving or not joining the
community. When you look at how many users of hudson we found who
were changing/writing code around it, it didn't make sense that so few
were active in the community and giving that code back. I think in
many cases hudson isn't seen as a strong, open, active community to be
a member of. I know most of the people following this thread would
disagree, and we want to change that perception and the hurdles around
getting more people involved.

>
> > As for my comment about our IT department, I think you missed the
> > point.  We are in the process as we speak to move this infrastructure
> > to our IT department.  It has not been there up until now and I could
>
> So there will be a second migration within Kenai itself?

No. The move we are doing now to kenai will be our infrastructure.
I'm just saying you can't really pick on it until we're done with the
migration ;) Up until now it was hosted by a third party sun had
hired. It took us a while to get things in place where we started to
host things ourselves.

>
> > go into why it has taken so long, but I need a drink in my hand to go
> > through it all again and my therapist recommends I don't start
> > drinking before noon. :)
>
> *snickers* I think I can understand that, I just don't think it jives
> with the whole concept that there's a large contingent of non-vocal
> w/r/t Hudson community yet vocal w/r/t Sonatype/Oracle. I don't think
> it would have taken so long to get changes in place if it were true. I
> don't think the migration, and another internal one to follow?, would
> have been such a low priority if it were true.

Unfortunately the changes were some technical, some infrastructure and
many political. Hosting something the size of java.net as well as
trying to migrate thousands of active projects between platforms is
not an easy task. It took us a while to make sure we could do it.
Moving it all over will probably have some hiccups, but we wanted to
minimize them as much as we could before we went live. The end result,
however, will be a much stronger platform to build from.

> I'll take you at your word and suggest I've misunderstood the gravity
> of these migrations, or perhaps the relative importance for these
> people/groups to be able to have current deployments (and thereby be
> relatively unconcerned with infrastructure outages lasting weeks).

thanks. If we have an opportunity for a beer someday I can give you
some highlights. It's been quite a process.

> > I think some skepticism is healthy, but I am getting the feeling you
> > are predisposed to not liking anything Oracle has to say.  I hope that
> > we can move beyond that and start working together.  I think we have
> > common goals w.r.t. the success of hudson and I think we can make it
> > better and stronger than it is today and really reach a lot of new
> > users and developers.
>
> There's always room for improvement, but I don't like it that we risk
> alienating our existing user base (which is considerable[1]: that's
> nearly __ 23,159 __ installs as of 3 months ago if you don't have
> access to the Google Doc) to bring in non-vocal people who don't seem
> to be contributing.

Agreed. We don't want to come in and scare everybody that everything
is going to change. But at the same time, some things need to change
in order to grow. I think those two things can co-exist and be done
with the involvement of the community that has gotten hudson to where
it is today. You can't make everybody happy with every decision, and
there will be decisions made that probably wouldn't be Oracle's first
choice, but in the end, if the community sticks together and grows, we
all get what we want.

-ted

[ted]

On Nov 29, 11:59 am, Jesse Farinacci <jie...@gmail.com> wrote:
> Greetings,
>
> On Mon, Nov 29, 2010 at 2:31 PM, ted <ted...@gmail.com> wrote:
>
> > Jesse, the reason we needed to reach out to so many of the users
> > "offline" was that they are not participating in the community.
>
> I guess I don't see any advantage to proactively engaging people who
> are so clearly not interested in engaging the community to which they
> silently belong. I think you address some of this later, so let's
> continue..

I'll consolidate my answer with similar items below.

> >  The
> > goal was get more of them back here.  Now that we are making progress,
> > we are bringing the conversations back to the community.  There is no
> > fork or secrets.
>
> My comment was directed at Jason when he said that nearly 9 months
> ago, shortly after the Call to Arms against java.net's inferiority,
> that he was in discussion with Oracle back then. Where are the mailing
> list archives of that discussion coming back to the community table?
> That means for 9 months there have been senior level people discussing
> strategy outside of any user level arena. While it may not have been a
> deliberate conspiracy, it sure as hell isn't a community process by
> any stretch of imagination.

Having some people talk about an initial proposal that you can then
submit for feedback/review/changes to a larger audience is pretty much
how anything real gets done. You can't start with an open canvas to
an unlimited set of people. I wasn't involved in a lot of what is
being referenced with sonatype/oracle, but if 9 months passed I doubt
seriously that the discussions were active that whole time. I see it
more as a statement of how long it takes to get anything done at a
certain level. I connected with Jason a month or so ago when we were
soliciting feedback about how to make hudson better. We were closing
in on being able to host things ourselves and thought it was time to
engage the community. We didn't want to engage everybody and then go
dark for 6+ months while we implemented a hosting platform.

>

> > We are trying to fix what we were told is broken
> > with the current hudson community.
>
> Here I believe one constant area for improvement which you've
> mentioned, and also bugs me, is getting changes through the system.
> Using git, and to a large part GitHub, in conjunction with Gerrit is
> something that has been discussed and planned for quite a while now.
> Even the most casual glance at GitHub's statistics, and the virtual
> Whose Who of technical leadership that swear by it, shows the game
> changing power that it provides. And at the risk of being overly
> dramatic, once we couple that with Gerrit I expect a new renaissance
> for Hudson.

Like I said, we don't have anything against github. I just want to
make sure we make a decision where all parties understand all
consequences. I think we will be in agreement with this by the time
it is settled. The hudson community was talking about git for a while
and it was only escalated because of this migration debacle. My goal
is to get that settled down and back running smoothly on java.net and
then talk about good things to come.

> > The simple fact is that there are
> > some people who are not comfortable voicing their opinions in an
> > environment which is centered around a small set of very vocal
> > people.
>
> I am reminded of the adage: the wheel that does the squeaking is the
> wheel that gets the grease.

I answer this below.

> >  If you read the tone of your replies to me, they don't seem
> > very willing to have an open conversation.  They seem unnecessarily
> > aggressive.  A lot of people will shy away when they see things like
> > this.
>
> Dale Carnegie I'm not, but it's rather irrelevant. I find it extremely
> hard to believe that so many users, let alone high profile ones, would
> have so much invested in an application like Hudson and yet remain
> completely silent on these issues. Who are these people? If you won't
> or can't reveal them, then as far as I'm concerned they do not exist.

What we actually found was a whole bunch of users who took a version
of hudson and never looked back. I wouldn't call these forks since it
was internal use only, but they grabbed the code and made a bunch of
changes for their own usage. Some did so because they thought it
would be easier or didn't care so much about re-contributing or
refreshing. Some did this because they had licensing concerns or
tried to get changes into the core and couldn't and became
frustrated. We found a couple examples of the latter that were about
to fork hudson into new communities because of that. We were able to
stop both of them and will hopefully bring them into the main hudson
community as active members. You can appreciate these kinds of
conversations don't lend themselves to a public forum.

So I think it isn't about community members sitting quietly as it is
about people needing to get things done and leaving or not joining the
community. When you look at how many users of hudson we found who
were changing/writing code around it, it didn't make sense that so few
were active in the community and giving that code back. I think in
many cases hudson isn't seen as a strong, open, active community to be
a member of. I know most of the people following this thread would
disagree, and we want to change that perception and the hurdles around
getting more people involved.

>

> > As for my comment about our IT department, I think you missed the
> > point.  We are in the process as we speak to move this infrastructure
> > to our IT department.  It has not been there up until now and I could
>
> So there will be a second migration within Kenai itself?

No. The move we are doing now to kenai will be our infrastructure.
I'm just saying you can't really pick on it until we're done with the
migration ;) Up until now it was hosted by a third party sun had
hired. It took us a while to get things in place where we started to
host things ourselves.

