I could have sworn we were going to update the lyx web site. Then we had a
minor disagreement, and everything stopped. (Or perhaps things stopped
because people actually started hacking code.) Anyway, as I see it, the
current problem is, do we use php3 or regular html for the LyX home page?
Here are a few options:
(1) Use PHP3
Advantages:
- sexy cookies, et al.
- Very easy to create web page template (e.g. the navbar), to keep the web
site up to date
Disadvantage:
- some mirrors don't have php3
(2) Use plain HTML
Advantage:
- Easy to mirror
- Don't need to learn php3 to code it (Well, the amount of php3 you need to
write a new page is pretty minimal.)
Disadvantage:
- adding a new page requires editing every one of the existing pages to add
it to the navbar. Yuck. Also, hard to keep a consistent look & feel.
I thought up a (3) and (4) today.
(3) Use php3 magic to create "default" html pages. I have to imagine that
php3 has some method to create html pages, perhaps using the default values
for cookies (like the "long names" cookie) etc. Then the LyX home page could
create *.html from *.php3 and mirrors that don't have php3 could mirror the
html pages.
(4) If (3) doesn't exist, I could probably create a simple perl script that
would understand php3 "include" and (simple) variables. That would allow us
to use the main things we want php3 for.
These might have the advantages of (1) and (2) above without the
disadvantages.
Basically, it would be nice to do *something* so that we could get the web
site into cvs already, and expand our mirroring.
The truth is, I personally don't see why we need *so* many mirrors. IMO, we
could just use a few php3-aware mirroring sites. If others disagree, then
we should come up with a different method. But let's not do *nothing*. Once
the web site gets standardized, then while the real hackers are fixing the
kernel, we idiots can work on the less glamorous stuff.
-Amir
