Re: In the beginning

On Wed, 2003-02-05 at 04:49, Lauri Watts wrote:
> Hi all,
> 
> I wasn't on the list yet when Aaron wrote his mail, so excuse thread breaking 
> and strange quoting.
> 
> > o A document format. We need to settle (quickly) on a final document format. 
> > Docbook is a candidate, but we need to look at what sort of methodology we 
> > wish to employ. Do we want to work on both documents in parallel files, or 
> do 
> > we want to work on folding both documents into one "metafile" that can be 
> > processed to present either a KDE- or GNOME- specific final result. The 
> > latter sounds more ambitious to me, but also perhaps a good way of doing 
> > things if only because it allows us to truly share the common parts of the 
> > resulting documents.
> 
> I can't see any reason to not use DocBook.  If another format is at all 
> useful, there either are stylesheets to get there, or someone can write xslt.  
> The advantages are enormous: The content is already mostly there and marked 
> up, it could be processed immediately by standard tools of both desktops, and 
> it can go into things like the translation workflow immediately.  I can't 
> think of any disadvantages - enighten me if there are some.
> 
> This could be done very easily and very quickly, by breaking out the files 
> this way:
> 
> * One gnome-index.docbook using Gnome's normal customization layer to OASIS 
> DocBook
> * One kde-index.docbook using KDE's normal customization layer to OASIS 
> DocBook [1]
> * One file per section in the HIG (name it something like <section>.docbook).  
> * Include those sections that are agreed upon by both parties directly into 
> both documents (this is as simple as defining a single entity, and then 
> including that entity in the index.docbook files)
> * Optionally, up to two extra files per section in the HIG (named something 
> like gnome-<section>.docbook and kde-<section>.docbook) holding content that 
> is either not applicable to the other desktop, or cannot be agreed upon.  
> Each index.docbook above would include the appropriate content.
> 
> Then we could attack in two ways.  Either 
> 1: Initially, all the sections could be of status "undiscussed", which would 
> give immediately a complete HIG document, but with much of the content 
> potentially subject to change.  Then work through it in some kind of order. 
> or
> 2: Each section is added as it is agreed upon here and finalised.

I think (2) is clearer.

-Seth



-- 
open-hci mailing list
[email protected]
https://listman.redhat.com/mailman/listinfo/open-hci