Mailing-List: contact general-help@logging.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Logging General" <general@logging.apache.org>
Received-SPF: pass (hermes.apache.org: domain of nicko@neoworks.com designates
 80.168.17.114 as permitted sender)
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [VOTE] Graduation of log4cxx and log4net
Date: Fri, 26 Nov 2004 23:01:12 -0000
Message-ID: <DDEB64C8619AC64DBC074208B046611C1FFE45@kronos.neoworks.co.uk>
Thread-Topic: [VOTE] Graduation of log4cxx and log4net
Thread-Index: AcTTm7pkXIyI0lptSfWHabAlgnGtuAAbJ+bA
From: "Nicko Cadell" <nicko@neoworks.com>
To: "Logging General" <general@logging.apache.org>

=20

> Indeed,  I  am  certain   that  all  those  involved  appreciate  your
> frankness.  At this stage,  in order  to build  a wider =20
> community, my advice to  you would be  actively seek ways  in=20
> which to  enlarge your committer base. On  way to get there,=20
> is to  grant committer status to
> those developers who  show that they have got a clue.   Note that I am
> not  suggesting to  get nonchalant  about it,  but only =20
> slightly less conservative. It  boils down  to keeping the =20
> clueless out  and giving those who show promise a  chance.=20
> Compare  this with the HTTPD project
> which  is   reputed  to  grant  comittership  status   only  to  those
> contributors who show consistent  and real committment to the=20
> project, say for at least 6 months.

So long as we can maintain quality then the more committers the better.
I will shake the tree and see who falls out.


=20
> Regarding to jira/bugzilla or not to jira/bugzilla question,=20
> I believe that the  decision regarding such a  technical=20
> matter is  best left to the people  actually doing  the work.=20
> I  am fairly confident  that the Board would sustain this=20
> view, if it ever came to that.

I haven't used jira at all, but it sounds like jira may be a good fit. I
will investigate the log4cxx jira.


=20
> Keep in mind  that there must be a convenient  way for=20
> contributors to submit patches.  The log4net@  lists=20
> apparently strip attachments; not your fault but mine.=20

The lists allow plain text attachments but strip binaries including
ZIPs.=20
This, I think, is a good thing.


> Having=20
> contributors post their patches at SF does not  come through =20
> very  well.  You  should  consider completing  your migration=20
> away  from SourceForge. The  mailing lists should  be closed=20
> down, for example by refusing  new posts and new=20
> subscriptions. Remove all file releases at SF. (Even under=20
> incubation, you can distribute a "snapshot"  release  here at=20
>  Apache.)  Close  down  the bug  tracking system, the  CVS=20
> repository,  etc. In short,  there should  be nothing left =20
> at http://sourceforge.net/projects/log4net/  except  pointers=20
> to http://l.a.o/log4net/.

The sourceforge lists require moderator approval and now have almost no
traffic (apart from spam ;) I should probably close them up now. The SF
site is only used to provide the downloads of the old versions of
log4net. We are planning our first release from the apache codebase
(ensuring the documentation is in sync with the code is the main
remaining item) and when that happens we will certainly not continue to
link to the SF site.

I am slightly wary about eradicating the old versions entirely. I know
they can't be served by apache. Is it still possible to download a
version of log4j that is pre-Apache?


> Given that  you have the  formal backing of  this PMC, I=20
> think  we can submit  our proposal  to  graduate log4net as =20
> soon as it becomes a no brainier case.  I expect we can reach=20
> that stage by early next year.

Sounds good to me.
Thanks for your support.

------------
Nicko Cadell
log4net development
http://logging.apache.org/log4net