>

> > go into why it has taken so long, but I need a drink in my hand to go
> > through it all again and my therapist recommends I don't start
> > drinking before noon. :)
>
> *snickers* I think I can understand that, I just don't think it jives
> with the whole concept that there's a large contingent of non-vocal
> w/r/t Hudson community yet vocal w/r/t Sonatype/Oracle. I don't think
> it would have taken so long to get changes in place if it were true. I
> don't think the migration, and another internal one to follow?, would
> have been such a low priority if it were true.

Unfortunately the changes were some technical, some infrastructure and
many political. Hosting something the size of java.net as well as
trying to migrate thousands of active projects between platforms is
not an easy task. It took us a while to make sure we could do it.
Moving it all over will probably have some hiccups, but we wanted to
minimize them as much as we could before we went live. The end result,
however, will be a much stronger platform to build from.

> I'll take you at your word and suggest I've misunderstood the gravity
> of these migrations, or perhaps the relative importance for these
> people/groups to be able to have current deployments (and thereby be
> relatively unconcerned with infrastructure outages lasting weeks).

thanks. If we have an opportunity for a beer someday I can give you
some highlights. It's been quite a process.

> > I think some skepticism is healthy, but I am getting the feeling you
> > are predisposed to not liking anything Oracle has to say.  I hope that
> > we can move beyond that and start working together.  I think we have
> > common goals w.r.t. the success of hudson and I think we can make it
> > better and stronger than it is today and really reach a lot of new
> > users and developers.
>
> There's always room for improvement, but I don't like it that we risk
> alienating our existing user base (which is considerable[1]: that's
> nearly __ 23,159 __ installs as of 3 months ago if you don't have
> access to the Google Doc) to bring in non-vocal people who don't seem
> to be contributing.

[Alan]

Ted-
Perhaps at this point you can step back and restate your goals for
this thread?

IMO, the issue at hand is where the source code will live. Once a
decision is made and migration work is done, people can start
committing again and some work can be done to restore the ability to
make releases.

Trying to be as brief as possible, here are the areas I've seen in
this thread and my thoughts on those:

- Discussion on stability

For me this is a non-factor in this discussion. I'm willing to accept
that java.net stability will be improved by this migration and that a
variety of factors made it take a long time to get going.

- Oracle wants to grow the community and get back more code
contributions

Great! But again, for me this is a non-factor in this discussion..
unless, Ted, you think that java.net will help these efforts? I think
most people on this alias have a good idea about the popularity of
github and how pull requests make it easy to bring code contributions
into a project. Ted, if you think java.net as the source repository
factors into this part of the discussion, please explain to us how.

- Where source will live

Again trying to be as brief as possible, here is how I read this part
of the discussion:
Andrew ran a trial migration to GitHub, sent out a message about it,
got a lot of support, then sent a "PROPOSAL:" message to propose the
plan.
Ted sent his message saying "For now, however, we are going to stay on
the java.net
infrastructure" and "Because it is open source, we can't stop anybody
from forking it."

So this brings me back to: Ted, what are you looking to achieve with
this thread? As you simply stated an outcome rather than offering a
counter-proposal with reasons why people might choose to support it,
I'm wondering how you expect the core contributors to react to this?

Looking forward to your reply,

- Alan

[teilo]

svn url is https://svn.java.net/svn/hudson~svn

/James

On Nov 29, 3:52 pm, Jacob Robertson <jacob.robertson.w...@gmail.com>
wrote:

[teilo]

On Nov 29, 8:42 pm, ted <ted...@gmail.com> wrote:

> No.  The move we are doing now to kenai will be our infrastructure.
> I'm just saying you can't really pick on it until we're done with the
> migration ;)

But there-in lies the problem.
There was no schedule that we the community had knowledge of - even
one week later it is still not done.

All this should have been piloted to death before coming anywhere near
a real project - at which point it should not take more than a day to
do it all.

So whilst we can not pick on it - you will excuse me if I have some
reservations about if it will actually be any good.

My current view given that the web based svn viewer adds random stuff
to code is that this (java.net beta) is not in a good state and won't
be for a long while :-(

for example compare the following

https://svn.java.net/svn/hudson~svn/trunk/hudson/pom.xml
and
http://java.net/projects/hudson/sources/svn/content/trunk/hudson/pom.xml?rev=37124
the latter is littered with random semi-colons
<url>https://hudson.dev.java.net/<;/url>

/James

[teilo]

stick a trailing slash on it as well :-/
https://svn.java.net/svn/hudson~svn/

On Nov 29, 10:55 pm, teilo <teilo+goo...@teilo.net> wrote:
> svn url ishttps://svn.java.net/svn/hudson~svn

[ted]

Alan, in a nutshell the goal of this thread is to bring the
infrastructure and source location of hudson back to where it was a
week ago, and that is java.net. It is also to let people know that
Oracle would like to be involved in any decision about the
infrastructure since we own it and want to continue to provide it.
SVN on java.net is available now for checkins. That is where people
should be committing their code. We hope to have the mailing lists
and other supporting pieces up in the next day or so at which time
everyone should use that. Once we get all of that back in running
order, we should continue the discussion of some proposed changes such
as a git-based repository and where that might live. Let me know if
you have any other questions.

telio, the lead of every java.net project that is currently being
migrated was notified about this migration ahead of time. KK received
that email ahead of time. Our goal is that we are shooting for
everything to be functional again on 12/3, but in the hudson case we
hope to be done sooner to help with the issue that the community was
not made aware of this change.

-ted

[ted]

Alan, in a nutshell the goal of this thread is to bring the
infrastructure and source location of hudson back to where it was a
week ago, and that is java.net. It is also to let people know that
Oracle would like to be involved in any decision about the
infrastructure since we own it and want to continue to provide it.
SVN on java.net is available now for checkins. That is where people
should be committing their code. We hope to have the mailing lists
and other supporting pieces up in the next day or so at which time
everyone should use that. Once we get all of that back in running
order, we should continue the discussion of some proposed changes such
as a git-based repository and where that might live. Let me know if
you have any other questions.

telio, the lead of every java.net project that is currently being
migrated was notified about this migration ahead of time. KK received
that email ahead of time. Our goal is that we are shooting for
everything to be functional again on 12/3, but in the hudson case we
hope to be done sooner to help with the issue that the community was
not made aware of this change.

-ted

On Nov 29, 2:01 pm, Alan <mindl...@dev.java.net> wrote:

[Kohsuke Kawaguchi]

2010/11/29 ted <ted...@gmail.com>:

> I am sorry for the inconvenience caused by our moving the hudson
> project to the new java.net.  We sent out an email to all of the
> projects (including hudson) warning them of this change and all of

> them except hudson had planned for it.  As KK pointed out in his
> email, he somehow missed the heads-up email our IT department sent
> him.

Since this point was mentioned repeatedly, I'd like to make sure facts
are correctly presented.

The week of Nov 22nd was assigned for the 2nd wave GlassFish community
migration, and the plan was that the general java.net projects migrate
in batches in the coming weeks of December [1]. This was well
publicized, and naturally projects that belonged to the GlassFish
community was well prepared. For some reasons, Hudson was in the list
of this migration wave, even though it was not a part of the GlassFish
community. But since it was just one of 161 projects, no one in the
GlassFish community nor the Hudson project noticed it beforehand.

On 19th Friday around noon pacific time, the first and the last e-mail
that notifies the migration was sent to dev, users, and owner aliases
of all the projects in the 2nd wave migration. The e-mail was never
acually delivered to users and dev, because the sender was not in the
subscriber list. The e-mail to the owners alias was delivered to me
and Winston, but both of us failed to notify the community in time. In
my case, on this Friday I was flying back from Utah to San Francisco,
and failed to read the e-mail in a timely fashion.

