You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by "William A. Rowe, Jr." <wr...@rowe-clan.net> on 2007/04/04 19:02:38 UTC
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Martijn Dashorst wrote:
> Also, the whole idea of the Incubator is to
> withhold releases from the general public.
Just to clarify - I don't think 'withhold' is a good description.
Release - but with no specific expectation of persistence at the
ASF is probably a better description. E.g. "here's code" is fine,
"here's a community" would be premature. Public releases of all
of our incubating projects IS goodness if it attracts more people
to the incubating communities, and increases their chances for a
successful graduation and project lifespan.
Also - folks have turned *another* vote thread into a discussion
thread. *please* change subject lines, that's why some have started
to ignore your requests to vote :-) [VOTE] is a thread that requires
an immediate response and should contain the responses.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by robert burrell donkin <ro...@gmail.com>.
On 4/6/07, Niclas Hedhman <ni...@hedhman.org> wrote:
> On Friday 06 April 2007 15:55, robert burrell donkin wrote:
> > most critical issues i encounter are issues with source. the source
> > can be checked at any time. potential issues with source should be
> > addressed as soon as possible. (and yes, i know henri arrived here
> > long before me.)
> >
> > the best approach would be to run RAT on the trunks of all incubating
> > podlings.
>
> So for Maven built projects, this should go into master POM file to be
> included automatically...
stefan developed an ant task (which ships with RAT) and jochen a maven 2 one
the problem is that the reports are hard to interpret ATM
RAT needs to read detached audit meta-data before it can conclusively
work out whether an anomaly is a real issue or not but doesn't ATM
- rboert
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Niclas Hedhman <ni...@hedhman.org>.
On Friday 06 April 2007 15:55, robert burrell donkin wrote:
> most critical issues i encounter are issues with source. the source
> can be checked at any time. potential issues with source should be
> addressed as soon as possible. (and yes, i know henri arrived here
> long before me.)
>
> the best approach would be to run RAT on the trunks of all incubating
> podlings.
So for Maven built projects, this should go into master POM file to be
included automatically...
Cheers
--
Niclas Hedhman, Software Developer
I live here; http://tinyurl.com/2qq9er
I work here; http://tinyurl.com/2ymelc
I relax here; http://tinyurl.com/2cgsug
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Niclas Hedhman <ni...@hedhman.org>.
[Should be off-line, but I think we programmer's need to be pointed into some
cultural enlightenment every now and then to broaden our horizons.]
On Friday 06 April 2007 11:44, Craig L Russell wrote:
> But please send a link to the meaning of kreti & pleti! I love
> learning new idioms and it's not in the first 20 Google hits. :-(
Hmmm... I think it is a germanic expression, still used in Sweden. I found
this (in swedish)
ETYMOLOGI: jfr d. kreti og pleti, t. krethi und plethi, ytterst av hebreiskt
ursprung, förekommande i ä. bibelöversättningar som benämning på två folkslag
i konung Davids livvakt, keretéer o. peletéer. I vardagligt bruk numera en
föraktlig beteckning för allehanda löst folk, personer utan bildning l.
samhällsställning, den breda allmänheten, vem som helst
Free translation; ultimately of hebrew origin, existing in older Bible
translations as definition of two groups of people in king David's life
guard, keretéer and peletéer. In every day use it is now a disrespectful term
for all kinds of lesser people, people without education, low status in
society, the broad public, anybody.
I found this fairly interesting email;
http://www.domainofman.com/forum/index.cgi?noframes;read=1383
which actually discusses a much larger scope, but still covers the term, and
the uncertainties around it.
Cheers
--
Niclas Hedhman, Software Developer
I live here; http://tinyurl.com/2qq9er
I work here; http://tinyurl.com/2ymelc
I relax here; http://tinyurl.com/2cgsug
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Craig L Russell <Cr...@Sun.COM>.
Hi Niclas,
On Apr 4, 2007, at 8:52 PM, Niclas Hedhman wrote:
> On Thursday 05 April 2007 05:08, Craig L Russell wrote:
>> I have a hard time understanding how a podling can consider itself
>> ready for graduation without having anything worth sharing, and don't
>> understand what the point is of having an audit of anything short of
>> a build artifact.
>
> Assuming for a second that you are not speaking of Wicket;
>
> A podling is ready to graduate when the community operates
> TheApacheWay and
> has cleared the legal hurdles, not when it has something that can
> be used by
> kreti & pleti, the general public.
I was very very careful to use the words "without having anything
worth sharing" by which I do not mean the hoi polloi. Sharing work
efforts doesn't mean shrink-wrapped.
But please send a link to the meaning of kreti & pleti! I love
learning new idioms and it's not in the first 20 Google hits. :-(
> Often the "usefulness" comes before the community and legalities,
> but from the
> Incubator's point of view, it is not a concern.
