home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cbmvax!snark!esr
- From: esr@snark.thyrsus.com (Eric S. Raymond)
- Newsgroups: comp.windows.x
- Subject: Looking for skilled X hacker to help fix Xlife
- Message-ID: <1jl47Z#4zGwVx9MyJTb4OKv6T0lY9bl=esr@snark.thyrsus.com>
- Date: 10 Dec 92 15:14:42 GMT
- Distribution: world
- Lines: 38
- Validation-Errors: snark=7Pl3VC721COt8BPgNm3mSL5T
-
- A couple years back I got my hands on the source to Xlife 2.0 (the last version
- publicly released, to my knowledge) and made quite a few improvents --- on-line
- help, improved commands, a better pattern-loading facility.
-
- What I was really after, though, was to generalize Xlife into a program capable
- of handling table-driven cellular automata with more than 2 states (retaining
- the original fast code for the 2-state case). The Codd automaton, Wireworld,
- things like that. I did most of the coding for this.
-
- Trouble is, the display didn't work. The program refused to display colored
- cells at scales above 1 pixel per cell. After some back-and-forth with the
- two original authors (a pair of CMU students who've since dropped out of sight
- and no longer answer email at the addresses I have) we arrived at the theory
- that my X server (an archaic X11R3) was buggy.
-
- So there it lay for many months. Now I have a fast X11R5 --- and the program
- fails in a *completely different* way. Now it will display cells at scales
- above 1, but not at scale 1 --- the opposite of the previous behavior. And
- it's prone to core-dumps.
-
- I hasten to add that the 2-state B&W case, ordinary Life, works beautifully. I
- didn't break that --- matter of fact I've sped it up a good deal. The problem
- is solely in the program's interaction with X when color is used to indicate
- cell states.
-
- With this posting, I am formally crying uncle. X has baffled and defeated
- me. I'm an expert C/UNIX programmer and I'm normally very good at picking
- up new interfaces quickly, but X's is sufficiently complex and this enough
- of a heisenbug that I can't get a grip on it.
-
- Rather than sit on the Xlife code for another year hoping for lightning to
- strike, I'm looking for an good X hacker to help me figure out just what in
- the bleedin' hell is wrong and fix it so we can release Xlife 3.0 and let
- *everybody* have the improvents I made.
-
- Any volunteers? Reply by email, please.
- --
- Eric S. Raymond <esr@snark.thyrsus.com>
-