Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2016-03-10

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:21 idiosyncrat_ I finally got curious enough to investiage whether bird brains are better at parsing than traditional grammars:
01:22 idiosyncrat_ From the paper http://phys.org/news/2016-03-sy​ntax-unique-human-language.html , I extracted this grammar
01:23 idiosyncrat_ S -> D ABC . S -> ABC D . S -> D . S -> ABC .
01:23 idiosyncrat_ The grammar is in grammophone format, *not* SLIF format.
01:24 idiosyncrat_ I used grammophone to test it -- http://mdaines.github.io/grammophone/#
01:24 idiosyncrat_ Results:
01:24 idiosyncrat_ Bird brains are smarter than LL(1), or top-down, parsers.
01:25 idiosyncrat_ However, the grammar is LALR(1), meaning yacc/bison could handle it.
01:25 idiosyncrat_ (For Marpa, of course, I haven't bothered testing it -- the grammar is not even a real challenge.)
01:28 idiosyncrat_ I can't imagine birds putting up with yacc's awful error handling, and I'm not even going near the question of whether we know the full language of Japanese great tits -- they might well need the extensibility that yacc/bison lack.
01:43 ronsavage Not only that: Birdbrains don't use Marpa!
02:39 ronsavage joined #marpa
02:48 ilbot3 joined #marpa
02:48 Topic for #marpa is now Start here: http://savage.net.au/Marpa.html - Pastebin: http://scsys.co.uk:8002/marpa - Jeffrey's Marpa site: http://jeffreykegler.github.io/Marpa-web-site/ - IRC log: http://irclog.perlgeek.de/marpa/today
04:01 idiosyncrat_ A new blog post: http://jeffreykegler.github.io/Ocean-of-Aw​areness-blog/individual/2016/03/parus.html
04:10 ronsavage That post yet again demonstrates the power of Marpa. I dream of the day we can write posts which only say what Marpa can do, and never again have to waste effort saying what other parser can't do.
04:12 idiosyncrat_ I'd been using convoluted phrases like "traditional parsing" to describe the current non-Marpa state of the art.
04:12 idiosyncrat_ I like the new phrase "sub-Parus parsing"
04:13 idiosyncrat_ :-)
05:04 ronsavage Yes! A moment's thought tells me how cunningly sarcastic that name is.
05:55 idiosyncrat_ Good night!
06:09 ronsavage joined #marpa
08:42 iarna joined #marpa
09:16 ronsavage joined #marpa
10:40 maybekoo2 joined #marpa
12:09 Pursuit joined #marpa
12:27 kaare_ joined #marpa
18:26 idiosyncrat_ joined #marpa
18:27 idiosyncrat_ Response to my latest blog post seems muted so far,
18:27 idiosyncrat_ though I am quite pleased with it.
18:27 idiosyncrat_ It's a very small compact way of showing why & how Marpa is different and better ...
18:28 idiosyncrat_ and just how inadequate our traditional (sub-Parus) expectations for parsers are.
18:32 idiosyncrat_ Inadequate to represent the linguistic capabilities of a bird whose brain weight is less than 1/20 of an ounce.
18:57 maybekoo2 joined #marpa
19:02 VsyachePuz idiosyncrat_: grammophone is interesting tool
19:07 idiosyncrat_ Yes, I wish I'd had grammophone earlier.
19:07 idiosyncrat_ Someone should contribute a Marpa frontend to it. :-)
19:08 VsyachePuz idiosyncrat_: exactly my first thought
19:09 VsyachePuz I am not going to implement it, but noted, that there is no earley in the list of algorithms
21:06 ronsavage joined #marpa

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary