This is the mail archive of the mailing list for the DocBook project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Why no MAP/SHORTREFs in DocBook?

On Tue, 12 Dec 2000, Michael Smith wrote:

> Trevor, you wrote:
> > In using the DocBook DTD in anger recently I've found it very
> > inconvenient that there is no use of the MAP feature or at the least
> > hooks, by which I mean parameter entity declarations, so the user
> > can define them themselves.
> My guess is maybe because the DocBook TC and docs have made it clear
> at least since v3.1 that as of v5.0, the only official DocBook will be
> XML compliant.

I was hoping that ignoring this retrogressive step I could avoid it. :-)

> As you know, XML -- for good or bad -- forbids all
> markup minimization, including SHORTREF and USEMAP.

Back in the days when XML was conceived as "implementable in a couple of
weeks by a computing science student" the lack of minimisation made some
sense to me. However, as XML has grown that goal has been forgotten as has
the usefulness of and desirability for minimisation. But that's probably
off-topic for this list.

> > The sort of thing I'd expected to be part of DocBook was for
> > &RS;&RE; sequences to be interpreted as occurences of a para start
> > tag. ...
> Out of personal curiosity and at the risk of revealing my ignorance,
> why the particular character sequence &RS;&RE; ? I realize it must
> have special significance, but I don't know enough to recognize it.
> I'm also genuinely curious to know in what sort of scenario that kind
> of minimization is useful. Is it because you're marking up legacy
> documents? or typing tags in manually in a text editor?

Both. When it's a choice between Word and ed I'll choose the latter every

> Wasn't markup minimization abandoned in XML because many SGML folks
> had already given up on it as making tool development too difficult?

See above. :-)

> That is, while minimization makes it easier to author documents using
> simple text editors, it makes things much tougher for the developers
> who have to write parsers to handle minimized doc instances.

This has always been my beef with XML: Developers restricting the features
that the users get because they (the developers) find it hard, actually
just simply inconvenient, to implement. That minimisation has been
implemented in SGML parsers demonstrates that it can be done. :-)

> A while back, I came across a vintage (1996) opinion on this same topic:
> I'm guessing you've heard all the arguments already, and still would
> like to have minimization.

Heard them in 1985 (sic) didn't believe them then. Don't believe them now.

> As I said above, as someone muchh less
> experienced with this stuff, I'm curious to know in what kind of
> situations you find it useful (or necessary).

Whilst we may be generating a whole pile of markedup documents the pile of
unmarked legacy documents is bigger. And will remain so for decades to
come. Taking away user-friendly features isn't going to help that
conversion process.

In as much as my comments may be deemed off topic for this list they are
germain to a discussion on the whole rationale behind DoCbook.

Regards, Trevor

British Sign Language is not inarticulate handwaving; it's a living language.
Support the campaign for formal recognition by the British government now!


<>< Re: deemed!

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]