Perl 6 - the future is here, just unevenly distributed

IRC log for #crimsonfu, 2014-12-18

crimsonfu - sysadmins who code

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

All times shown according to UTC.

Time Nick Message
00:52 marcoceppi joined #crimsonfu
00:56 hydrajump can anyone recommend a rails developer whose available for a couple of days job working on an existing rails 4 app?
00:57 prologic permissive?
00:57 prologic hydrajump if you said python I'd say sure yeap me :)
00:57 prologic :P
00:58 hydrajump hehe thanks :D
00:58 hydrajump not easy finding good devs who actually do what they say they will
00:59 prologic that's true
01:17 pdurbin those lazy devs
01:32 prologic yeah :P
01:32 prologic damn them all to hell :)
01:32 prologic oh wiat
01:32 prologic I'm a dev
01:32 prologic rats :(
01:32 prologic got my new server working last night
01:32 prologic a little VX11 CRUX VM at hetzner.de
01:33 pdurbin prologic: if you want URLs to be clickable in the logs made by ilbot3, you have to add http://
01:43 prologic heh
01:43 prologic http://hetzner.de
01:43 prologic there :P
01:44 prologic anyone heard of this statusnet / gnusocial thing?
01:44 prologic any opinions?
01:51 pdurbin statusnet is no more, but yes it lives on a gnusocial, I guess. the guy who wrote statusnet has since created http://pump.io
01:54 prologic oh?
01:54 * prologic looks
01:55 pdurbin prologic: node.js! async! you'll love pump.io!
01:57 prologic https://1realtime.net/prologic
01:57 prologic cool :)
01:57 prologic I created an account on their hosted version of pump.io
01:57 prologic I'll try it out over the next few days
02:06 * pdurbin has an account at https://microca.st/pdurbin ... "2 years ago" on the more recent post
02:07 pdurbin prologic: I left some notes here: http://irclog.greptilian.com/spanworm/2013-02-26#i_3540
02:08 pdurbin looks like I opened this ticket: https://github.com/e14n/pump.io/issues/440
03:45 prologic So i pump.io decentarlized ?
09:28 * prologic is hacking up a gopher daemon in circuits
09:28 prologic :)
11:29 pdurbin prologic: yes. decentralized and federated. rare in that way
13:59 prologic https://bitbucket.org/prologic/cgod
13:59 prologic pdurbin, cool so I should run a pump instance on my server yeah?
13:59 prologic get in on this oss/decentralized/federated social crap :)
14:28 dotplus prologic: "circuits"?
14:28 dotplus I don't think you can mean ASIC, right?
14:33 prologic ASIC?
14:33 prologic https://pypi.python.org/pypi/circuits
14:37 pdurbin this confusion never comes up in #circuits ;)
14:37 prologic no indeed :)
14:48 dotplus https://en.wikipedia.org/wiki/Application-specific_integrated_circuit
14:53 rruma joined #crimsonfu
14:59 dotplus huh. circuits sounds interesting. the purpose of it is to make it easy to build network service daemons?
15:02 prologic sure
15:03 prologic or web apps
15:03 prologic or anything that does I/O
15:06 dotplus how does it do performance-wise? multi-thread/multi-process? python for that sort of thing just makes me think of multiprocessing and "fails at anything but small scale", but that might be because the primary app I'm thinking of (bcfg2) was horrid.
15:10 dotplus mrrh, you people are too interesting. I have work I'm supposed to be doing
15:12 prologic async I/O
15:13 prologic event-driven
15:13 prologic real world performance is acceptable for what the framework does
15:13 prologic raw banchmarks put circuits (the core) at ~50k events/s on my hw and ~150k event/s with PyPy
15:14 prologic circuits.web fairs far worse but it's quite a feature rich web framework
15:14 prologic so benefits from load balancing :)
15:15 prologic circuits is design to scale :)
15:15 prologic you should do as little CPU bound work as possbile in any given event handler
15:15 prologic CPU bound work should be shoved off to a Worker (circuits.Worker)
15:57 tfhartmann joined #crimsonfu
16:13 rruma left #crimsonfu
18:48 tfhartmann joined #crimsonfu

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

crimsonfu - sysadmins who code