Sonatype to stick with Hudson

[Morten A-Gott]

Jason van Zyl's response to the Hudson/Jenkins debacle:
https://gist.github.com/805941

Please! Won't somebody please think of the children!!!

[Cédric Beust ♔]

On Tue, Feb 1, 2011 at 7:13 AM, Morten A-Gott <morten.an...@gmail.com> wrote:

Jason van Zyl's response to the Hudson/Jenkins debacle:
https://gist.github.com/805941

Not really surprising. I wish Kohsuke the best but I think that at this point, the Hudson brand has too much mindshare for Jenkins to be successful.

--
Cédric

[Fabrizio Giudici]

On 02/01/2011 04:47 PM, Cédric Beust ♔ wrote:
>
>
> On Tue, Feb 1, 2011 at 7:13 AM, Morten A-Gott

> <morten.an...@gmail.com <mailto:morten.an...@gmail.com>>

> wrote:
>
> Jason van Zyl's response to the Hudson/Jenkins debacle:
> https://gist.github.com/805941
>
>
> Not really surprising. I wish Kohsuke the best but I think that at
> this point, the Hudson brand has too much mindshare for Jenkins to be
> successful.
>

Well, I'm not sure. Being a pretty-geeky kind of products, Kohsuke has
got a great visibility. For myself, I still don't know what I'll do.
I've taken the habit to upgrade once in a few months (unless something
immediate is needed) and I'll wait for a bit to see where the wind will
blog...

--
Fabrizio Giudici - Java Architect, Project Manager
Tidalwave s.a.s. - "We make Java work. Everywhere."
java.net/blog/fabriziogiudici - www.tidalwave.it/people
Fabrizio...@tidalwave.it

[Fabrizio Giudici]

On 02/01/2011 06:53 PM, Fabrizio Giudici wrote:
>
> Well, I'm not sure. Being a pretty-geeky kind of products, Kohsuke has
> got a great visibility. For myself, I still don't know what I'll do.
> I've taken the habit to upgrade once in a few months (unless something
> immediate is needed) and I'll wait for a bit to see where the wind
> will blog...
>

Oh my God! I meant where the wind will bloW :-O

[phil swenson]

It's not going to be hard to move to Jenkins.  Wait until it's stable and move.  Hudson will wither as Oracle's implementation stagnates and developers quit updating their plugins.  That's my take.

IMO Oracle doesn't understand the development community.  Especially free/open source.  They understand selling giant, crappy enterprise products and little else.

--
You received this message because you are subscribed to the Google Groups "The Java Posse" group.
To post to this group, send email to java...@googlegroups.com.
To unsubscribe from this group, send email to javaposse+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.

[Cédric Beust ♔]

On Tue, Feb 1, 2011 at 9:53 AM, Fabrizio Giudici <fabrizio...@tidalwave.it> wrote:

   Jason van Zyl's response to the Hudson/Jenkins debacle:
   https://gist.github.com/805941


Not really surprising. I wish Kohsuke the best but I think that at this point, the Hudson brand has too much mindshare for Jenkins to be successful.

Well, I'm not sure. Being a pretty-geeky kind of products, Kohsuke has got a great visibility

Pure speculation on my part, but I suspect that 95% of Hudson users have no idea who Kohsuke is nor do they know anything about the fork.

-- 

Cédric

[phil swenson]

RT @tom_enebo: Notice a trend in this Hudson-themed thread? http://bit.ly/dLmCJg.

2011/2/1 Cédric Beust ♔ <ced...@beust.com>

[Cédric Beust ♔]

On Tue, Feb 1, 2011 at 2:46 PM, phil swenson <phil.s...@gmail.com> wrote:

RT @tom_enebo: Notice a trend in this Hudson-themed thread? http://bit.ly/dLmCJg.

Ten people unsubscribed from a 7000+ person mailing list. See what I mean?

Let's see how many subscribers the Hudson list has in a month, it will be a better indicator.

--
Cédric

[Thomas Ferris Nicolaisen]

And the other 5% are us who actually set up Hudson installations and maintain them for the rest of the organization.

The reason I've pushed Hudson so much over the last years is that it such an easy product to use. It has an amazingly high "just-works" factor. You can drop it right in, with practically zero upgrade-instructions.

I think this has a lot to do with Koshuke's way of working. If something is hard or doesn't work out of the box, he fixes it fast, always focusing on ease-of-use for the majority.

