On 26 Oct 2004, at 20:46, Bertrand Delacretaz wrote:
So I cannot refrain from sharing this idea, although I have no code to back it up yet...
Similarly, however not even _hinting_ that this is something I'd want to push myself, your set of goals and techniques still somehow teased me in doing a cross-tabulation with Daisy features...
GOALS
G1. Generate our docs and website dynamically, directly from the SVN repository accessed over http
Not from SVN, but publishing a website is definitely a feature.
G2. Give access to older versions of the docs using standard SVN mechanisms (tags etc)
Likewise, for individual documents. Making the publisher configurable so that one can say: "show me a website with this (version) field set to this value" could be possible though. It's hardcoded now to show the latest version, but one can navigate (per document) to older versions easily.
G3. Index the latest version of the docs, including structured fields (keywords, target audience, components mentioned, etc), to implement "prepared queries" (as links, simply) to improve our docs' accessibility
Oh yes. Metadata & queries. Show queries as documents. Dynamically include documents into other ones. Ditto for query results. Use queries in navigation.
TOOLS / TECHNIQUES T1. Get content from SVN, editing is considered a separate problem
The default Daisy front-end app is definitely Wiki-like: authoring & publishing are one.... in a positive sense however: they share the same skinning system.
T2. Build an index with Lucene, triggered via SVN post-commit hooks, uses a live Cocoon instance to generate an easy to index XML document for Lucene. Include metadata fields as mentioned in G2 above, generated from (enhanced as compared to now) document content
We do Lucene indexing - and we add metadata field content to those indexes, so that documents about a certain keyword can be found even if the exact word doesn't exist in the document itself, only in the metadata.
T3. Generate pages using a live Cocoon instance, maybe Forrest. SVN tags "pass through" the URLs to give access to older releases of the docs.
See G2.
T4. Use queries like "find all documents which talk about sitemap matchers" to build navigation pages semi-automatically.
See G3.
Please mind: I do not want to push this or make this into some "competitive offer". Consider it a thought experience. I hate documentation stored in SVN/CVS with a passion, as much as I hate the dual documentation pools (xdocs/Wiki).
</Steven> -- Steven Noels http://outerthought.org/ Outerthought - Open Source Java & XML An Orixo Member Read my weblog at http://blogs.cocoondev.org/stevenn/ stevenn at outerthought.org stevenn at apache.org
