Re: The gnome-help.2



Alexander Kirillov <kirillov@math.sunysb.edu> writes:

> I wouldn't expect MathML to be implemented. First, all docs are
> written in DocBook, and I do not know of any way to combine MathML
> with DocBook (I may be wrong, of course).

I don't think it's possible in a very nice way with DocBook SGML.  But
if/when the GDP uses the XML version of DocBook, it should be easier.
In XML written with one DTD (ie DocBook) you can embed a piece of
document in another DTD (eg MathML, or anything else that'd be
useful).  I think this is done using namespaces (see
http://www.w3.org/TR/1999/REC-xml-names-19990114/ for the spec).

I believe this is why GNOME apps tend to use XML for their data
formats - you can then embed Bonobo things in the same file, even
though the `parent' app doesn't understand that bit of the file - it
just delegates it to the Bonobo component.  Or at least, if they don't
do that, it seems like a good idea to me.

> Second, MathML is pretty new and not supported by majority of
> browsers. I'd assume that it would require too much work from
> developers to make Nautilus render MathML properly.

Mozilla now has experimental MathML support.  And I think Nautilus
will embed Mozilla at some point (if it hasn't already), although I
guess it might not be the default HTML renderer for help browsing.

> On the other hand, some things (like sub/superscripts, and many math
> symbols, such as arrows, sum and product sign, etc) are possible
> with standard DocBook - they are defined as entities, see files
> /usr/lib/sgml/ISOams*. E.g., the sum sign is &sum;. The right
> question is, of course, to make sure that docbook2html converter
> which will be used for gnome 2.0 understands these entities and
> produces html which is properly rendered by Nautilus and other
> browsers. We should mention this to jrb, who is working on this
> converter.

The problem with using entities like that is that it isn't semantic
markup, which is the main point of using DocBook in the first place.
If you implement an equation like that, it makes it much harder for,
say, a speech-synthesising help browser to work out something useful
to do with it.

> PS. I wish this whole documentation system were based on Latex
> rather than DocBook - it would give so much more freedom, and I like
> LaTeX much, much more... But of course, it is too late to cahnge it
> now, and would create other serious diffculties

The main difficulty being that LaTeX isn't very good at semantic
markup.  It's heavily optimised for printing.  I would guess that most
reading of help will be done on screen, which LaTeX isn't very good
at, and it's *really* bad for speech synthesis, etc.

    Andy
-- 
Andy Smith                                             ams70@cam.ac.uk
St John's College, Cambridge, CB2 1TP, UK               (07989) 453208





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]