The project was locked down on 21st Sunday in preparation of the
migration, and I noticed on 22nd Monday morning about what's really
happening, after a plugin developed reported that he can no longer
commit to the repository.

IMO, even if I were to read the e-mail right after it was sent, it
still doesn't give the community nearly enough time to plan/coordinate
anything (for the project the size of Hudson) --- By Friday noon
pacific time, everyone in Europe and Asia had already headed home for
a week, and the east coast will follow in a few hours.

[1] http://weblogs.java.net/blog/communitymanager/archive/2010/11/03/javanet-begins-migration-collabnet-kenai-infrastructure
--
Kohsuke Kawaguchi

[Vic]

It appears that (some) Hudson developers will have to accept Oracle
managers as setting direction.
The alternatives is that (some) 'Hudson' developers could fork the
project formally known as Hudson, if Oracle owns the Hudson tm (even
if it does not, developers have no lawyers to fight it). As a user of
Hudson, I need committers/developers of this project and that's whom I
would follow. And I hope that their donated time is not spent on
politics or management.
In Summary, I hope there is a new Hudson based code base fork.
Maybe similar to http://www.documentfoundation.org/foundation vs
Oracle Open Office, they forked). I doubt funding or free hosting
would be any issue.

[Winston Prakash]

Thought of keeping quiet until actual engineering work starts again after these logistical details are worked out. But decided to make one exception and reply to this.

I spoke to the manager who is responsible for hudson migration to find out why the community members are not notified. She mentioned all members of the project are notified on Dec 19th, which surprised me because I did not see the e-mail in the hudson user or dev mailing list.

It turned out, before the migration, dev and user mailing list was notified. Unfortunately, looks like, who ever send out the e-mail was not subscribed to the mailing list, so the announcement did not go through and reached the community members. Here is the actual e-mail send out.

---

-------- Original Message --------

Subject:	Please Read: This project will be migrated to Kenai, starting 10 PM, 11/21/2010 (PST)
Date:	Fri, 19 Nov 2010 11:25:22 -0800
From:	Ed Bratt <ebr...@dev.java.net>
Reply-To:	ebr...@dev.java.net
To:	us...@hudson.dev.java.net, ow...@hudson.dev.java.net, d...@hudson.dev.java.net

This project will be migrated to Kenai in the wave-2 migration effort. This project will be locked, starting at 10 PM Pacific Standard Time, Sunday November. 21. We will work to unlock this project as soon as possible however, we cannot predict the exact schedule for importing projects into Kenai. The whole effort is expected to take about two weeks, possibly three. Projects will be activated at [project].java.net as soon as all data are imported into Kenai.

Please watch http://kenai.com/projects/glassfish-migration/pages/Home
for updates regarding progress.
If you have any questions about the projects which are going to be migrated, please review the complete list, here.

Actions you should take

User Credentials
If you have not activated your new java.net user credentials, please do that now. (beta.java.net, select login. If you did not receive your welcome e-mail, click "lost password.")
Sync your local source copy
You should take action now to Sync. your source code copies.
Projects from Collabnet servers will be read-only, once they are locked. However we will be switching projects over to Kenai, as they become ready so we cannot give a precise target date for this, for each project. Once they are switched, you will not be able to access the old repository from Collabnet.
E-Mail
Starting, this Sunday, e-mail to this project will cease since that is part of the migrated data. Once the project is re-enabled, the same set of project lists will be at [List]@project.java.net (remove the .dev).
Learn about the new Java.Net
Familiarize yourself with the new java.net. Browse to beta.java.net and read the help documentation. Or, you can read the orientation and other information we have compiled at the GlassFish Migration WIki.

Getting Started
FAQ
Migrating your GlassFish site

Prepare to update your web-site files
Project Site Managers and Administrators should look for instructions that you'll need to follow, to update your web-site. You can review the migration procedure at the GlassFish Migration Wiki "Guide for Migrating Your Site".
CVS will migrate to SVN
Reminder -- The migration effort will attempt to migrate all CVS archives to SVN. If this cannot be done without error, the original CVS archive will be attached to your project. SVN has stronger integration with Kenai so this will be your best option. If there are errors and your repository is not switched, we can discuss options for additional conversion methods.
Repository Archives
Repository archives, exported by Collabnet will be saved indefinitely so, if your project determines a need for this, you can request it. However, do not expect to receive these before Mid. January (unless there is an urgent issue).

Thank you,
-- Ed Bratt

---

[joelhealy]

On Nov 29, 7:33 pm, ted <ted...@gmail.com> wrote:
> Alan, in a nutshell the goal of this thread is to bring the
> infrastructure and source location of hudson back to where it was a
> week ago, and that is java.net.  It is also to let people know that
> Oracle would like to be involved in any decision about the
> infrastructure since we own it and want to continue to provide it.

Well that is Ted's goal for this thread. As a representative of the
silent majority of long time Hudson users, my goal for this thread is
to say this:

Fork this sh*t!

It is obvious even to an ignoramus like me that Oracle has no interest
in supporting an open project that they don't control. If I am wrong,
then they would be willing to support a forked community ruled open
project based on Hudson and infrastructure they don't control. Right?

The user community will support the development community not Oracle.

Sorry for being so offensive, but I am feeling pretty offended myself.

-joel

[joelhealy]

On Nov 29, 7:33 pm, ted <ted...@gmail.com> wrote:

> Alan, in a nutshell the goal of this thread is to bring the
> infrastructure and source location of hudson back to where it was a
> week ago, and that is java.net.  It is also to let people know that
> Oracle would like to be involved in any decision about the
> infrastructure since we own it and want to continue to provide it.

[ted]

In reading posts like vic's and R Tyler's blog on this topic [1] (I
think would have been more constructive to keep the conversation here
instead of dividing across mediums and websites), I want to make sure
I summarize what this thread was about (again).

Oracle is not at an impasse with Hudson developers. We *are* hudson
developers. We responded to a bunch of quick changes that were being
made to the hudson infrastructure that were causing alarm with some of
it's users/contributors. The migration of java.net sparked it all,
and we are remedying that situation. The repository and mailing lists
are back and working again on java.net. I believe that we should take
a step back from making a bunch of quick changes that scatter the
assets of hudson across many different domains and solutions and make
make sure we are doing the right thing for the community and project
as a whole.

I have stated multiple times that Oracle is in favor of moving to a
git-based repository, including possibly github, and we just wanted
some time to evaluate what that means and the best way to achieve it.
Somehow that simple request was taken as we are trying to stop the
will of the developers. That seems pretty unfair. I'm not sure why
things need to turn into sensationalized posts based on speculation of
all the bad things that Oracle might do. We are here, trying to work
in a community we have been part of for almost a year, trying to make
it better for everyone and promote it's growth and potential.

-ted


[1] http://www.hudson-labs.org/content/whos-driving-thing

[ted]

Joel, what is your actual issue? Oracle participates in hundreds of
open source/standards projects that we don't own and we are not asking
to control this. We want to coordinate it just like every other open
source project has a single point of coordination that takes feedback
from the community, and just like KK was doing before he left. Your
statement is not based in any fact. If you can please clarify your
exact issues sticking to real facts and examples, I would be happy to
address them.

-ted

[Nigel Magnay]

So can I take from that Oracle is willing to donate the Hudson name and associated logo rights to the community, as a gesture of goodwill ?

You see, the problem is that you're talking a lot about building a better community, but there's the whiff of Oracle treating the developers, because of the remaining copyright over the name, acting as if it is the "first amongst equals" - that we all get a say, but really it's just theatre - the decision and veto lies with you. As abayer pointed out earlier, that's no way to build a community, particularly with this impression that other discussions are happening out of sight. 

