On Fri, 2008-02-22 at 10:34 -0500, Jared Smith wrote:
On Thu, 2008-02-21 at 17:30 -0800, Karsten 'quaid' Wade wrote:
While all this toolchain talk is in the air ...
We've got several major activities going on that are relying upon our current-and-working toolchain:
Just to be clear... I'm *not* pushing for a change to a new toolchain before F9. I think it would be awfully rash of us to do so at this point in the game.
I also don't want anybody to mistake my enthusiasm for publican as somehow knocking the current tool chain. I'll be the first to admit that I don't know the current tool chain very well. Additionally, I don't think we've tested publican enough (as far as the translations go, especially) to know whether or not it's a good fit for what we do. What I *am* enthusiastic about is how quickly publican allows people to hit the ground running, especially those who are new to DocBook and their tool chains.
Clearly, the interest in publican and it's (apparent) ease to get started with an XML book is a feature we've been missing in the Fedora Docs toolchain.
Honestly, though, we have been pursuing the higher gain. Across open source projects, the wiki is where the developers and other contributors do their community documentation. When we focused exclusively on XML, we had very interested or enabled contributors.
Some years back I spoke with the Mozilla Dev documentation folks, who had then changed from DocBook XML to a fully wiki-based system:
http://developer.mozilla.org/en/docs/Main_Page
Switching to the wiki, they saw 10x increase from developers, even the ones who knew DocBook well enough. In addition, many new contributors came in to help, which increased the editorial and content group.
It's unclear to me how much these features of publican matter to Fedora Docs. For example, the people who have dropped by #fedora-docs looking for publican help seem to be working on their own content. While maintaining tools that allow the creation of free content is a part of the Docs charter, it's definitely a lower priority than enabling Fedora contributors to create content for Fedora. Right now that means helping to maintain the wiki:
http://fedoraproject.org/wiki/DocsProject/WikiGardening
If I had to prognosticate at this point, I'd say that in two years' time, we'll have taken the best pieces of publican and the current tool chain and welded them together.
An open community development from this point forward is definitely to be desired.
- Karsten