Yoav Shapira wrote:
Hi,
I'm missing something fundamental. What would a JSR Expert Group
have to do with this? We are talking about the API jars for
completed JSRs, right, and maybe other specs if there are any that
require similar machinations? (I can't think of any...)
I raised the EG commit point for cases of completed JSRs that later
have corrections or errata, or simply things like typos in the
JavaDocs or even implementation bugs in the examples if any are
included with the APIs themselves. We've seen this happen every now
and then (not often at all, but it does happen, maybe once per year)
with the Servlet and JSP APIs in the Tomcat world.
Sure - but in that case, the EG should/is required to do an update on
the spec itself, that would be somehow labeled, and we'd update from
there, right?
It would be great to get EGs to come fix our code, but I don't think
that would happen except in rare cases, and those would be just because
we happen to have a community member on the EG.
For a long time the relevant CVS modules (jakarta-servletapi,
jakarta-servletapi-5) only granted commit access to Servlet/JSP EG
members, not all Tomcat committers, and if some of the EG members were
busy elsewhere, typos and corrections to the spec stuff would take
forever to fix. This is unfortunate because when these rare things do
occur, they are a high-visibility issue that should be addressed
immediately.
Right. Agreed. Maybe we don't have external EG-only commit rights in
the future ;) ?
(The situation is now better for these specific APIs since we moved to
SVN and I believe the entire Tomcat PMC may have commit privilges to
the tomcat/servletapi tree, but I'd like to ensure it doesn't happen
in this new specs project).
I see. Thanks.
geir
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