By coming on board with the community and giving up the name in this way you could completely eliminate this issue, and many people would be reassured of your good intentions by taking part as an equal partner in what we all, collectively, decide. If not, there's always going to be a lingering doubt, git/github/etc notwithstanding, of whether it just might have been better to bite the bullet now, and fork.

[Nord, James]

Ted,

I think this stemmed from the wording of your original post which sounded dictatorial rather than an objection to the move with a counter proposal.

> For now, however, we are going to stay on the java.net infrastructure.
> We believe it is important for hudson to stay connected with the rest
> of the the java community

I certainly took that to mean the Oracle "we" and not the Hudson (development) community we.

I think what you are trying to say is (and please correct me if I am wrong):

Oracle would like the code base to stay on the new java.net infrastructure.

If it is missing features (or stability) then we (the community development or user base)
can raise these and we (the community) can then discuss any potential solutions.
The solutions could be that Oracle provide the missing features in java.net in a
timely fashion or it could be that we should consider moving (some) of the infrastructure
elsewhere.

/James

-ted

[1] http://www.hudson-labs.org/content/whos-driving-thing

**************************************************************************************
This message is confidential and intended only for the addressee. If you have received this message in error, please immediately notify the postm...@nds.com and delete it from your system as well as any copies. The content of e-mails as well as traffic data may be monitored by NDS for employment and security purposes. To protect the environment please do not print this e-mail unless necessary.

NDS Limited. Registered Office: One London Road, Staines, Middlesex, TW18 4EX, United Kingdom. A company registered in England and Wales. Registered no. 3080780. VAT no. GB 603 8808 40-00
**************************************************************************************

[Marcelo Silveira]

Since Ted brought the silent community to the table and I'm part of
it, I want my voice to be heard against his arguments.

+1 for the fork

It's a shame this thread motivated me to post my first message to the
list. I hope the OSS part of Hudson surpass the Oracle style.

Cheers and thanks a ton to the development community.

> This message is confidential and intended only for the addressee. If you have received this message in error, please immediately notify the postmas...@nds.com and delete it from your system as well as any copies. The content of e-mails as well as traffic data may be monitored by NDS for employment and security purposes. To protect the environment please do not print this e-mail unless necessary.

[ted]

James, I see your point. I think the way you worded it is more in
line with our sentiment. We are trying to bring the infrastructure
back to where it was 10 days ago (all on java.net) and then talk about
the changes we want to make as a community. Last week saw multiple
changes being made without any input from most of the community. I
appreciate that some of it was because of the java.net migration, but
rather than leave it scattered we want to make sure we are doing
things for the right reasons. We see a lot of benefits of having some/
all of this on java.net and would like that to be part of the
discussion before final decisions are made by the community. In
earlier discussions java.net was seen as an obstacle, where we believe
the new kenai-based infrastructure will be a benefit. I think that
deserves more discussion.

-ted

> This message is confidential and intended only for the addressee. If you have received this message in error, please immediately notify the postmas...@nds.com and delete it from your system as well as any copies. The content of e-mails as well as traffic data may be monitored by NDS for employment and security purposes. To protect the environment please do not print this e-mail unless necessary.

[Juha Heimonen]

One more from silent community, being my first post and that on rather
unpleasant subject. If, for some reason, there should be a fork then
fork it.
I just wonder what this has to do with _users_, because I kinda think
that there would still be users-list for asking stuff, hudson.war
available, plugin updater etc, regardless if the development is done
in github, java.net,
well, little do I know, I'd just kinda think that maybe there aren't
that many silent developers really, who wouldn't have their say on the
subject, and it _seemed_ that the developers were quite unanimous on
how things should be.

Thank you, dev-community for the work you've done, and you seem to
have a good thing going on.

br, Juha H

[Alan]

On Nov 30, 10:25 am, ted <ted...@gmail.com> wrote:
> We see a lot of benefits of having some/
> all of this on java.net and would like that to be part of the
> discussion before final decisions are made by the community.  In
> earlier discussions java.net was seen as an obstacle, where we believe
> the new kenai-based infrastructure will be a benefit.  I think that
> deserves more discussion.

I am open to discussion. I ask everyone reading this to do your best
to put aside your feelings about Ted's handling of this situation so
far (hard for me too), and be open to discussion as well.
Ted, you have acted in a very corporate manner.. the VP steps in and
makes the call. Probably this is how you are used to operating.
However, you can see the reaction to this approach in this arena.
If you can start your next message with something like "Can we hold
off on the migration to github for now? Here are the reasons I think
we should use java.net for source repository, ...", then I for one
will read it and consider it. To date I have not seen any requests
from you, nor any reasons ("We see a lot of benefits".. but no
details).

Thanks,
- Alan

[R. Tyler Croy]

On Tue, 30 Nov 2010, Juha Heimonen wrote:

> One more from silent community, being my first post and that on rather
> unpleasant subject. If, for some reason, there should be a fork then
> fork it.
> I just wonder what this has to do with _users_, because I kinda think
> that there would still be users-list for asking stuff, hudson.war
> available, plugin updater etc, regardless if the development is done
> in github, java.net,
> well, little do I know, I'd just kinda think that maybe there aren't
> that many silent developers really, who wouldn't have their say on the
> subject, and it _seemed_ that the developers were quite unanimous on
> how things should be.

Regardless of how this all plays out, things should still operate perfectly
fine for users. Thanks to a generous donation of space and bandwidth we have
packages and war files mirrored on OSUOSL:
http://ftp.osuosl.org/pub/hudson/

We will be doing whatever we can to ensure the continuity of Hudson progress
for users, since largely, they are us :)

> Thank you, dev-community for the work you've done, and you seem to
> have a good thing going on.

Appreciate the well wishes, having started out in the OpenBSD community I can
say with certainty that the Hudson users and developers communities are some of
the friendlist around :)

- R. Tyler Croy
--------------------------------------
GitHub: http://github.com/rtyler
Twitter: http://twitter.com/agentdero

[Roger Vaughn]

Count me as part of the silent community, too. I do not particularly
want to see a fork, but I sure don't want to see Hudson fall prey to
the "first among equals" or "sole custodian" models either. Over the
past couple of days I've been forced to poke into the Hudson code (in
particular the MSBuild plugin) and I thought that GitHub was much,
much easier to use than the old SVN repos, and much more convenient.
If I have only one request it's this - please, please include detailed
descriptions on the "official" project pages! There are so many forks
on GitHub, and there are already several different repos for Hudson
and its plugins, that it can be very difficult to tell which ones are
supposed to be the "reference" or "official" repos. (There is an
excellent example here: https://github.com/sitaramc/gitolite)

I have to say that in no way did the Hudson development community ever
discourage me from contributing. Ted, specific examples are really
needed to back that statement up. There were only two factors that
kept me from contributing: 1) I had no need. Hudson and its plugins
have, until now, served my needs as-is. 2) Personally I found the old
java.net infrastructure slow, byzantine, and extremely hard to use. I
avoided "vanilla" java.net projects like the plague, and only hudson-
ci's custom skinning made it usable. The broader open source hosting
community is years ahead of the java.net standards, and Oracle would
be better served to try to integrate more with that than to isolate
Java projects behind this "java.net walled garden". Yes, that's really
what it feels like. Visiting a java.net project is like a step back
into history. I appreciate that Oracle expects to improve it, but why
keep trying to reinvent what the broader community has already
addressed?

Project Kenai may very well be a big improvement, but that needed to
be demonstrated to the various project communities before decisions
were made, and before the old java.net sites were taken down. The best
way to do that would have been to run parallel sites. Yes, it may have
caused some short-term confusion, but at least it would have given the
developers a chance to see what they were getting. And it's hard to
claim that the old site suddenly going dark didn't create confusion.
Ted, in reference to GitHub alone you say that "we just wanted some
time to evaluate what that means and the best way to achieve it", and
yet KK (and through him the community, at least in intention) was
given at best two days notice for the whole infrastructure going dark?
Well you know, regarding the migration to Project Kenai and the new
java.net infrastructure, "we just want some time to evaluate what that
means and the best way to achieve it." That consideration MUST flow
both ways.

