lockywolf has quit [Remote host closed the connection]
<erkin>
Does only-in have any effect on performance, or is it just a code separation precaution?
<erkin>
My program takes a lot of time to start up and according to the profiler, it's almost entirely made up of requires, so I'm trying to optimise that part.
iyzsong has quit [Read error: Connection reset by peer]
<nisstyre>
tonyg: awesome thanks!
rain2 has quit [Quit: WeeChat 1.6]
iclon_ has joined #racket
rain1 has joined #racket
iclon__ has quit [Ping timeout: 244 seconds]
orivej has quit [Ping timeout: 248 seconds]
Sgeo__ has quit [Quit: Leaving]
q9929t has joined #racket
soegaard has quit [Quit: soegaard]
q9929t has quit [Quit: q9929t]
euhmeuh has left #racket [#racket]
ubLIX has quit [Quit: ubLIX]
soegaard has joined #racket
vraid has quit [Quit: Leaving]
soegaard has quit [Client Quit]
<samth>
erkin: only-in does not affect performance
<samth>
erkin: a few things to do in that situation: use `#lang racket/base` instead of `#lang racket`, eliminate unnecessary requires using the check-requires tool, use lazy-loading with `lazy-require`
miskatonic has joined #racket
casaca has quit [Ping timeout: 258 seconds]
pera has joined #racket
moldybits has quit [Read error: Connection reset by peer]
casaca has joined #racket
moldybits has joined #racket
<greghendershott>
erkin: Also use `raco make` directly or via `raco setup` to compile to bytecode as much of your own stuff as possible
dustyweb has quit [Remote host closed the connection]
dustyweb has joined #racket
<greghendershott>
Mainly how `only-in` helps is preventing surprises -- some future version of a module `provide`s a new binding that happens to conflict with one of your own.
soegaard has joined #racket
casaca has quit [Ping timeout: 258 seconds]
casaca has joined #racket
refpga has joined #racket
miskatonic has quit [Quit: ERC (IRC client for Emacs 24.5.1)]
casaca has quit [Ping timeout: 258 seconds]
casaca has joined #racket
nckx has left #racket [#racket]
clacke_movim has left #racket [#racket]
dbmikus has joined #racket
clacke_movim has joined #racket
casaca has quit [Ping timeout: 245 seconds]
jcowan has joined #racket
casaca has joined #racket
sauvin has quit [Ping timeout: 245 seconds]
casaca has quit [Ping timeout: 245 seconds]
casaca has joined #racket
casaca has quit [Ping timeout: 245 seconds]
soegaard has quit [Quit: soegaard]
simendsjo has joined #racket
orivej has joined #racket
_whitelogger has joined #racket
iclon_ has quit [Remote host closed the connection]
iclon has joined #racket
miskatonic has joined #racket
ym has quit [Quit: Leaving]
pera has quit [Ping timeout: 272 seconds]
soegaard has joined #racket
simendsjo has quit [Remote host closed the connection]
simendsjo has joined #racket
sagax has joined #racket
YuGiOhJCJ has quit [Quit: YuGiOhJCJ]
pera has joined #racket
rubic88 has quit [Remote host closed the connection]
rubic88 has joined #racket
rubic has joined #racket
manualcrank has quit [Quit: WeeChat 1.9.1]
manualcrank has joined #racket
rubic88 has quit [Ping timeout: 246 seconds]
manualcrank has quit [Client Quit]
manualcrank has joined #racket
simendsjo has quit [Ping timeout: 272 seconds]
manualcrank has quit [Quit: WeeChat 1.9.1]
manualcrank has joined #racket
casaca has joined #racket
pera has quit [Ping timeout: 258 seconds]
pera has joined #racket
libertyprime has quit [Ping timeout: 245 seconds]
<nisstyre>
tonyg: I like that goblins is based on ocaps
<nisstyre>
I'm not sure if I would agree the actor model gives you ocaps for free though
<nisstyre>
because it doesn't necessarily give you revocation, unless you implement it yourself using proxy actors or whatever
<nisstyre>
but it looks very up my alley anyway
miskatonic has quit [Quit: ERC (IRC client for Emacs 24.5.1)]
<nisstyre>
looks like it does address revocation though
rubic has quit [Remote host closed the connection]
rubic has joined #racket
Arcaelyx has joined #racket
m1dnight_ has quit [Ping timeout: 272 seconds]
efm has quit [Ping timeout: 258 seconds]
efm has joined #racket
efm has quit [Ping timeout: 248 seconds]
rubic has quit [Remote host closed the connection]
rubic has joined #racket
efm has joined #racket
soegaard has quit [Quit: soegaard]
rubic has quit [Remote host closed the connection]
ubLIX has joined #racket
<jcowan>
Certainly you don't get revocation for free. Actors encapsulate their friends in local variables of the closure. In general I don't know how you'd do revocation at all without either two-way communication or a centralized revocation list.
<erkin>
samth, greghendershott: Thanks for the advice, I'll try to cut down on standard library imports.
<endformationage>
nisstyre: You may also be interested in the COAST architectural style, and its reference implementation: http://isr.uci.edu/projects/coast/
<nisstyre>
endformationage: yeah I saw COAST before but I wasn't too sure what the status of it is
<endformationage>
Yeah, saddly seems no longer developed> :\
<endformationage>
nisstyre: I'm currious have you read the dissertation on it by Gorlick?
<nisstyre>
endformationage: no, I've mainly been reading stuff from Mark Miller and a few other related things
<nisstyre>
like the stuff about sandstorm.io
<nisstyre>
I was actually thinking of trying to use Cap'n Proto for some stuff
Sgeo has joined #racket
<endformationage>
nisstyre: I really like the ideas COAST puts forth. I'm puzzled why there's not more discussion about it amongst the ocap community. You might consider reading the abstract/intro.
<nisstyre>
endformationage: I'll take a closer look
<endformationage>
I came across it via my interest in Actors and ocap.
<nisstyre>
endformationage: it does sound pretty robust
<nisstyre>
I think if it were ported to some other languages (like ES) it would have a better chance of catching on
* jcowan
likes Cap'n Proto
<endformationage>
It embeds computation exchange (mobile code) in object-capability infrastructure. /mouthful
<nisstyre>
endformationage: the problem seems to be that it does so much, it's hard to fully understand what it's capable of (pun intended)
<nisstyre>
the actor model itself is really easy to understand
pera has quit [Ping timeout: 272 seconds]
<endformationage>
I had the same feeling at first, but ended up reading through the dissert., and them through 2 others which used it, and all the other papers I could find on it.
<endformationage>
In the default implementation Motile/Island, Islets are essentially actors.
notzmv has joined #racket
ricekrispie has quit [Quit: YEET]
pera has joined #racket
libertyprime has joined #racket
pera has quit [Ping timeout: 246 seconds]
rubic88 has joined #racket
notzmv has quit [Remote host closed the connection]