Transition to Mallard?

[Phil Bull]

Hi Thomas,

On Wed, 2010-01-20 at 16:51 -0600, Thomas R. Jones wrote:
> > Quite aside from the merits of this discussion, which is definitely
> > interesting, this isn't an attitude that I want to see projected by
> > the documentation team to potential contributors and for that reason I
> > feel compelled to reply to it.
> And neither should shoving mallard down the throats of dissenting  
> opinion. There is no time constraints to speak of to push the "new"  
> project into implementation. What is the need for such a hurried  
> implementation? It's obvious to the list that yourself and a few  
> others are spearheading the development and possible use of mallard.  
> So why can't you and the community allow others to review and comment  
> before a before a decision is made?

A decision certainly hasn't been made yet, and I don't think it's
anyone's intention to shove Mallard down anyone's throat. I don't think
any sort of full implementation is on the table yet. Rather, it would be
interesting to see how porting an existing DocBook document over to
Mallard goes. I think that musicvideophotos.xml (Ubuntu branch) is a
good candidate. This wouldn't be released to users. It would just be for
our testing.

The whole point of this thread was to court opinion on a possible
transition to Mallard. You and a few others have provided some strong
arguments against a transition, and we're discussing them, in what I
hope is a satisfactory manner. At some point, I hope we'll reach a
consensus on the best course of action.

> > I think the next stage for this discussion is probably to simply try a
> > migration to Mallard and see how it goes: there are no strings
> > attached to such a testing branch, and it would enable us to
> > understand better the issues.
> Feel free to proceed with quality testing. Testing be the key word.
> 
> Personally, I am afraid that do to personal preference you and a  
> select few others are leading the ubuntu doc project into peril. Not  
> only will the project be affected but those that contribute behind us.  
> I for one want to make the right decisions and not negatively affect  
> either.

Of course, I want us to make the right decisions too. Part of this
process needs to include the evaluation of new technologies, especially
if a major upstream project is using it, and especially if it could
solve a lot of long-standing problems.

I certainly do have a personal preference for Mallard, which is why I
initiated this discussion. I wanted to see what people thought. No
decisions are going to be made right away, and people on the whole seem
to be favouring the approach of evaluating Mallard this release cycle
and potentially going for implementation the next. I would have
advocated a bolder strategy of doing the whole migration this cycle, but
that seems unjustifiable.

So, I think a sensible course of action from here would be for those
unacquainted with Mallard to look into it and compare it with DocBook
(e.g. [1]), and for those advocating Mallard to look at the alternative
route of modifying our DocBook toolchain.

Thanks,

Phil

[1] - http://projectmallard.org/1.0/docbook.html

-- 
Phil Bull
https://launchpad.net/~philbull