There is already too much confusion and fragmentation in the online
project-hosting and repository space. I really hate the thought of
Oracle contributing more fragmentation to that, and furthering Sun's
old walled garden approach.

In short, I'm in favor of the move to GitHub AND the move to Google
Groups. In addition to the improved functionality, that also gives me,
as a user, fewer services I have to learn, sign up for, and keep track
of, since I already use them for other projects.

[Henrik Lynggaard]

HI

I sincerely hope that is not going to happen. There is no need to
fork.

[Henrik Lynggaard]

Hi

Since everybody seems to be voicing their opinion I thought I might as
well. I am for the most part, part of the silent majority and a casual
contributer.By casual I mean I have contributed to 2 plugins and a few
other patches.

I have only been following the debate on the sideline until now, and I
honestly can't be bothered to answer each mail individually so I just
lump everything together here.

1) Ted/Oracle has already admitted several times that there was a
screw-up in the migration and notification. Mistakes happen and they
fessed up to it. That should be the end of that part of the debate,
please stop re-iterating the same complaint over and over again. We
heard you the first (and second and third) time. Calm down ;-)

2) I think the decision to spread Hudson's infrastructure in 117
different pieces all over the Internet is premature. As a casual
contributer/user I get very little benefit if any from it and I do not
see the need for rushing things. It might be that github is a better
social hub for open source scm hosting, but to me I see more value
from having all Hudson things in one place.

3) I am undecided about the move to git. To me it has a higher entry
barrier than subversion. I use Netbeans daily but it has no git
support (or no easy support). I tried to download NbGit but it threw
exceptions when trying to clone a hudson plugin repo and failed to do
anything. Hudson in SVN worked out of the box. Maybe I am doing it
wrong but that was my user experience.

4) if Roger's statement about the multiple hudson repos on github is
true, it allready sounds way too complex and fragmented. I confuses me
and I see no value added.

5) To me it sound like Oracle wants to add any features that we are
missing from github, so instead of complaining let us see what is
missing and fix it. That will not only help us but all projects on
java.net.

6) I have to say that the "Overwhelming support" for moving to github
is I bit overstated in my mind. Going by "Proposal: Hudson plugin
repositories on GitHub " and "PROPOSAL: Formal migration to Github
+Google Groups " I see roughly 40 reponses from 10-15 people. I know
most if not all of them contributes far more than me, but with 300+
plugins I don't really see it as the majority of developers.

7) I agree that both the old and the new java.net is missing some
features, but lets get those identified to Oracle can start working on
them or at least relay how feasible they are.

8) To me forking is a no go, it simply looks like something people
want to do because their toes got hurt during the migration. I
sincerely hope the senior contributers do not buy into the mob rules.


If I had the lead I would properly do something like this:
* stay on java.net svn
* stay on the current Jira (which is oracle backed as far as I know)
* Push to get a better wiki on java.net. Confluence sounds like a good
choice.
* Evaluate the difference in features between github and java.net-git.
Let us see how easy it is to get the missing features into java-net-
git and only if there are string reason move to github

In the ideal world I would properly set up a build forge consisting
of Hudson,Maven repo, Jira, Conflucence, SVN+git+mercurial SCM,mailing
lists with NNTP support, a bunch of hudson slaves for selenium
browser testing in different browsers.

Best regards
henrik

[Les Mikesell]

On 11/30/2010 11:26 AM, ted wrote:
>
> I have stated multiple times that Oracle is in favor of moving to a
> git-based repository, including possibly github, and we just wanted
> some time to evaluate what that means and the best way to achieve it.
> Somehow that simple request was taken as we are trying to stop the
> will of the developers.That seems pretty unfair.

How else can anyone outside interpret that? It sounds pretty unilateral
to me, ignoring a decision that seemed to already have been made.

--
Les Mikesell
lesmi...@gmail.com

[joelhealy]

On Nov 30, 11:33 am, ted <ted...@gmail.com> wrote:
> Joel, what is your actual issue?  Oracle participates in hundreds of
> open source/standards projects that we don't own and we are not asking
> to control this.  We want to coordinate it just like every other open
> source project has a single point of coordination that takes feedback
> from the community, and just like KK was doing before he left.  Your
> statement is not based in any fact.  If you can please clarify your
> exact issues sticking to real facts and examples, I would be happy to
> address them.
>
>    -ted

Ted, my actual issue is Oracle's attempt to control Hudson, and your
attempt to convince us that you are taking actions which are in the
best interest of the Hudson community.

Fact: Oracle totally botched the java.net migration from the
perspective of the Hudson community.
Fact: Oracle had every right to handle the migration of their own
infrastructure any way they wanted.
Fact: The Hudson community is under no obligation whatsoever to use
the Oracle infrastructure.
Fact: Oracle wants to coordinate the Hudson project and act as a

single point of coordination that takes feedback from the community

just like KK was doing before he left.

Fact: When the Hudson community reacted negatively to the botched
java.net migration and took steps collectively to mitigate the damage,
Oracle VPs did not offer fact-based reasons why it would be in the
best interest of the Hudson community to return to java.net, but
instead tried to assert control and dictate a rollback.

Ted, you have every right to try to control the Hudson project on
behalf of your employer. Please just don't insult my intelligence by
trying to convince me that you are acting in the best interest of the
Hudson community. Fact: You are not!

If you believe that the Hudson project needs a single point of
coordination that takes feedback from the community just like KK used
to do, might I suggest that you ask the Hudson community who they
think would be an appropriate single point of coordination and then
abide by their decision. Assumption (not fact): I don't think the
Hudson community would support your appointment to that position.

I have no problem with your attempt at a sales pitch for Oracle
control. All I ask is that you don't insult me by telling me you only
have my best interests at heart.

Oracle is not evil. It is also not Hudson's Guardian Angel. Keep it
real bro.

[ted]

Joel, thanks for the reply. I still don't think all of this skeptisim
is warrented, but let's leave that be for a minute and let me ask you
a simple question. What do you think Oracle gains by wanting to
control hudson? We don't make any money from it. If we were to ever
charge people for it the community would fork immediately. So what do
you think is behind this big conspiracy you claim I am running?

Oracle is a big company and I think a lot of judgment is being made
towards it as a whole, instead of talking about me and my team who are
involved with hudson. If you have real examples of how we ruin and
control projects, I would love to hear it. If you want to know what
it's like to work on an open source project that my team leads, talk
to the developers and users of Trinidad. It is an apache project that
we started years ago. You can also talk to the leads of dozens of
other standards and projects that we have been a key part of, like
JSF. Nobody is perfect, but regardless of what you think about these
technology, I think we have done really good things for the projects
that we have been actively involved in.

As far as hudson goes, we like hudson. We use to use a lot of cruise
control here but started to use hudson once we acquired sun. In
talking to other companies that we do business with, we heard a lot of
rumblings about how it is impossible to get changes into the core of
hudson, or that the licensing is a mess and people can't repackage
it. We also stopped 2 different fork attempts. Stopping a fork
attempt sounds like something that is good for hudson. Don't you
think?

In this thread you have heard some other opinions about lack of
process and fragmentation of assets. That is also what we heard. We
would like to commit more things internally and to our customers and
encourage more use of hudson and see the product grow. Our concerns
are very similar to some we have heard, which is that it is hard to
make changes, the core is not very pluggable with the options to use
different technologies and the licensing makes it impossible for us to
ship it along with our developer tools to try and get more of our
customers using it.

