<ijp>hmm, guile-sqlite3's makefile hardcodes 2.0 <ijp>naughty naughty wingo <ijp>heh, I didn't realise truncate-file worked on string output prots <ijp>hmm, actually it doesn't, that code is just broken <ijp>oops, spoke too soon <ijp>shanecelis: dunno, but I did my write up of my thompson bug <ijp>why on earth does the lua lexer have a separate for returning source info <ijp>it should have said 'procedure' <ijp>it's weird, but recently, I seem to be dropping words a lot <ijp>one thing that might make sense, is for the lexer to return newlines <ijp>that way you can use it as a token, like ; in the parser <shanecelis>ijp: Oh, awesome! I'll check it out. Heh, nice domain name. :) <shanecelis>ijp: This is cool! I never thought of Thompson's bug as a Quine. Briliant! <shanecelis>ijp: "it's own survival." -> "its own survival." <shanecelis>ijp: Great write up, ijp. I submitted a link to it everywhere. <shanecelis>ijp: Sorry, I probably should have asked you before posting them. <nalaginrut>maybe reddit/hacknews brings huge clicks, I'm afraid ijp build his server with Guile inner server? <shanecelis>nalaginrut: No, the error page was Apache wasn't it. <nalaginrut>hmm...it's unreasonable that Guile could run (display "scheme sucks!\\n"), it should quit <nalaginrut>well, if someone put (display "C sucks!\\n") in the head of program, the compiling speed of Guile will increase crazy <ijp>shanecelis: it's a free internet, I can't police where I get linked from :) <ijp>if I were a saner person, I'd have done it completely static <dje42>Is it ok to use SCM_SMOBNAME (and SCM_SMOBNUM) ? ***cmatei_ is now known as cmatei
*ijp sings copyrighted birthday celebration song *davexunit slaps ijp and sings the free software song instead <davexunit>now everyone get yourselves to cambridge tomorrow :P *dsmith-work remembers the article in Dr Dobbs by rms. ***johnny is now known as Guest7515
***Guest7515 is now known as johnny__
<aidalgol>Is the same seed used for every session? <ijp>the quick'n'dirty solution is (random-state-from-platform) but I don't like the idea of libraries doing that <ijp>unless they maintained their own seed <aidalgol>And regarding this -> "and if we should expose the seed so other people can test their commands." do you plan on allowing user-commands a la fsbot? <ijp>no, I meant for testing <ijp>but I don't have a good reason to need that yet <aidalgol>Can't you get the seed with the REPL, anyway? <ijp>yeah, either with the function I already mentioned, or by setting *random-state* <aidalgol>Why don't you like libraries using random-state-from-platform? <ijp>I don't like libraries touching global state at all <ijp>so what's going to happen is, that issue will remain there for a month or so, then I'll give in, and add it :) <ijp>(random-state-from-platform) <aidalgol>I'm looking into making it maintain its own seed right now, actually. <aidalgol>ijp: OK, fixed I think. Could you confirm? <ijp>ah, it seems I had a misunderstanding about (random-state-from-platform) <ijp>I was under the impression that it mutated the state *aidalgol wishes the Racket docs were available in Info. <aidalgol>ijp: Any more things we should create issues for? <taylanub>Anyone heard news from bipt, since GSoC is over ? Last repo update is from 5 weeks ago on the -sym branch. <ijp>aidalgol: those were the only two things in my todo.org <ijp>I just thought it would be simpler to make two issues rather than commit that file <aidalgol>Unless we were using something like repo.or.cz