Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2016-06-21

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

All times shown according to UTC.

Time Nick Message
00:13 stmuk joined #moarvm
01:48 ilbot3 joined #moarvm
01:48 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
02:22 lizmat joined #moarvm
06:12 domidumont joined #moarvm
06:17 domidumont joined #moarvm
07:46 nwc10 http://ianozsvald.com/2016/06/20/results-for-which-version-of-python-2vs3-do-london-data-scientists-use/ -- interesting. if you read it, please read the "Bias" bit at the end
08:00 moritz if I'd start a new python project, it'd certainly be in 3.4+
08:01 moritz but legacy, legacy, legacy
08:01 nwc10 me too*. and yes, legacy
08:01 nwc10 * I wanted something not-perl to validate XML I generated. But I found a command-line thing that seemed to be C, that was already installed
08:23 rurban_ joined #moarvm
08:34 cognominal joined #moarvm
09:03 arnsholt Yeah, Python 3 is a lot nicer
09:03 arnsholt When I looked at the differences in semantics and such between 2 and 3 when I started Snake, I basically looked at old-style vs. new-style classes and went NOPE!
09:03 arnsholt All of the nopes, in fact
09:04 timotimo yeah, never do old-style classes
09:05 arnsholt Yeah, I wanted to do something moderatly clever in a Python 2 project, involving descriptors and couldn't figure out why the descriptors weren't called
09:05 arnsholt Then I realized: it's old-style unless you explicitly inherit from object
10:54 domidumont joined #moarvm
11:36 domidumont joined #moarvm
11:46 brrt joined #moarvm
11:46 brrt nwc10: interesting
11:47 brrt otoh, descriptors are such a dumb design
11:47 timotimo what's that again?
11:48 brrt its...
11:48 brrt an object
11:48 brrt which takes the place of an attribute on another object
11:48 brrt and which has three methods: get,set,delete
11:48 timotimo oh
11:48 brrt e.g.
11:49 moritz a layer of indirection over an attribute
11:49 moritz much like returning a Proxy in Perl 6
11:49 moritz python makes heavy use of them
11:49 brrt class Foo(object): bar = descriptor(); f = foo(); f.bar = 3; (calls f.bar.__setitem__(f.bar, f, 3)
11:50 brrt perl6 has containers, which (imho) is more sane
11:50 timotimo we also have NativeRefs which basically are those descriptors :P
11:50 timotimo those that are such a dumb design
11:50 brrt not sure i agree
11:50 moritz the weird thing in python land (IMHO) is that from the outside, you have to know if something is an attribute or a method
11:50 timotimo maybe :)
11:51 brrt that what moritz says
11:51 moritz but if you access it as an attribute, there might still be a method hidden behind it (trhough such a descriptor)
11:51 timotimo right
11:51 timotimo NativeRef works in terms of assign and decont
11:51 brrt on the inside, you somehow have to store something in the object hash, without conflicts pretty please
11:51 brrt on the inside, you don't necessarily know what 'name' your attribute has
11:52 timotimo hm
11:52 brrt i'm very confident that descriptor are the wrong way to approach the question of 'magic attributes'
11:52 timotimo magibutes
11:53 brrt but yeah, proxies are also like that
11:53 timotimo proxies don't let you set a name without mixing in an anon role with a .^name on it, eh?
11:54 brrt except that you probably create a proxy per object (or per invocation), whereas the descriptor is a class-global instance
11:55 timotimo ah, yeah, the proxy needs to close over the instance you want to access
11:56 lizmat joined #moarvm
11:59 brrt \o lizmat
12:00 lizmat brrt o/
12:03 timotimo <o
12:05 jnthn o/ lizmat
12:05 jnthn Enjoying YAPC?
12:05 jnthn Or TPC as I heard it's now called...there's one less FAQ :)
12:06 nine New FAQ: "What happened to "YAPC"?"
12:06 brrt and will we be calling it TPC::EU next year?
12:08 jnthn And if it's TPC::EU, will the UK be allowed to host it again? :P
12:09 moritz well, Kiev wasn't really "EU" either :-)
12:10 brrt we'll host it in switzerland or norway
12:10 moritz in both!
12:10 jnthn Goodness, the beer prices... :P
12:10 moritz Poland, maybe? :-)
12:10 moritz closer to me, cheaper to eat and drink
12:11 brrt we can also do YAMC for a moarvm conference
12:12 timotimo yam yam
12:12 * jnthn doesn't mention how nice it'd be to have it in Prague at some point, 'cus he knows that's a case of "well volunteered" :P
12:12 brrt which is just us 5 or 10 guys hanging out
12:12 brrt prague is an excellent idea
12:12 brrt it's settled then
12:12 timotimo i need to offer brrt hugs
12:12 brrt YAMC::EU 2016 in prague
12:12 timotimo this year, already?
12:13 timotimo we only have have a year to plan it, absolute-maximum
12:13 brrt :-D
12:14 moritz well, a smaller conference needs less planning
12:14 moritz also, please spell it MOARCONF
12:15 brrt awesome
12:16 nine +1 for MOARCONF
12:16 timotimo COANF
12:16 jnthn heh, a MOARCONF would be fun
12:21 lizmat jnthn: sorry, was a way for a bit having breakfast
12:22 lizmat TPC is nice, hope the videos of yesterday will be up today
12:22 lizmat the guys were working on it last night already
12:23 lizmat MOARCONF is an excellent idea, BTW
12:23 lizmat perhaps after Alpine Workshop ?
12:23 nwc10 I assume it's more likely to be a Moar Unconference (come hackathon)
12:23 lizmat or maybe *at* the Alpine Workshop ?
12:25 timotimo i'm not sure we want to pull in the whole UN for this
12:27 brrt when is the APW?
12:32 brrt 2-3 september
12:32 brrt hmmmm
12:35 brrt oh well, i can always try to get free time for that :-)
13:28 lizmat joined #moarvm
14:08 domidumont joined #moarvm
14:46 rurban_ joined #moarvm
16:01 cognominal joined #moarvm
16:59 domidumont joined #moarvm
17:30 lizmat joined #moarvm
18:57 FROGGS joined #moarvm
19:01 lizmat joined #moarvm
19:32 mst___ joined #moarvm
19:32 japhb joined #moarvm
20:05 lizmat joined #moarvm
20:10 lizmat joined #moarvm
20:38 cognominal joined #moarvm
21:45 lizmat joined #moarvm

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