felix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Upayavira ...@odoko.co.uk>
Subject Re: How close to graduation?
Date Thu, 10 Aug 2006 12:22:49 GMT
Marcel Offermans wrote:
> On Aug 9, 2006, at 16:50 , Upayavira wrote:
>> Richard S. Hall wrote:
>>> I guess the main issue for me is, I don't really know how to structure
>>> and synchronize what we have on the wiki with what is on the incubator
>>> web page.
>> Yep. And that's what we need a volunteer for. Did I get the impression
>> Alex was volunteering?
> Let me summarize my view on where we stand with this.
> Basically the website consist of content from one of two sources:
>  - documentation authored by developers and users;
>  - reports and documents generated by the build process.
> The challenge is how to mix these two artifacts together into one
> consistent site.
> We would like all authored documentation to be in the confluence wiki.
> For confluence we have already installed and configured the auto
> exporter plugin, so we can create static copies of this content
> automatically (although the auto exporter only allows us to export to a
> local directory).
> For the documents generated by the build process there is no automated
> process in place yet because we don't have a continuous build server
> (and according to Upayavira's mail earlier today we are unlikely to get
> one right now). That means that at best these artifacts will be
> generated manually and it probably makes sense to generate them for
> every release we do.
> So, apart from the actual content, we have the following technical issues:
>  - How to get the files from the auto exporter to the right location (or
> the other way round, how to redirect from the felix homepage to this
> auto exporter location)?
>  - How to merge both sources into one consistent site (this is probably
> not too hard considering the fact that we can provide a template for the
> auto exporter which could simply include links to all the build
> generated content).
> I have no problem volunteering to help with this, but I'm not familiar
> with Apache infrastructure rules and guidelines, so some assistance on
> these tech issues would be very welcome.

We need to do the simplest thing possible right now, IMO, before the
September deadline (Sept 20th). The simplest way to plan this would be
to say we've got until the end of this month.

Remember, the requirement is to have a website that is reasonably
informative, not to have a state of the art CMS. That can be done post

I'd suggest we:
 * Use confluence to author the pages we want on our site
 * When we're happy with them, someone manually converts these to xdoc
 * We use Anakia or Maven (as is) to produce a very simple site using
   these xdocs
 * We're done.

Then after graduation, we can look into how to use Confluence as a more
effective backend, how to configure the confluence template, how to get
the content from Brutus into SVN, and from SVN onto Minotaur (the master
www server for Apache, currently also people.apache.org).

One deviation from this I'd consider in the short timeframe would be to
look at whether we can create a template for the auto-exporter that
would export xdoc instead of HTML. With that we could run a simple
script on one of our own machines that would wget the content (or some
such), run Maven or Anakia, then commit src and generated docs to SVN.

That would be neat, and would fit sufficiently well with our existing
approach to site maintenance.

Regards, Upayavira

View raw message