Regarding the problems that do sneak in, I think we have well-educated patience with the Hudson/Jenkins release rhythm. Could be that 1.396 doesn't work for me, or some plugin isn't working, then I just wait a couple of days for the next release. Maybe we send a mail to the list and try nailing the problem, cause we know that Koshuke will actually have a look at it, and help or fix within hours, days tops.

I'm not sure if Oracle can follow this rhythm, or even keep the product so easy to use/upgrade, even with Sonatype trying to push them in the right direction. Maybe they'll even introduce an obligatory "Register this product at Oracle.com" step during installation, or some other "user friendly" stuff :)

So my finger is on the "Hudson will stagnate, and users will gradually migrate to Jenkins" scenario, like Phil says.

[Manfred Moser]

This interesting insight popped up on the jugleaders mailing list and was posted by Stephan Jansen from BEJUG/parleys.com and I think it should be shared and bring some light into things as well.

> Guys,
>
> During an IOUC session last week, Ted Farrell debriefed us on his view
> on the Hudson debacle (too bad it wasn't recorded).
> But basically it came to this (and please correct me if I'm wrong):
>
> Kohsuke Kawaguchi (aka KK) has worked for many years on Hudson during
> business hours paid by Sun Microsystems.
> KK feels that he is Mr. Hudson and could make money with the project
> by taking "his" baby and the community over to CloudBees.  The VC
> behind Cloudbess even placed some pressure on different parties
> involved to make this happen!
>
> It's obvious that Oracle wouldn't allow this to happen.
>
> Now the fork has happened and both parties can now compete on
> different implementations.  Oracle (fyi - Ted Farrell is chief
> architect for Tools and Middleware at Oracle) is fully committed to
> continue the investment in Hudson (see also
> http://wiki.hudson-ci.org/display/HUDSON/Home)
>
> There's always two sides to a story and when money and pride is
> involved it can get very ugly.
>
> Cheers,
>
> Stephan

[Joe Sondow]

Certainly there's money and politics involved, but what makes Hudson
valuable in the first place is Kohsuke's philosophy towards ease of
use, rapid improvement, and extensibility, as other people have stated
in this thread. CruiseControl was a great idea when it came out, but
it was all but forgotten a few years after Hudson showed how
continuous integration could be done far better.

If Ted Farrell is the new head of Hudson, check out his track record
on his executive bio:
http://www.oracle.com/us/corporate/press/Spokespeople/016474
"Mr. Farrell is responsible for the technical and strategic direction
of Oracle's frameworks and development tools products, including
Oracle JDeveloper, Oracle Application Development Framework (ADF),
Oracle Metadata Services (MDS) and Oracle WebCenter Framework."

I haven't heard much interest from Java developers in those products,
compared with interest in competing products. I think over the next
few years Hudson will evolve into something that aligns better with
Oracle's strategic objectives. Maybe it will integrate with other
Oracle products. What it probably won't focus on is following the
evolving needs of engineers. Consider Oracle's decision to remove Ruby
support from NetBeans, perhaps because Oracle has no strategic plan to
monetize Ruby. What good parts of Hudson will Oracle consider
unprofitable?

On another note, the open issues list for Jenkins is an interesting
read:
http://issues.jenkins-ci.org/secure/IssueNavigator.jspa?mode=hide&requestId=10230

[Michael Neale]

yeah it is unfortunate, the brand is powerful. (I work with Kohsuke @
cloudbees).

Who knows if it will really take off, it is interesting to watch. The
brand pull is powerful.

If Jenkins takes off and hudson withers - shows power of community
against oracle. Cool.

If Hudson powers on, and everyone ends up using that again, who knows
what it mean to users (assuming it stays open source, and things don't
creep in as oracle add-ons).

What it does mean for open source, is that who owns "the brand" (ie
name trademarks etc) is kind of important. At least in the short term
of a project. It has made me think that open source projects where the
name is owned by a non foundation as "risky" (ie what if that name
ends up in the hands of someone.. you know... like hank scorpio or dr
evil). Projects which have TM owned by a foundation (eg tomcat -
apache, jetty - eclipse, for example) are much safer (at least in a
free community sense).

Interesting to watch.



On Feb 2, 2:13 am, Morten A-Gott <morten.anderseng...@gmail.com>
wrote:

[Manfred Moser]

Jason van Zyl shines a bit more light on this whole disaster. Interesting insights in terms of code quality, IP, process and stuff.

http://www.sonatype.com/people/2011/02/hudsons-bright-future/

Jelly and forked artifacts certainly bring up some bad memories for me and that does not exactly provide confidence using Jenkins.. 

manfred