People
Repo info
Activity
  • Jul 29 2018 14:38
    madmalik commented #9
  • Jul 29 2018 12:43
    madmalik commented #9
  • Jul 29 2018 12:43
    madmalik commented #9
  • Jul 29 2018 11:02
    uazu commented #9
  • Jul 29 2018 08:46
    madmalik commented #9
  • Jul 28 2018 09:03
    uazu commented #9
  • Jun 07 2018 08:36
    cipriancraciun opened #10
  • Jun 03 2018 12:29
    madmalik commented #9
  • May 30 2018 20:40
    pliniker commented #9
  • May 28 2018 19:29
    jolby commented #1
  • May 25 2018 10:07
    madmalik edited #9
  • May 24 2018 19:57
    madmalik opened #9
  • May 06 2018 15:44
    cipriancraciun commented #8
  • May 06 2018 09:02
    madmalik commented #8
  • May 04 2018 14:15
    cipriancraciun commented #8
  • Apr 28 2018 19:58
    madmalik commented #8
  • Apr 27 2018 12:52
    mattgreen commented #6
  • Apr 26 2018 17:13
    uazu commented #8
  • Apr 26 2018 16:11
    madmalik opened #8
  • Apr 26 2018 11:09
    pliniker commented #7
kyren
@kyren
honestly, no
the behavior of arrays is just TOO surprising imo
or, the array like behavior of tables, I mean
madmalik
@madmalik
yeah, makes sense
kyren
@kyren
the # operator is exactly defined, it makes.. "sense", it's just really very frustrating
I guess.. it's fine until you want to actually use tables, it's fine because you can make userdata types that sort of fill in the gaps, like you could make a lua that just has a single bespoke array userdata always available or something
madmalik
@madmalik
on painpoint i haven't encountered personally but i can imagine exist is interfacing with formats like json
kyren
@kyren
which you actually very much have to do if you interact with something like json, the ambiguity between [] and {} and null values in arrays will bite you
oh haha, great minds :P
madmalik
@madmalik
xD
kyren
@kyren
the runtime behavior of the array vs the map part and guessing at the runtime complexity is actually NOT that big of a deal I think, the worst part is that you're missing array semantics, but I suppose if you have a strong enough userdata story it doesn't matter that much, and that's sort of what lua is for in my mind?
madmalik
@madmalik
mhh, yeah
kyren
@kyren
which is also why it bothers me so much that bindings are inherently slow, and which is why I'm writing my own lua
madmalik
@madmalik
thanks for the long answer! its really interesting
kyren
@kyren
everyone always tells me that haha
madmalik
@madmalik
@kyren btw, since you thought about a name for what you called arena. the graph theory term for a subgraph with no connection to the outside would be a closure, which is unfortunately already taken in our domain. But what do you think about something like fence or barrier?
Bruce Mitchener
@waywardmonkeys
island
Edd Barrett
@vext01
Quick shout out for the Benchwork'18 workshop: https://conf.researchr.org/track/ecoop-issta-2018/benchwork-2018-talks
You are all benchmarking, right? ;)
(Deadline this friday)
Peter Liniker
@pliniker
I've had a lot going on in my evenings and weekends over the past 4 weeks so book progress has been minimal and sporadic. My unmerged WIP commits are being published here https://pliniker.github.io/book/master/ and the issue tracker is here https://github.com/rust-hosted-langs/book/issues
The direction the book will take is towards implementing Immix, given that Immix peforms generally well in various dimensions.
Peter Liniker
@pliniker
Other benefits are that @YorickPeterse has an implementation for Inko and there's another implementation in Rust out there connected with a published paper
btw @kyren - the RustConf closing keynote! That's very cool!
Yorick Peterse
@YorickPeterse
@pliniker if you want to know anything about how I did Immix, just shout :)
Peter Liniker
@pliniker
oh I will! :grinning:
Yorick Peterse
@YorickPeterse
maybe one day I can actually figure out how to turn that into a nice talk and have an excuse to go to RustConf
Dylan DPC
@Dylan-DPC
haha
madmalik
@madmalik
hey everyone who is still here: should we transfer to this channel into the rust discord?
since the rest migrated more or less i don't have gitter open all that much, i imagine thats the case for other people too
Aaron Weiss
@aatxe
Doesn't affect me. I'll continue to have both open either way. :wink:
madmalik
@madmalik
i don't feel strongly about it (and maybe we all should wait until discord is really where rust will end up...)
but i think in the long run it'll be better if all the groups end up in the same place/service
madmalik
@madmalik
the whole chat situation is a mess -.-
Aaron Weiss
@aatxe
Indeed! The one good thing that's come of it though is that I found out that Riot finally hired the UX designer it desperately needed and they're releasing a new UI Soon™.
madmalik
@madmalik
yay :) that sounds good
i hope that works. sometimes its hard to tack on a good UX when it wasn't a consideration from the start
Aaron Weiss
@aatxe
madmalik
@madmalik
that look indeed very nice
all in all, i just may have to accept to have 3-4 different electron apps running... still less hassle than irc
Aaron Weiss
@aatxe
You can always use Rambox and run all the Electron apps in one program. :wink:
madmalik
@madmalik
mhh, i'll try that
Aaron Weiss
@aatxe
There's a couple other apps of that sort, like Franz.
Maybe you'll like one of them. :smile:
madmalik
@madmalik
:D
the first thing that comes to my mind when i hear "franz" in the context of computers i think of franz lisp
Peter Liniker
@pliniker
@madmalik I've just this week logged into the rust-lang discord for the first time after seeing Steve Klabnik's IRLO post. It still looks like there's a fair amount of... discord... about using any single chat platform, my opinion, for now, is wait-and-see
(I'm still working on an immix implementation and documenting it, as I get time, btw. Definitely not inactive!)
madmalik
@madmalik
@pliniker wait-and-see seems fair
and yay for non-inactivity ;) i just started a new job and don't get much done in my freetime right now, but i'm also still working on some stuff related to this group