I think putting the hudson infrastructure on a real, scalable platform
and providing more structure and openness to the community will
greatly help the product and attract more developers and users.
That's it. No evil plot. We happen to own the assets and rights to a
product that we really like and want to see it succeed so we can
continue to use it internally and provide our customers with
integrated solutions that make them more productive. I know you have
been assuming the worse, but try to imagine what a company with the
resource of Oracle could do for hudson w.r.t. infrastructure, coding
resources and exposure.

I think there is room for improvement with hudson today. That doesn't
mean controlling it or loosing the community. It means the opposite
of that.

-ted

[rossjudson]

If you really need a name, I hereby donate mine. ;) It's only one
letter different.

Ross Judson

[Shawn Bower]

>
> I think putting the hudson infrastructure on a real, scalable platform
> and providing more structure and openness to the community will
> greatly help the product and attract more developers and users.
> That's it.  No evil plot.  We happen to own the assets and rights to a
> product that we really like and want to see it succeed so we can
> continue to use it internally and provide our customers with
> integrated solutions that make them more productive.

As oppsed to github? They have a mere 1.5 million projects hosted at
github which seems pretty scalable to me. If you want the community
to be more productive then lets use github which has emerged as a
standard in the development community. Honestly as a user of
java.net I can say Im thoroughly unimpressed with it and have no
belief that new version will be the least bit better. Oracle's
motivation is to pimp java.net which is just sad.

> I know you have
> been assuming the worse, but try to imagine what a company with the
> resource of Oracle could do for hudson w.r.t. infrastructure, coding
> resources and exposure.

The resources of even a large corporation like Oracle pale in
comparison to the open source community. The project is better
without Oracle trying to assert control. If Oracle had any clue about
dealing with open source they would support the hudson community
instead of turning it against them. Let's not pretend this is just an
issue with Oracle and the Hudson community this is an issue with
Oracle and open source.

http://blogs.techrepublic.com.com/opensource/?p=2003

[TrashHalo]

> Oracle is a big company and I think a lot of judgment is being made
> towards it as a whole, instead of talking about me and my team who are
> involved with hudson.  If you have real examples of how we ruin and
> control projects, I would love to hear it.

I agree that people are becoming uncomfortable with oracle as a whole
and not specifically your team. Not sure if we in this context means
oracle or we means your team. But its not a stretch to say oracle has
been making so rather unpopular decisions since taking over sun.

*The continuation of the Apache dispute
https://blogs.apache.org/foundation/entry/statement_by_the_asf_board1
*The Google/Android lawsuit
*The nomination of Hologic to the JCP http://www.jroller.com/scolebourne/entry/stacking_the_jcp_election
*Oracle effectively forcing the OpenSolaris community to fork
http://dtrace.org/blogs/ahl/2010/08/27/the-future-of-solaris/
*The request for members to Document Foundation to step down because
of a "conflict of interest" http://wiki.services.openoffice.org/wiki/Community_Council_Log_20101014
*The countless misunderstandings (paid java, mysql, etc) due to lack
of communication from Oracle

Wether Oracle is doing the right thing or not it can't be denied there
is a section of the developer community that is put on edge by these
actions and I think it is causing people to respond to you with much
more skepticism and doubt. One only has to look at the comments on
reddit or hudson labs to hear tons of developers chanting fork.

>Stopping a fork attempt sounds like something that is good for hudson. Don't you think?

Forks aren't inherently bad. Projects are forked every day on github.
They only becomes an issue if one party makes it an issue. Often code
is shared between forks of open source projects. Though it does seem
like fork is a bad word at Oracle. Its what they quickly declared
LibreOffice and gave it this bad connotation.

[TrashHalo]

> >Stopping a fork attempt sounds like something that is good for hudson.  Don't you think?
>
> Forks aren't inherently bad. Projects are forked every day on github.
> They only becomes an issue if one party makes it an issue. Often code
> is shared between forks of open source projects. Though it does seem
> like fork is a bad word at Oracle. Its what they quickly declared
> LibreOffice and gave it this bad connotation.

I found the quote that best describes what I was trying to say about
forks

"The Oracle employees who are members of the OpenOffice.org project
and who expressed themselves these past days have displayed a
disturbing lack of understanding of Free and Open Source Software;
LibreOffice is, after all, and until proven otherwise, a downstream
version of OpenOffice.org, and as such deserves inclusion into the
OpenOffice.org community. I can only imagine what it would be like if
Debian was rejecting the Ubuntu employees among its teams, calling it
a fork. As for the fork itself, and because we’re still a downstream
version of OpenOffice.org, forks become forks only when one of the
boys refuse to play ball with the others; and the Oracle team of
OpenOffice.org just did that."
http://standardsandfreedom.net/index.php/2010/10/22/leaving-the-openoffice-org-project/

[Ken Liu]

Ted -

Seriously? It's pretty obvious what makes github great.

To name a few features:
- Performance (responsiveness of web site)
- Clean user interface
- Pull request management
- Social features (messaging, following, repository watching)

Frankly, even if kenai were to replicate all of the features of
github, it still wouldn't be as desirable as github itself since so
many other projects are hosted there and so many other developers have
github accounts. This makes the barrier to entry very low because if
you already have a github account (as practically every serious OSS
developer does), you can just fork with a single click.

On Mon, Nov 29, 2010 at 2:16 PM, ted <ted...@gmail.com> wrote:
> Andrew, Tyler, if you have a list of the github features that you guys
> use the most that wouldn't be available with the kenai version, that
> would be helpful to get.  You can email them to me directly or post
> them here.  This isn't intended to talk you out of it, but I am
> curious as to what we are missing for a kenai-based git solution.

[Nigel Magnay]

I think there is room for improvement with hudson today.  That doesn't
mean controlling it or loosing the community.  It means the opposite
of that.

   -ted

Ted, that's excellent news.

Can I suggest that Oracle now moves pretty quickly to move the Hudson name into the public domain or over to the developer community? (I'd say just an announcement would be sufficient, but sadly as you correctly identify, your team may be totally honorable but the destruction of goodwill caused by the OpenOffice, OpenSolaris and continued failure to honour the TCK agreement with Apache is likely to make people skeptical - but, as you say, Trinidad shows it *can* be done). That way you totally defuse the issue by removing the possibility of communiques being accidentally interpreted as veiled threats, and Oracle both comes on board as a great partner and repairs some of its image with the community. And as you say, Oracle looses nothing by doing this, it's a win / win.

I'm sure as a generous donator of IT services the community is happy that Oracle is our 'platinum sponsor'.

[vivo]

(silent majority +1)

This discussion goes in circles.

@Ted:
1) from your point of view: Who is "empowered" to decide where the
source code should be hosted?
1a) If the active developers are voting to move to Github, why do you
mind?
1b) if they move, why would that be a "fork", wouldn't the code on
java.net rather be the fork?
2) You are claiming to speak for some other people. I am sorry, but
this has no weight in an _open_ source project. Arguments have weight.
2a) Which changes were rejected?
3) I do not know enough about trademarks, but can you tell us where
(when) the name Hudson was registered by whom?

[Conny Kreyßel]

Another silent member.

First of all, i cannot see how does the hudson community have a clear
future when the hudson tm is not in the hand of the community!

Second, Infrastructure: I think your discussion can I reduce to 'how
can we bind the hudson community to kenai'. But in Feb 2010 it was not
clear that Kenai has a bright future [1] and Oracle try today to bind
some key projects. Without resounding success.

