Re: Successor to DocBook
- From: Shaun McCance <shaunm gnome org>
- To: karderio gmail com
- Cc: Matthew East <mdke ubuntu com>, gnome-doc-list gnome org
- Subject: Re: Successor to DocBook
- Date: Thu, 13 Jul 2006 10:45:35 -0500
On Wed, 2006-07-12 at 13:34 +0200, karderio wrote:
> [snipsnip snip, here's what's left]
> > > > It's not really so sudden. The idea of replacing DocBook has
> > > > floating around for years now. Discussions have happened in
> > > > of places, including IRC. (Sidebar: somebody please create an
> > > > client that doesn't suck. Allow me to select a range of
> > > > and send it to a mailing list for public consumption and
> > > > with no more than three clicks.)
> What is wrong with xchat-gnome - apart from the stability and the fact
> that it is far from finished ? I'm sure a plugin for this
> could be easily hacked up for this...
To be honest, I'm just difficult. I could never get xchat-gnome
to run before. I thought gnomechat was promising too, but it
looks dead. While I want a simpler, Just-Works IRC client, I'm
also used to some of the advanced features I've managed to learn
in xchat, so...
> > I have looked at DITA. It looked somewhat promising at first,
> > but we have a fundamental difference in design philosophy. My
> > design philosophy is "as simple as possible, and no simpler."
> > This could also be restated as "make life easier for everybody
> > involved."
> Would/could we have Mallard import/export utilities to/from
> etc ?
There are two problems with converting Mallard to DocBook:
1) It is not quite as semantically rich. The resultant DocBook
would not be quite as perfect as that done by hand. It would
still be better than DocBook created from HTML or Moin though.
2) Mallard employs topic maps, and we'd have to figure out a sane
way to serialize those into a linear structure. This, by the
way, is something we'll have to give thought to anyway if we
want to be able to print entire documents.
There is one major problem with converting DocBook to Mallard:
You won't get topic maps. We'd have to arbitrarily choose a
chunk depth and creates nodes and pages as best we can, but
a human being would have to do surgery to make it into really
good help. But hey, tools that do half the job can still be
As for DITA, is there a need? Is anybody here doing stuff in
DITA right now? DocBook migration is important, because all
of our current documentation is in DocBook.
] [Thread Prev