All times shown according to UTC.

Time Nick Message
01:52 roarde joined #copyfree
03:05 libman GitHub doesn't recognize those two aforementioned Mlanguages, ex.[…]=language:mythryl
03:07 And neither does Wikipedia...
03:12 There's a FreeBSD and OpenBSD lang/myrddin ports, which is some sign of life...
03:21 I think being included in ports of at least one Wikipedia-worthy copyfree OS should be a prerequisite for including a language in my comparison matrix.
03:28 roarde joined #copyfree
04:39 libman Added some preliminary notes on copyfreedom in the OCaml family:[…]
04:54 Let me know if there's another ML-related package ecosystem (other than Smackage and OPAM).  GitHub project license types for Standard-ML, Elm, UrWeb, Idris, Agda, and Oz are dominated by "Unknown"s.
05:03 Since many copyfree languages can be grouped into families (mainly either List or ML), I think doing so would make for a more readable list.
05:04 is three stars away from 4000.  ;P
06:11 p4cman joined #copyfree
06:41 roarde joined #copyfree
10:52 roarde joined #copyfree
11:13 roarde joined #copyfree
13:38 roarde_ joined #copyfree
13:45 libman joined #copyfree
15:20 libman apotheon: what do you think of the name OS/NAP (can be pronounced "oh snap") for a hypothetical pure-copyfree distro project?
15:21 It's also a throw-back to Ye Olde Real Operating Systems like OS/400, OS/2, etc.
15:43 p4cman joined #copyfree
16:16 apotheon libman: Sure.
20:29 libman At first glance, Myrddin binaries are a bit on the heavy side -[…]/bench/helloWorld
21:09 apotheon Holy shit, compilation is fast there.
21:09 Holy shit, Go launch time is slow.
22:24 libman It's not a good benchmark, just to identify which languages produce unreasonably bloated native binaries.
22:31 The only observation so far is that D, C#, and Go have a binary / dependency obecity problem, and stripping / compression isn't a good solution.  Haskell, Standard-ML, and Chicken seem reasonably lean now, and Nim never had that problem.
23:02 apotheon I don't see C++ in there.  I wonder how it compares.
23:26 libman That is just a quick draft.  I'll expand it later.  It will go into the "lean" column of the copyfree proglang comparison matrix.
23:28 The UPX experiment was just an ad-hoc idea and should be removed.  Binaries should be `shh`ed to a barebones box to show they can work without runtime.
23:30 Any binary can be compressed, with much decompression overhead time.  It doesn't solve the fat binary problem.
23:34 I think Nim is the only language that will get straight A's on this matrix: module ecosystem copyfreedom, no unfree deps / tooling, portability, lean binaries, safety features, and performance.
23:37 Also, instead of popularity I'll measure "grass-roots community size": IRC and mailing list / web forum activity especially.

← Previous day | Index | Server Index | Channel Index | Today | Next day → | Atom Feed | Search | Google Search | Plain-Text | plain, newest first