timotimo no problem at all 00:11
01:48 ilbot3 joined 05:36 brrt joined
brrt good *, #moarvm 05:39
nwc10 good *, brrt 05:49
brrt pre-coffee/post-tea here 05:50
05:53 domidumont joined
brrt it's about time i write a blog i think 05:54
05:59 domidumont joined 06:56 brrt joined
brrt by the way 07:04
tensorflow is sooo cool
it's like a turbocharged version of R
… now hang on a second 07:05
we have shaped arrays
what if perl6 would become the easiest, fastest way to use libraries like tensorflow 07:06
nwc10 "people would still use Python 2.7"? (I'm not sure if this is a joke. Or if even if it is, if it's funny) 07:07
(but it's certainly a backhanded compliment) 07:08
brrt hehe
yeah, it is, in a way
staying power
that is what keeps perl5 alive as well 07:09
aside from being, altogether, a pretty good language
yesterday i had another of my oracular predictions
(my last one was that machine learning would be a 'bog-standard, MySQL-and-a-webpage level component in 15 years. I was wrong. It's going to be much, much, much, sooner) 07:10
within a similar time period (let's say 10 years), we're going to have most of our existing codebase slowly rewritten or simply dwarfed by 'high-assurance' systems in 'safe' languages
however, rust is not going to be the leader of that pack
why not? because rust is pain 07:11
nwc10 I've not used rust. From the outside, what I like about the (intent at least) is that it's intended to be able to be used without a GC (at least, a subset of it) 07:12
and that it's intended to make concurrency easy
the latter should be a given of anything "new" that folks plan to migrate to, given that smart watches are now multicore
brrt i can probably answer why rust is pain, with a short and example I gleaned from a blog post 07:14
.. maybe i can find the blog post in question
m-decoster.github.io//2017/01/16/f...borrowchk/ i think this was it 07:15
the gist of it
a Student struct cannot have a reference to the Classes he's supposed to follow, because the borrow checker, hates, hates, hates circular references, and you should have an effective 'join table' structure instead 07:16
so rather than have a pointer to a list of classes, you now have to manage a table of { class, student } pairs
is that more flexible? certainly
is that however overkill and sometimes even just the Wrong Thing? almost certainly as well 07:17
doesn't the developer know best? Not in rust-land he/she don't
that's why rust is pain
07:41 AlexDaniel joined, zakharyas joined 07:54 zakharyas joined
jnthn morning o/ 08:38
lizmat o/ 08:40
jnthn Hm, tensorflow
I was looking at that momentarily at $dayjob and thinking "hmm, Perl 6 binding"...no way that'll be something I get a $dayjob excuse to do, alas. 08:41
brrt alas indeed 08:46
jnthn already has enough free time projects too :) 08:47
08:51 ilmari[m] joined 10:19 zakharyas joined 10:29 brrt joined 11:23 brrt joined 11:29 brrt joined 12:15 ilmari[m] joined 12:32 brrt joined
Geth MoarVM/master: 11 commits pushed by (Jonathan Worthington)++, jnthn++
review: github.com/MoarVM/MoarVM/compare/3...9f9a2a045b		 13:03
13:17 brrt joined
Geth MoarVM: cono++ created pull request #602:
getport op code		 13:41
13:47 brrt joined
Geth MoarVM/master: 4 commits pushed by cono++, (Jonathan Worthington)++ 14:09
15:11 zakharyas joined 15:28 brrt joined 16:33 brrt joined 16:53 domidumont joined 17:06 AlexDaniel joined 17:37 zakharyas joined
AlexDaniel m: say 42 17:52
camelia 42
17:59 committable6 joined 18:17 geekosaur joined 19:36 Ven joined 19:46 Ven_ joined 19:57 zakharyas joined 21:03 Ven joined 21:38 robertle joined 21:43 nebuchadnezzar joined 21:47 robertle joined 21:52 robertle joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!