OK since we appear to have been split up into two teams, and I'm only 
interested in the Long-Term-Team, I'm going to highlight the areas of 
discussion thus-far (together with my own opinions ;), and hopefully we 
can move this onto the aforementioned web-based discussion board rsn.

1. The filesystem structure.
    People have been primarily focused on what the URLs will look like, 
   but we need to see what changes (if any) are required or deriable
    to implement the filesystem structure.
    The only major issue I can think of here is that of duplication, but 
   I'm rather hoping we can link rather than duplicate information.
    What we don't want are symlinks. I don't like that idea in cvs.

2. The URL structure.
    This is very important. Several plans have been published, with 
major   differences voiced over what constitutes documentation, software,
    projects, releases, etc.
    One thing that has a few voices "for", and none "against", is that
    URLs must not have a filename extension, and thus you get
    /page/to/webpage rather than /page/to/webpage.html. I'm in agreement
    here that filename extensions should not be used in URLs, although
    they may be used in the filesystem filenames since we should be able
    to instruct the server software what to load and how to intepret it.
    We should also ensure that all URLs are in lowercase.

3. The server software.
    Must this site be served up by Netscape software? Can Apache not be 
   used if we have the need?

4. Do we use CVS?
    I'm in favour of using CVS myself. The main arguments against it are:
    - cvs prevents some people from getting involved. I don't think this 
     is a serious problem. People who want to make changes to the site
      are probably intelligent enought to figure out cvs or submit their
      changes to a module owner.
    - Because the site is so large, it should use a database. IMO, a
      database brings with it it's own set of problems and it's use
      should be limited to obvious database applications such as
      bugzilla.

5. What standards should we use?
    I think it's generally accepted that we should at all times use W3C
    standards, although a mass conversion at once may not be viable.
    Eventually we'll get there :). Question is, what standards do we use?
    Should we forget the thoughts of being visually-appealling in
    Netscape 3/4, IE3/4, etc., and focus on structure of the document
    while using CSS 2 for complete presentation? Should we use XHTML or
    HTML4.01 Strict or Transitional? There seems to be problems with
    XHTML as noted before, so I'd vote for HTML4.01 Strict, although if
    XHTML is a possibility we should seriously explore it.
    Should we use tables, or should we be really CSS-orientated? Anyone
    got statistics on what browsers visit mozilla.org ?

6. We need to sort out human resources.
    Everyone planning on working on this needs a bugzilla account now.
    We need to modularise the various parts of the site, e.g.
    Documentation, Projects, Software/Releases, etc., then get owners for
    these modules.
    Once we are modularised, we need to work out the finer details of
    what needs to be done to meet our goals, each problem must have a
    bugzilla bug id to work on.

And then the work begins.

We need to make a few primary descisions on the topics raised here, only 
then can we begin to actually work on this stuff.

James Green


Reply via email to