cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tim Larson <>
Subject Re: [Vote] Removing Woody
Date Tue, 09 Mar 2004 20:58:14 GMT
[Oops, let's try this again, this time to the ml]

----- Forwarded message from Tim Larson <> -----

From: Tim Larson <>
To: Stefano Mazzocchi <>
Subject: Re: [Vote] Removing Woody

On Tue, Mar 09, 2004 at 02:56:39PM -0500, Stefano Mazzocchi wrote:
> Reinhard P?tz wrote:
> >variant B: remove Woody after 2.1.6 release
> +1
> we give our users time, tools and help to show that we do want to keep 
> innovating, but we *do* care about long term stability.
> Woody was an alpha block so we have all the rights in the world to 
> remove them under their feet, but since it was a very successful alpha 
> block (and for this reason is becoming the seed for the officially 
> supported Cocoon Forms framework), we make an effort to make the 
> transition as smooth as possible.
> So:
>  1) we created the Cocoon Forms block.
> From now on, CForms will indicate the content of that block. Woody will 
> indicate the content of the woody block that is scheduled to be 
> decommissioned.
>  2) we created tools to ease the migration [we need to put some docs on 
> how to do this]
>  3) we freeze the woody block (this forces development to happen in the 
> cforms block) now and forever. no changes are getting there. it's read 
> only and just for short legacy and migration purposes.
>  4) we schedule to remove this after *two* releases, to give time to 
> everybody to adjust.
> This should make everybody happy: it doesn't hurt to have a frozen 
> block, as long as everybody understands its status [so we need to 
> reflect this in both docs and samples!!]

[Please let me say sorry in advance for the tone of this email.
If you do not mind listening to me vent for a moment, maybe we
can smooth this out.]

Why would we tell our users to "eat bugs" until they are ready to do a
full migration?  Why could we not do minor (from an architectural
standpoint, but not necessarily from the user's view) bug fixes while
we are still holding the block?  I just don't get the point of this.
First we tell them we are breaking all of the interfaces via a rename,
then we tell them that they must choose to migrate right now or get no
bug/security fixes.

In case you are wondering, this is not just a theoreticaly problem.
It was recently discovered that the woody transformer eats element names
under certain circumstances, and I have a fix submitted by a user.
It seems likely that other bugs will also appear over the course of the
two releases you mentioned above.  Maybe this is all a moot point, since
we might have to remove cocoon from the ASF downloads anyway due to the
licensing issues currently floating around.

Like you say above, we have the right to pull the code whenever we
feel like it, but that does not necessarily make it advisable.


Well, that's what I think.  I will of course live with whatever we
decide as a group, and stop grumbling (perhaps) after the decision
is made.  Part of why I am upset right now is I see the potential
for the work I have been doing for the last few months at my current
employment dangling on the edge of being tossed because of the risk
my employer perceives from the fast pace of open source development.
I know if I face this then others do too, and the only way I know
to confront it is to face it down with well conceived development
and maintenance policies that we together closely adhere to.

Sorry for venting, I hope my intentions came across clearly.
--Tim Larson

----- End forwarded message -----

View raw message