I agree. The Incubator is responsible for guiding the podling in the
Apache Way and not for passing judgement on the usefulness of the
podlings' efforts. That's best judged by the community.
>
> So, I would claim that it is definately a positive sign that a
> podling is able
> to cut a 'non-functional' release, or a 'functional' release that
> explicitly
> will not be supported, future-proof and having other qualities
> typically
> expected from Apache projects.
I agree. That's how I understand what the "incubating" disclaimers
are all about. "Take a look if you like, use it if you choose, but
don't assume it's going to be any better than it is right now."
But I was specifically objecting to a proposal that a release audit
would be a good addition to the incubator process:
>> IMHO we need to alter the process so that we have an explicit audit
>> when the community feels (by vote) that it has a build process in
>> place. the code doesn't need to be ready but the build does and the
>> codebase needs to be ready for audit.
>>
>> - robert
I'm wondering if there really is a need for YAP (yet another process)
to handle a release by a podling. There are lots of ways of labeling
a podling release, and you can indicate whether you want folks to try
it out or just look at its release artifacts. But it still has to
look like a release with disclaimers, licenses, and ip clearances
completed.
I think the incubator is really more concerned about a podling that
appears to claim it's a real Apache project when it's still just in
incubation.
Speaking specifically of Wicket now, I think it's great that the
community has progressed to the point of wanting to demonstrate that
they know how to release stuff, even though it isn't ready for
general use. And I think that 1.3.0-incubating-alpha has enough
disclaimers in it to scare off most casual prospective users.
For what it's worth,
+1 for going through the release process for Wicket with the
disclaimers that the Wicket team has discussed... It looks like they
are "getting it".
Regards,
Craig
>
>
> Cheers
> --
> Niclas Hedhman, Software Developer
>
> I live here; http://tinyurl.com/2qq9er
> I work here; http://tinyurl.com/2ymelc
> I relax here; http://tinyurl.com/2cgsug
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
Craig Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:Craig.Russell@sun.com
P.S. A good JDO? O, Gasp!
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Niclas Hedhman <ni...@hedhman.org>.
On Thursday 05 April 2007 05:08, Craig L Russell wrote:
> I have a hard time understanding how a podling can consider itself
> ready for graduation without having anything worth sharing, and don't
> understand what the point is of having an audit of anything short of
> a build artifact.
Assuming for a second that you are not speaking of Wicket;
A podling is ready to graduate when the community operates TheApacheWay and
has cleared the legal hurdles, not when it has something that can be used by
kreti & pleti, the general public.
Often the "usefulness" comes before the community and legalities, but from the
Incubator's point of view, it is not a concern.
So, I would claim that it is definately a positive sign that a podling is able
to cut a 'non-functional' release, or a 'functional' release that explicitly
will not be supported, future-proof and having other qualities typically
expected from Apache projects.
Cheers
--
Niclas Hedhman, Software Developer
I live here; http://tinyurl.com/2qq9er
I work here; http://tinyurl.com/2ymelc
I relax here; http://tinyurl.com/2cgsug
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Eelco Hillenius <ee...@gmail.com>.
> I have a hard time understanding how a podling can consider itself
> ready for graduation without having anything worth sharing, and don't
> understand what the point is of having an audit of anything short of
> a build artifact.
Yes, that would be strange :) However, in the case of Wicket, we are
talking about a development branch that is intensively being worked on
by several people. We have a stable release (1.2) out on sourceforge,
but the development release (1.3) is for Apache. Several developers
and quite a few users are using the development branch for their
systems (including production sites), so the code is ready for use
alright. However, we have a couple of API breaking features to be
implemented and we like to have them resolved before we put out
official releases for users. At the same time, we are trying to go
ahead with incubation. For that, we're mostly concerned about having a
repeatable build process with checks etc to ensure it produces
releases that are acceptable for ASF. As we understand it, we submit
builds here to be audited on legal aspects and procedure, not so much
end-user functionality.
Regards,
Eelco
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Johan Compagner <jc...@gmail.com>.
>
> IMHO the current process is fine but needs to be documented better.
> Podlings should be encouraged to release stuff that they think is
> usable outside their small world of developers who check out from svn
> and build from source. The incubator is set up to review and approve
> releases without passing judgement on the usability, just the
> "legality".
>
> I have a hard time understanding how a podling can consider itself
> ready for graduation without having anything worth sharing, and don't
> understand what the point is of having an audit of anything short of
> a build artifact.
Are you now talking in general or specific also for wicket?
because first we don't have a small world of developers we have plenty of
users already
and we already share because we already have real releases of our code base
(1.2)
So this release is really for making a apache approved build of wicket that
all the
legal issues are resolved.
johan
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by robert burrell donkin <ro...@gmail.com>.