I think that we (the community) should combine the best tools to get a
modern distributed ALM [2] Platform to develop hudson:

Project-, Task-, Change-, Release & Test Management: JIRA (or GitHub
Issue-Feature, but JIRA is real smarter)
Documentation & Knowledge Management: Confluence (or GitHub Wiki - i
dont know the feature set)
Community Communication: Mailing Lists on Google Groups (GitHub has no
mailing list feature?)
Version Control Systems: Git on GitHub (real distributed and easy
contribution!)
Build & Test: Maven
Continuous Integration: Hudson :-)
Artifact Repository: own Sonatype Nexus Instance or deploy to
oss.sonatype.org [3]

And that all should be available under one second level domain (like
hudson-ci.org is) to be clear to any out there that he access a real
hudson page.

I think thats all and if we can setup and manage such infrastructure,
we are able to move strong forward.


[1] - http://www.h-online.com/open/news/item/Confusion-over-Sun-s-Kenai-hosting-platform-924331.html
[2] - http://en.wikipedia.org/wiki/Application_lifecycle_management
[3] - https://docs.sonatype.org/display/Repository/Sonatype+OSS+Maven+Repository+Usage+Guide

[Jesse Glick]

To drift even further off topic:

On Nov 30, 4:47 pm, Henrik Lynggaard <henrik.lyngga...@gmail.com>
wrote:
> I use NetBeans daily but it has no git support [...]. I tried to download NbGit

Official Git support is under development for 7.0.

[Jesse Glick]

On Nov 29, 1:27 pm, jvanzyl <jason.van...@gmail.com> wrote:
> Although I prefer github and do Maven work there (I, in fact, hate
> working with SVN at this point)

Same here. Trying to do a simple feature branch in SVN is quite error-
prone, and waiting for a slow server just to get file history is
unpleasant.

> the canonical repository is at Apache
> for the official Apache Maven releases.

Note that with a DVCS such as Git it is simple and safe to maintain an
official repository which is _mostly_ a mirror of actively changing
repositories elsewhere such as GitHub; bidirectional synchronization
is trivial. The "official" designation is a matter of convention which
can be changed without interrupting ongoing work. So while decisions
about which VCS to use, and (in the case of a DVCS) how
subrepositories should be organized, should be made carefully since
they are hard to change later, the decision about where a DVCS should
be "hosted" can be made more lightly.

[Kohsuke Kawaguchi]

I'd also like to call the attention to the silent changes to the terms
of use of the Wiki and the issue tracker.

For example, if you see the footer of [1] and click "Terms of Use",
you see a link to [2]. This used to redirect to [3], but now it goes
to [4].

And this drastically changes the way the submitted contents are handled.
In its original ToU of [3], it was a very open license:

> 4. CONTENT SUBMITTED TO SUN
> 4.1 Sun does not claim ownership of the Content You place on the Website ...,
> and You grant Sun and all other users of the Website an irrevocable, worldwide,
> royalty-free, nonexclusive license to use, reproduce, modify, distribute, transmit,
> display, perform, adapt, resell and publish such Content (including in digital form).

But in the new ToU [4], such rights are only granted to Oracle and not
"all other users of the Website".

That is to say, the community has no rights to the contents it
created. I don't think this makes sense for the Hudson community, and
I sincerely hope that this is unintentional and the correction will be
made, perhaps to the ToU of the java.net [5].

[1] http://wiki.hudson-ci.org/display/HUDSON/Home
[2] http://www.sun.com/share/text/termsofuse.html
[3] http://web.archive.org/web/20080822035452/www.sun.com/termsofuse.jsp
[4] http://www.oracle.com/us/legal/terms/index.html
[5] http://home.java.net/javanet-web-site-terms-use

2010/12/1 Nigel Magnay <nigel....@gmail.com>:

--
Kohsuke Kawaguchi

[Bernard Wilson]

>
> That is to say, the community has no rights to the contents it
> created. I don't think this makes sense for the Hudson community, and
> I sincerely hope that this is unintentional and the correction will be
> made, perhaps to the ToU of the java.net [5].
>

All your bugs, are belong to us.

[Andrew Bayer]

That terms of use change is very, very bad - Ted, can you comment on this?

A.

[Thomas Fürer]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

that is exactly what I would say, thanks Henrik ...

+++ don't fragment the hudson base
+++ use JIRA as before, because it is the best issue tracking tool ever
(IMHO)
+ try to get Confluence for the wiki

o for me SVN works fine - no need to change something

- --- stop talking about how is the bad one...

just my 50 cents

tom

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJM9oIFAAoJEMIrIJt+IWii9YMIAMr1XmSXJ8ShUl3b7wP7pVf7
EPC0Wpgs8XGFD9nVMHCYfweAy/XAyMLRXy9kI3gG0TyEYYYgMvc3fq1wa6aMo2sA
oU9+zX5y8K1dAhNDpZTOunwU5sYCey62SG5Ftm61oc4YedZPGagGgTk3VB2HwvqC
owi65opkNg6cwtRjF+r0W7PtTjfZ4auu4KVL0gYN9b++0BEhoChJUZO36xo8GHVJ
Wlc75nE6Uv0EqgsVmK1H370jSP6acxLF14gdHQOoQFLhQ8g2JF+5Efm47pfkQ9ks
/Qw0nHYkKUARvMhtCNw9pO2NejvK++UYps05eAYURHrE8hIDM4UgQJ9b/RkyinY=
=NFtj
-----END PGP SIGNATURE-----

[Andrew Bayer]

It now sounds like that's just a broken link on java.net - the java.net terms of use are at http://java.net/terms_of_use and are far more acceptable.

A.

[Conny Kreyßel]

I have not found any hudson trademark, that oracle have registered.

Look at http://tess2.uspto.gov/ (Owner name == Oracle)

Based on what rights can oracle own hudson?

[pelegri]

I pointed this out to Justin and he says it is a redirect bug; the ToU
for Java.Net has not changed; see [10], and that this will get fixed
ASAP.

I saw some of the sausage-making on the GlassFish migration side, so I
have no reason to believe that is not the case.

- eduard/o

[10] http://twitter.com/oracletechnet/status/10014701714939905

On Dec 1, 7:16 am, Kohsuke Kawaguchi <k...@kohsuke.org> wrote:
> I'd also like to call the attention to the silent changes to the terms
> of use of the Wiki and the issue tracker.
>
> For example, if you see the footer of [1] and click "Terms of Use",
> you see a link to [2]. This used to redirect to [3], but now it goes
> to [4].
>
> And this drastically changes the way the submitted contents are handled.
> In its original ToU of [3], it was a very open license:
>
> > 4. CONTENT SUBMITTED TO SUN
> > 4.1 Sun does not claim ownership of the Content You place on the Website ...,
> > and You grant Sun and all other users of the Website an irrevocable, worldwide,
> > royalty-free, nonexclusive license to use, reproduce, modify, distribute, transmit,
> > display, perform, adapt, resell and publish such Content (including in digital form).
>
> But in the new ToU [4], such rights are only granted to Oracle and not
> "all other users of the Website".
>
> That is to say, the community has no rights to the contents it
> created. I don't think this makes sense for the Hudson community, and
> I sincerely hope that this is unintentional and the correction will be
> made, perhaps to the ToU of the java.net [5].
>
> [1]http://wiki.hudson-ci.org/display/HUDSON/Home
> [2]http://www.sun.com/share/text/termsofuse.html
> [3]http://web.archive.org/web/20080822035452/www.sun.com/termsofuse.jsp
> [4]http://www.oracle.com/us/legal/terms/index.html
> [5]http://home.java.net/javanet-web-site-terms-use
>

> 2010/12/1 Nigel Magnay <nigel.mag...@gmail.com>:

[pelegri]

((as a reminder/introduction - I used to work at Sun and was KK's manager at
the end his tenure there. When Sun was acquired by Oracle, the GlassFish
and the Hudson groups went to different groups at Oracle and I moved with
the GlassFish team. I recently left Oracle))

A while ago, I had proposed to KK to keep core and plugins in two separate
(Java.Net) repositories because core requires SCA (OCA now), while plugins
does not. KK countered with the benefits of a smooth transition between
"plugin developer" and "core developer". At the end we didn't change
anythng because because Hudson was the most successful of Sun's projects in
terms of participation and I though I couldn't/ shouldn't argue w/ success.

