Perl 6 - the future is here, just unevenly distributed

IRC log for #darcs, 2016-05-11

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

All times shown according to UTC.

Time Nick Message
00:52 thorkilnaur_ joined #darcs
01:39 Riastradh joined #darcs
01:47 ilbot3 joined #darcs
01:47 Topic for #darcs is now Darcs/Pijul Sprint, Helsinki, 6th-8th May http://darcs.net/Sprints/2016-05 | http://darcs.net/ | logs: http://irclog.perlgeek.de/darcs/ | darcs 2.10.3 is out http://darcs.net/Releases/2.10
02:06 mizu_no_oto joined #darcs
02:17 pointfree joined #darcs
03:07 mizu_no_oto joined #darcs
04:34 Big_G joined #darcs
06:53 Weltraumschaf joined #darcs
07:13 inhortte joined #darcs
08:08 Weltraumschaf joined #darcs
10:13 pbgc joined #darcs
10:28 inhortte joined #darcs
11:13 Weltraumschaf joined #darcs
11:21 mizu_no_oto joined #darcs
11:55 inhortte joined #darcs
12:27 BitPuffin joined #darcs
12:48 pbgc joined #darcs
13:13 pbgc joined #darcs
13:28 Weltraumschaf joined #darcs
13:30 ocharles_ joined #darcs
13:30 ocharles_ hi all. http://darcs.net/manual/ says "2.8.4" - should that be updated?
14:02 sm hi ocharles_ .. absolutely, good catch
14:06 ocharles_ :)
14:13 roconnor joined #darcs
16:50 amgarchIn9 joined #darcs
17:09 Heffalump hmm, the release instructions don't mention updating the manual.. http://darcs.net/Development/ReleaseManagement
17:17 amgarchIn9 joined #darcs
17:29 amgarchIn9 joined #darcs
18:25 roconnor joined #darcs
18:59 roconnor joined #darcs
19:45 pbgc joined #darcs
19:45 pointfree http://hub.darcs.net/pointfree/kona (kona scm). Usually I don't like to share projects so early, but I guess it doesn't hurt.
19:47 pointfree (will be written in Rust most likely and licensed the same as pijul)
19:57 Heffalump you might want to talk to the pijul developers about the licence if you care what it should be, rather than just being compatible with them
20:01 pointfree I happen to also like AGPLv3
20:23 Heffalump cool, what will it do?
20:31 pointfree I wanted it to be easier to slice-and-dice changesets between commits after the fact, temporarily, and as needed.
20:34 Heffalump would you be able to recompose them to get the original one?
21:41 pointfree Good question. Splitting changesets would break that snapshot of user intent and be something like adding conflicting branches.
21:43 Heffalump if you couldn't, how would it differ from just re-recording in darcs?
21:59 sm yow... ekmett: "switching to github from darcs moved me very rapidly from having received two patches by email to roughly 500 collaborators and tens of thousands of patches. My goal is to get eyeballs on code to eliminate bugs and the ecosystem effects are amazing."
22:08 pbgc joined #darcs
22:41 pointfree Heffalump: I don't know of a way to do it in a provably correct way, but at the moment, I'm exploring using the changeset/patch as a context vector, a semantic fingerprint.
22:41 pointfree I will use a dense format of Sparse Distributed Representations instead of opaque hashes
22:41 pointfree Similar data overlaps where it is semantically identical. SDRs have some nice properties http://arxiv.org/abs/1503.07469
22:41 pointfree Common things in computer science such as generalization are somewhat complicated in localist memory...
22:41 pointfree but are simple Boolean functions like AND/OR with SDRs. I'm exploring supplementing patch theory with SDRs, semantic folding, etc.
22:41 pointfree http://www.cortical.io/static/downloads/semantic-folding-theory-white-paper.pdf

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