On Thu, Feb 12, 2015 at 10:18 PM, Tobias Famulla <uni@famulla.eu> wrote:
Hello Mailing-List,

I used Latex for a few years in university to create reports for
assignments and also to write my bachelor thesis (I would have liked to
use Context, but the right schema for citation was not available and I
had no time to create it myself).
Over the time I got a little bit frustrated with Latex, because it has
many modules and most of the time gets the job done, but writing Latex
can sometimes be quite hard sometime to me (you have to have the modules
installed, tweak around with charactersets, imagepositioning, ...).

In between I looked at much smaller and sleaker document representations
languages (asciidoc, restructuredText, Markdown) and writing in it is a
pleasure compared to Latex (I haven't really tried out Context but
looked over the documentation and it looked more promising but shares
the same design ideas).
Asciidoc is even able to declarate source code listings and formulas.
Never the less, the output to Pdf is not always the nicest one.

The reason why I now write to this list, is, that I imagine, that
Context could be the right processor to create beautiful PDFs out of
intermediate formats (DocBook 5 or Asciidoc). For the conversion to
Latex a module for asciidoctor (ruby implementation) is in developement.
The ideal system I imagine would be close to what is used with HTML and
CSS on the web: Having a easy to use file format to writing you
documents (Asciidoc or DocBook as intermediate format) and a system to
create the PDFs (maybe Context and a Context-Template)

So my main questions are:
- Are there straigt forward ways to create PDFs with Context using
Docbook 5?
- Are there "not that hard" possibilites to write extentions to Context
to do exactly that (maybe using Lua)?
- Does it make more sense, when using another input format like
Asciidoc, to write a converter which directly creates a
Context-document? (although it might be more versatile to use DocBook
for other formats like Markdown or DocBook itself)

Context alredy has a kind of xslt processor written in lpeg and embedded into the format.
--- see  for example http://wiki.contextgarden.net/XML
The DocBook is a huge specification, so 
I guess that a convert for ConTeXt takes a huge amount of work if you want to map everything --- but it is feasible if you plan to start with a small subset.
From this point of view, Docbook already has a xslt to latex, 
so  working on a xslt to context maybe makes more sense, if one accepts that context is still evolving.


--
luigi