In the current situation, we could treat each of these differently. It
would be ideal to keep the two in the same location, but, in principle they
could even go to the different forges. Not arguing for this (yet?), but
pointing it to encourage brainstorming.

In any case, one benefit of Oracle keeping a strong relationship w/ Hudson
core is that Oracle could defend the IP, if necessary. Given that Plugins
do not require SCA/OCA, that does not apply there.
--
View this message in context: http://hudson.361315.n4.nabble.com/hudson-java-net-is-alive-tp3058272p3067972.html
Sent from the Hudson users mailing list archive at Nabble.com.

[Bernard Wilson]

> In any case, one benefit of Oracle keeping a strong relationship w/ Hudson
> core is that Oracle could defend the IP, if necessary. Given that Plugins
> do not require SCA/OCA, that does not apply there.

Except for the fact that Oracle's association would make the (already infinitesimally tiny) chances of an IP shakedown by a 3rd party *more* likely, since Oracle has nice deep pockets.

In any case, Oracle distributes Hudson *regardless* or not of whether it owns it, it's liable for potential action (viz: SCO sued IBM, not the linux community). I'm sure that motivates /them/ towards cumbersome eclipse-style anal-retentiveness, but they can hardly act surprised if the developer community at large cares not one jot for it.

Linux does not require such SCA/OCA encumberance. A simple "Signed-Off-By:" trail in the commit logs (as agreeing to the terms in the contributors file) is sufficient for the OSDL lawyers, and keeps the barrier to entry low. Jeez, there's much bigger projects than Hudson with murkier IPR histories out there cracking along just fine.

[pelegri]

Bernard Wilson wrote:
>
>
> In any case, Oracle distributes Hudson *regardless* or not of whether it
> owns it, it's liable for potential action (viz: SCO sued IBM, not the
> linux community).
>
>

That's a good point (IANAL, etc, etc). I was seeing it more from the
perspective of protection for the little guys, like CloudBees, but your
argument seems reasonable.
--
View this message in context: http://hudson.361315.n4.nabble.com/hudson-java-net-is-alive-tp3058272p3068080.html

[Schloida Schmitt]

What the... ?

On 29 nov, 17:16, ted <ted...@gmail.com> wrote:
> Andrew, Tyler, if you have a list of the github features that you guys
> use the most that wouldn't be available with the kenai version, that
> would be helpful to get.

So you have a list of things Github has that Kenai does not? Easy,
Kenai does not have Github. Case closed.

[Schloida Schmitt]

On 29 nov, 17:16, ted <ted...@gmail.com> wrote:

> Nigel, what I am saying is that I believe the *final* decision of what
> to do w.r.t. infrastructure belongs to Oracle and that decision should
> be made according to the will of the community as it makes sense.

Hmmm... it sounds like a contradiction to me.

[Richard Bywater]

I'm sorry but I feel compelled to respond to this comment. Comments
like this aren't really very helpful IMHO.

GitHub obviously has a set of features which are useful when using Git
and Ted is simply asking for what features GitHub has that would be
useful to include with the new java.net infrastructure.

Comments like this simply drag on the "hostilities"...

Richard.

[Schloida Schmitt]

On 29 nov, 15:17, ted <ted...@gmail.com> wrote:
> I think this thread is a bit out of control.

No, but it seems Oracle is trying to get control of it.


> Oracle has been talking
> to a lot of the users of hudson who are not part of this very small
> core group on the forums.

Where are they? Do they have their own discussion list?


> Susan's job is to represent them until they
> feel the urge to reply directly to the mailing lists or forums.

So he's something like a senator that speaks in their name, is that
it?


> One of their biggest complaints is the lack of formal solicitation of
> comments/requirements before changes like this are made, and a lot of
> frustration around the inability to get their changed requests into
> the core of hudson.

Explain how formal solicitations should be done. With paper forms? A
discussion list like this one is enough.


> You all might not be aware of this, but the actual hudson user base is
> very large.  Much bigger than what you see on the mailing lists or in
> the forums.  The unfortunate part of that is how many of these users
> do not contribute to the core, and do not participate in these
> discussions.

Well, they should.


> They want to do that, but don't feel like they can be
> heard.

What's the deal? There's nothing they should be afraid of. We don't
bite.

Dude, even reading the name 'Oracle' hurts my brain these days.

[Bernard Wilson]

http://www.theregister.co.uk/2010/12/03/oracle_doesnt_own_hudson/

Oracle claimed that it acquired the Hudson trademark with its purchase of Sun Microsystems. But a well-placed former Sun Microsystems employee has contacted The Reg to say that Sun took an "explicit decision" not to apply for a trademark on the name Hudson. A search of the US Patent and Trademark Office's website throws up 623 trademarks for Hudson for many things, but not for the project Oracle owns.

...

Though the company does not in fact own the trademark, Oracle US recently applied for a trademark in the European Union. A search here reveals that Oracle applied for a trademark on October 29 of this year, just before Hudson users began forking the service by moving it off Oracle-owned hosting servers. It doesn't seem that the EU has granted Oracle the trademark.

So, Ted claims that they have 'nothing to gain' by controlling the Hudson community, and yet behind the scenes (NOT in front of the community), Oracle is busy trademarking assets (that Sun had explicitly decided not to), pushing forward employees as co-owners (at the same time as advancing those trademarks), and threatening to veto infrastructure moves they don't like.

That's not controlling?

No wonder they've not agreed to donating the name to the community. They don't even own it yet - despite Ted explicitly claiming that they did. 

Elephant in the room time guys; Oracle is institutionally incapable of playing an equal partner role. It demonstrates this time and again. It's got an agenda here, that much is clear.

If they truly have nothing to gain, then I'm sure they'll be happy to contribute to a Hudson fork.

[ymen...@gmail.com]

Yup that article is correct. Application nb 009482902 on the
2010-10-29.

Note that there is a process to oppose a trademark registration
(http://oami.europa.eu/ows/rw/pages/CTM/regProcess/opposition.en.do),
within the first three months after registration.

On Dec 3, 9:29 am, Bernard Wilson <bernard.wil...@live.com> wrote:
> http://www.theregister.co.uk/2010/12/03/oracle_doesnt_own_hudson/
>
> Oracle claimed that it acquired the Hudson trademark with its purchase of Sun Microsystems. But a well-placed former Sun Microsystems employee has contacted The Reg to say that Sun took an "explicit decision" not to apply for a trademark on the name Hudson. A search of the US Patent and Trademark Office's website throws up 623 trademarks for Hudson for many things, but not for the project Oracle owns.

> ...Though the company does not in fact own the trademark, Oracle US recently applied for a trademark in the European Union. A search here reveals that Oracle applied for a trademark on October 29 of this year, just before Hudson users began forking the service by moving it off Oracle-owned hosting servers. It doesn't seem that the EU has granted Oracle the trademark.

[Schloida Schmitt]

I thought I'd sense hostility in Ted's previous posts, maybe I had the
wrong impression.

On 3 dez, 01:24, Richard Bywater <rbywa...@gmail.com> wrote:
> I'm sorry but I feel compelled to respond to this comment. Comments
> like this aren't really very helpful IMHO.
>
> GitHub obviously has a set of features which are useful when using Git
> and Ted is simply asking for what features GitHub has that would be
> useful to include with the new java.net infrastructure.
>
> Comments like this simply drag on the "hostilities"...
>
> Richard.
>