Re: [docbook-apps] New Branch: website5

["Sina K. Heshmati" <sina@khakbaz.com>]

Stefan Seefeld wrote:
> On 05/20/2010 11:54 AM, Sina K. Heshmati wrote:
>> "Denis Bradford"<denis.bradford@verizon.net>  said:
>>
>>
>>> I realize this is not a trivial thing. Besides the layout, there doesn't
>>> appear to be much difference between the DocBook and Website (full)
>>> documents -- mostly a few elements at the top. But the big difference is
>>> in the processing, and that would no doubt require a lot of work to bridge.
>>>
>>> The benefits just might be worth the effort. Making Website a DocBook
>>> output option, instead of a separate dialect, would increase its value
>>> for technical documentation -- a low-tech, frameless alternative to
>>> Eclipse infocenters and HTML-based help browsers.
>>>
>> Some of these features are indeed useful for both DocBook HTML output and DocBook
>> Website but please note the fundamental distinction between DocBook and DocBook
>> Website. DocBook helps format documents and publish them anywhere, including the
>> Web whereas DocBook Website helps publish *websites* on the Web and only on the
>> Web.
> 
> I think things aren't quite as clear-cut as that: As an interesting
> use-case, consider Sphinx (http://sphinx.pocoo.org/), which started out
> as a "documentation tool". It is based on ReST (with its existing
> formatters), but adds superstructure to it.
> There, generating a "website" proper is just a side-product (or almost).

Right. Things might not be as clear-cut but agreeing on a definition helps us determine the scope of the project.

> Taking this to DocBook, I think it would be nice if DB Website content
> could be published into an inter-connected set of (pdf) articles or
> somesuch.  (Since the stylesheets for individual page content already
> exists, I imagine that the only new bits required would be templates
> that use the website structure information ("layout" or similar), and
> generate appropriate inter-document links from that, as well as other
> toplevel things such as an index, TOC, etc.)
> 
> But, that's just an idea, and may not fit into this GSoC project. Still,
> it's worthwhile thinking of this for some future work.

Thanks for sharing.

SinDoc