Re: Other observations on www.gnustep.org

[Wim Oudshoorn]

Adam Fedor <address@hidden> writes:

> My point was:
>
> i.e. unless I'm actively updating the links, I'm always copying over
> the FAQ with old links that will be there until the retemplater
> updates again.

I understand.  As far as I kan see there are two options:

* put in make a very small script which will fetch the 
  "links" file from the website and use that version.

* somehow keep the links completely separate from the generated
  document.  

* change the way the website is generated (not that I know how
  it is generated now), to something like:
                       
     Creator of changes -[checkin]---> website repository 
     website reposity   -[checkout]--> temporary location
     temporary location -[scripts]---> temporary location
     temporary location -[upload] ---> public website

  (this would mean that the faq.texi would be put in the
  website repository)     


Number 1 sounds impractical.  For number 2 my knowledge about
the current state of the HTML spec and browsers is insufficient.
So let me put forward a question:

   Is it possible to use css files to completely generate the link
   section?  That is, modify the css file to contain the link
   information it self.    I guess not, because the name
   suggests it is impossible. 

Finally number 3, might conceptually make sense but is probably
too much work and adds complexity.

So I am really hoping that 2 is possible.
I assume that something like php which allows includes etcetera 
is not desirable because you want to keep webpages static
and it adds another dependency.


Wim Oudshoorn.