WARNING: Severe geekdom herein
As mentioned on NewIdeas , I've changed the way this site is managed. No more will the pages be turned into XML by a markdown or emacs-wiki regular expression script, there to be converted using XSL into actual web pages. (Yes, I did do that.)
But now I have this "sitegen" thingy. Sitegen uses a real syntax driven parser to parse emacs-wiki and markdown syntax into an HTML model parse tree, then does interesting things to that tree, including apply templates, which is how that nifty site navigation tool shows up so consistently on the left on all these pages.
It also allows me to write nifty things like "%%categoryList(category='bliki')" have that expand to a list of all pages who are tagged with the category "bliki", which I'm using (surprisingly enough) to do my new WebLog type thing (Actually, it's a bliki . More on that later.)
I'm not posting the script just yet. Aside from it being horrendously undocument, there are a few sytax quirks that I want to clean up and a hack or two I want to remove from the syntax. I wouldn't mind someone else using it but I wouldn't want to inflict it on some poor innocent who thinks the language won't change.
Also, I intend to shortly parse actual HTML files and be able to apply all my nifty tricks to them, so a release should probably wait until that point.
Yes, I might at some point just use TWiki but I wanted to play around with the PERL recursive descent parser generator and syntax tree rewriting anyway.
(Of course, the biggest reason I might use someone else's software for thi site is so I don't have to figure out and deal with the differences between IE and Firefox.)