Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2017-06-14

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

All times shown according to UTC.

Time Nick Message
00:24 idiosyncrat More specifically, re the last point -- at long as it's clear we're not "squatting", but are reserving the name "Kollos" for a reason, I hope and expect we'll be allowed to keep it.
00:29 ronsavage JK: Do you see some value in not (yet) switching names?
00:30 idiosyncrat As long as the semantics are committed to Perl (which they will be for a while yet) it's Marpa::R3 or at best Marpa::R3/Kollos.
00:30 idiosyncrat Once a user is required to use Lua semantics, it's Kollos.
00:31 idiosyncrat In particular, right now character classes, event callbacks, semantic actions, etc., are all Perl.
00:31 idiosyncrat In Kollos, they'll all have to be Lua.
00:32 ronsavage Ah. OK.
00:32 idiosyncrat That means, of course, that some people may always prefer Marpa::R3 (or Marpa::Rn) even after Kollos is released.
00:33 idiosyncrat This also means I'm not in a real hurry to make it Kollos -- it means abandoning the user base I have, in favor on one that does not exist.
00:33 idiosyncrat s/in favor on one/in favor of one/
00:34 idiosyncrat To explain more, right now inside Marpa::R3 I'm creating a Kollos "lower layer".  It supports the Perl semantics on top of it.
00:35 idiosyncrat When the CPAN Kollos comes out it will be a "thin" covering for that Kollos lower layer.
00:36 idiosyncrat That means, if you're more committed to Perl semantics, for you it won't really be of much interest.
00:37 idiosyncrat OTOH, if you want efficient access from Perl to the much more efficient (and as yet unimplemented) Lua semantics, you'll like the CPAN Kollos more than Marpa::R3.
00:38 idiosyncrat Right now Kollos = Marpa + Lua.
00:38 idiosyncrat In time, it'll be more accurate to say Marpa = Kollos + Perl
00:44 idiosyncrat All that is kinda far off, and right now I'm working up new recognizer & lexer features, hoping to lure users to Marpa::R3.
00:44 idiosyncrat ronsavage: I'm not sure you're the entire Marpa::R3 user base right now, but I suspect pretty close to it.
00:46 idiosyncrat If anybody else is closely tracking Marpa::R3 I'd be happy if you'd come forward on this channel.  It's helpful for me to know how many are out there, and it plays some role in my decisions about moving to beta, how adventurous to be in terms of breaking backward compatibility, etc.
00:48 idiosyncrat My expectation is that the most difficult changes for a Marpa::R2 user are the ones I'm about to make, which (on the positive side) means if you wait until I've made them, you can expect a less bumpy ride on the Marpa::R3 alpha.
01:02 ronsavage As you probably guess: Right now I prefer to wait for the dust to settle on the event 'v' callback change before writing new code.
01:03 idiosyncrat ronsavage: Good thinking. :-)
06:43 ronsavage joined #marpa
09:35 ronsavage joined #marpa
13:29 kaare_ joined #marpa
14:06 kaare_ joined #marpa
14:11 kaare_ joined #marpa
14:30 kaare_ joined #marpa
14:43 kaare_ joined #marpa
14:59 kaare_ joined #marpa
15:04 kaare__ joined #marpa
23:02 ronsavage joined #marpa

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