On 4/4/07, Craig L Russell <Cr...@sun.com> wrote:
> On Apr 4, 2007, at 1:54 PM, robert burrell donkin wrote:
<snip>
> > IMHO we need to alter the process so that we have an explicit audit
> > when the community feels (by vote) that it has a build process in
> > place. the code doesn't need to be ready but the build does and the
> > codebase needs to be ready for audit.
>
> IMHO the current process is fine but needs to be documented better.
i agree that we need better documentation but i'm really not getting
the cycles ATM. i'd hoped that other people would step up and push it
forward since this is vital work.
> Podlings should be encouraged to release stuff that they think is
> usable outside their small world of developers who check out from svn
> and build from source. The incubator is set up to review and approve
> releases without passing judgement on the usability, just the
> "legality".
i've been approaching this problem from the wrong perspective
most critical issues i encounter are issues with source. the source
can be checked at any time. potential issues with source should be
addressed as soon as possible. (and yes, i know henri arrived here
long before me.)
the best approach would be to run RAT on the trunks of all incubating
podlings. unfortunately this means finding the cycles to add the
required features to RAT and i'm likely to be very short of energy in
the next few months :-/
- robert
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Craig L Russell <Cr...@Sun.COM>.
On Apr 4, 2007, at 1:54 PM, robert burrell donkin wrote:
> On 4/4/07, Thilo Goetz <tw...@gmx.de> wrote:
>> William A. Rowe, Jr. wrote:
>> > Martijn Dashorst wrote:
>> >> Also, the whole idea of the Incubator is to
>> >> withhold releases from the general public.
No, the idea of the incubator is to make sure that a project is
suitable for Apache before claiming to be an Apache project.
>> >
>> > Just to clarify - I don't think 'withhold' is a good description.
>> > Release - but with no specific expectation of persistence at the
>> > ASF is probably a better description. E.g. "here's code" is fine,
>> > "here's a community" would be premature. Public releases of all
>> > of our incubating projects IS goodness if it attracts more people
>> > to the incubating communities, and increases their chances for a
>> > successful graduation and project lifespan.
>>
>> +1. Doing releases to attract new contributors is essential to
>> podlings
>> who need to build a diverse developer community to exit the
>> incubator.
>> Wicket doesn't have that issue :-)
>
> IMHO we need to alter the process so that we have an explicit audit
> when the community feels (by vote) that it has a build process in
> place. the code doesn't need to be ready but the build does and the
> codebase needs to be ready for audit.
IMHO the current process is fine but needs to be documented better.
Podlings should be encouraged to release stuff that they think is
usable outside their small world of developers who check out from svn
and build from source. The incubator is set up to review and approve
releases without passing judgement on the usability, just the
"legality".
I have a hard time understanding how a podling can consider itself
ready for graduation without having anything worth sharing, and don't
understand what the point is of having an audit of anything short of
a build artifact.
Craig
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
Craig Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:Craig.Russell@sun.com
P.S. A good JDO? O, Gasp!
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by robert burrell donkin <ro...@gmail.com>.
On 4/4/07, Thilo Goetz <tw...@gmx.de> wrote:
> William A. Rowe, Jr. wrote:
> > Martijn Dashorst wrote:
> >> Also, the whole idea of the Incubator is to
> >> withhold releases from the general public.
> >
> > Just to clarify - I don't think 'withhold' is a good description.
> > Release - but with no specific expectation of persistence at the
> > ASF is probably a better description. E.g. "here's code" is fine,
> > "here's a community" would be premature. Public releases of all
> > of our incubating projects IS goodness if it attracts more people
> > to the incubating communities, and increases their chances for a
> > successful graduation and project lifespan.
>
> +1. Doing releases to attract new contributors is essential to podlings
> who need to build a diverse developer community to exit the incubator.
> Wicket doesn't have that issue :-)
IMHO we need to alter the process so that we have an explicit audit
when the community feels (by vote) that it has a build process in
place. the code doesn't need to be ready but the build does and the
codebase needs to be ready for audit.
- robert
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: discussion of release of Apache Wicket 1.3.0-incubating-alpha
Posted by Thilo Goetz <tw...@gmx.de>.
William A. Rowe, Jr. wrote:
> Martijn Dashorst wrote:
>> Also, the whole idea of the Incubator is to
>> withhold releases from the general public.
>
> Just to clarify - I don't think 'withhold' is a good description.
> Release - but with no specific expectation of persistence at the
> ASF is probably a better description. E.g. "here's code" is fine,
> "here's a community" would be premature. Public releases of all
> of our incubating projects IS goodness if it attracts more people
> to the incubating communities, and increases their chances for a
> successful graduation and project lifespan.
+1. Doing releases to attract new contributors is essential to podlings
who need to build a diverse developer community to exit the incubator.
Wicket doesn't have that issue :-)
--Thilo
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org