cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vadim Gritsenko <>
Subject Re: svn commit: r312968 - in /cocoon/branches/BRANCH_2_1_X: ./ src/java/org/apache/cocoon/components/source/impl/validity/ src/java/org/apache/cocoon/i18n/ src/java/org/apache/cocoon/transformation/ src/webapp/WEB-INF/ src/webapp/samples/i18n/translations/
Date Wed, 12 Oct 2005 01:57:20 GMT
Ralph Goers wrote:
> Vadim Gritsenko wrote:
>> Torsten Curdt wrote:
>>> ...shouldn't this only be changed in trunk instead of in branch?
>> IMHO, if OOMEs and severe usability issues are not fixed in branch, 
>> then trunk won't ever be used by anybody, isn't it?
> There has got to be a typo of some sort in your statement.

No :-)

Expanding a bit on it: if critical production issues are not fixed on the 
branch, then perception is that software isn't ready for production, hence 
shouldn't be used for implementing systems. Instead, something like Struts or 
whatever should be used, where there is a strong support present and any issues 
can be addressed. As a result, next major release might see even less users due 
to our inability to fix issues on current release (Think from user POV: "if they 
aren't fixing them now, what chances are they will fix'em later?").

I don't like to give such perception, and I'd rather have such issues fixed in 
the stable branch, as well as I'd like to grow Cocoon's userbase. Anything that 
helps run Cocoon non-stop for months is a good thing to have, in my book.

Remember, we are not talking about any new features here - rather, fixing bugs & 
increasing stability.

As somebody :-P mentioned on parallel "branch vs trunk" thread, and I'll quote:

     I guess I would say that it is business as usual with 2.1.x until
     2.1.8 is released.  I would then suggest that only very minor
     enhancements and bug fixes be added.  Hopefully, then 2.1.9 could
     be the final release on the branch.

And I agree with that. It is business as usual - fix bugs and what not till we 
have 2.1.8. Hopefully after that we'd have momentum on 2.2 large enough to make 
2.1.9 final release on the branch.


> I'm assuming 
> you meant one or both of the following:
> 1. If we keep adding stuff in the branch noone will use trunk.
> 2. If we don't add them in the branch then we will force people to use 
> trunk.
> Both of these amount to the same thing and I don't agree.  I don't have 
> any projects "on the fence" - i.e. nothing deliverable in the next 
> several months.  If I did I would certainly go with 2.1.8.  OTOH, since 
> my projects will be early next year I am already preparing folks for the 
> transition - upgrading to Maven 2, etc, because I want to take advantage 
> of the packaging improvements as well as all the things that have not 
> been ported back to 2.1.
> I'll use Maven as an example. Maven 1 is still in strong use with some 
> 1.x work still going on - yet here we are focusing in on Maven 2 even 
> though it still is in beta. Why? Because we want the features.
> In other words, the new features in 2.2 and the fact that we are 
> formalizing its release will, IMO, be enough to get people to switch.  
> If they aren't then we have completely missed the mark.
> Ralph

View raw message