www-jcp-open mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Henri Yandell" <bay...@apache.org>
Subject Re: [VOTE] New ASF/JCP Policies
Date Sat, 14 Jul 2007 09:10:42 GMT
On 7/13/07, Dain Sundstrom <dain@iq80.com> wrote:

> Hummm, it doesn't seem hard line to me to stop running TCKs, but I
> guess that is a difference of opinion.  I personally don't think
> certification is such a big deal.  This option will have some pain,
> but I don't think it will be nearly as painful as some of the others
> on the table.

It also has benefits.

* No waiting on the TCK to get a release out. My suggestion is that
TCKs hurt our release-early/release-often desire and so are a
detriment to our communities.

* No "the test is in the TCK, why write a test" mentality. This exists
on some projects for which there is a TCK, and means we spend less
effort on thinking about testing with the assumption that the TCK will
cover it. A detriment to our code.

* No hush hush/secret stuff. This diminishes the community who can
discuss the tests.

* No suddenly dead maintenance projects. It seems that the pain of
having to do a TCK can turn a project from active to inactive very
quickly; whereas without the TCK it would slowly wind down with bugfix
releases going out.

I want to do a Jakarta Standard Taglib release soon (JSTL spec). If
someone can tell me that 'Yes, you can release without the TCK', which
I think we've been doing anyway, then I've no interest in running the


View raw message