Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2014-12-07

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

All times shown according to UTC.

Time Nick Message
08:28 drrho joined #pdl
15:23 vicash mstpan blogs are informative
15:25 vicash sivoais: history repeats itself, especially in the programming world.. IBM mainframes had virtual machine segregation in the 70s
15:26 vicash Node.js will be reimplemented as something else 10 years from now by the new generation of programmers who are too dumb to ignore history
15:28 jberger node is too busy tearing itself apart to do much else atm
15:29 vicash that's the problem with frameworks
15:30 vicash everyone comes and tries their own version, then markets it, then someone comes along and creates something that does the exact same stuff with slightly different syntax and markets it again
15:31 jberger vicash: its worse than that
15:31 jberger they are trying to retroactively own parts of the project
15:31 jberger trademarks, npm etc
15:31 jberger they really are tearing it to pieces
15:31 vicash well that makes sense since after a while a corporation has to make money and they go the Microsoft route
15:32 jberger vicash: assuming the framework is your product then yes
15:32 vicash in this case that is true
15:32 vicash Joyent buying Node was a play to get the ownership
15:33 vicash because they can sell a full stack with their Postgres offering
15:33 jberger I still think a good rule of thumb is, open source the code that is not your money-making business
15:33 jberger then you get the bazaar effect for it
15:33 vicash that is correct
15:33 vicash otherwise you get Node.js, SystemD etc
15:34 jberger and keep your money-niche small enough that you get most of the open source benefit
15:34 vicash but MBAs dont know mathematics
15:34 vicash they only know addition
15:34 * jberger is so happy to be paid to work on a Perl codebase again!
15:35 jberger and at a company with good open-source understanding
15:35 vicash they add the revenue of each Node.js user and say it is worth billions so let's own it
15:35 jberger yes, until they do that and everyone leaves
15:35 * vicash works for myself.. tough at times but whatever i open source is not for making money on it
15:35 jberger node was already precariously positioned as the current fad, ready to fall
15:36 vicash that is because javascript developers have ADHD
15:36 vicash they jump from one framework to another for unknown reasons
15:36 jberger vicash: I would say its because the web community has ADHD generally
15:36 vicash good point
15:37 jberger Perl -> (PHP|Python) -> Ruby -> Node -> (Go?|Elixer?)
15:37 Mithaldu adult view: it's because most of them are very early beginners with little experience
15:38 vicash Mithaldu: agreed. also they dont want to understand how things work in extreme detail
15:38 jberger also, non-technical business-types want to build on the newest best thing
15:38 Mithaldu yes because they are beginners
15:38 vicash so they move from fadto fad->next
15:38 Mithaldu that will change over time if they have the mental capacity
15:39 Mithaldu frankly, you're passing judgement on people you don't even know :v
15:39 vicash that's true.. i learnt Perl 10 years ago for a parsing project and then looked at Java/Ruby/Python and said why need those at all.. still doing Perl..
15:39 vicash hahaha... similar to the judgments they pass that Perl is dead
15:39 jberger I'm starting to formulate an understanding of languages/frameworks as their ability to survive their eventual drop in popularity
15:39 Mithaldu yes, both are bad :v
15:41 jberger some of it has to do with the amount of adoption they had before the drop, sure
15:41 vicash jberger: that would make a good blog post.. if you had some historical data  on usage you could plot it using PDL+plotting library of choice
15:41 jberger some of it then is how good the thing was versus how much was just hype
15:42 vicash that's true like Ruby's string parsing was really slow in version 1.8 and may still be .. havent used it
15:42 jberger vicash: agreed, its something I've been thinking a lot about, but I've not been able to boil it down to something that sounds sane and unbiased
15:43 vicash jberger: make it like a non-static blog post.. something that updates the stats on a daily basis or a monthly basis.. then it wont be stale or unbiased since all you're showing is stats and discussing it
15:43 jberger Ruby was essentially an academic language that only gained popularity because of Rails, then once Rails-like frameworks popped up in other languages, Ruby has seen a drop off
15:44 jberger interestingly, heavy frameworks like Rails seem also to be dropping off somewhat
15:44 vicash yea I saw an image on Twitter from some conference citing Rails is dead
15:44 jberger though there will always be people who prefer a baked-in (opinionated) model layer for ease of use
15:44 jberger Rails/Django etc
15:45 jberger if I had to start a new project in Python now I would probably choose Tornado or something like it
15:46 jberger python's "yield from" continuations as webapp primatives is an interesting concept, I wonder if people will actually be able to grok it to the point of using it
15:46 vicash i normally use Dancer for my small web interfaces since i mostly dont do web development.. but i will try Mojo for the next one i build if I do .. i am interested in using the threading capabilities you wrote about a few months ago
15:47 Mithaldu try Web::Simple :v
15:47 jberger N.B. not threading, non-blocking in a single thread
15:48 vicash that's right.. non-blocking in single thread..
15:48 Mithaldu just say event-based :v
15:48 vicash haha.. yea like Node.js :)
15:48 jberger Mithaldu disagrees on Mojo, we agree to disagree on that :-)
15:48 vicash does it use AnyEvent internally ?
15:48 jberger vicash: it can, it doesn't have to
15:49 jberger it uses IO::Poll if EV is not installed
15:49 vicash Mithaldu: i have to try Web::Simple as well...
15:49 Mithaldu \o/
15:50 vicash jberger: i will be interested to hear Marc Lehmann's input on IO::Poll vs EV, biased or unbiased.. he is generally pretty blunt about the various event libs
15:50 jberger I thought mst's opinion of his own framework (W::S) was interesting here: http://shadow.cat/blog/matt-s-trout/mstpan-1/
15:50 jberger I have a feeling I know what shmorp would say
15:50 jberger I don't especially care anymore
15:51 Mithaldu jberger: fwiw, sri recently gave me another reason to distrust him
15:51 Mithaldu in an interview he described his catalyst situation as him having left it
15:51 jberger we are actually hoping that eventually some good bindings to libuv will show up and we can make a third Mojo::Reactor for it
15:51 jberger Mithaldu: you really aren't going to convince me
15:52 Mithaldu of what?
15:52 jberger mst and sri have made peace over the catalyst situation, and I'm ok with that
15:52 Mithaldu that's fine, still doesn't mean he needs to distort the facts
15:52 * vicash is afk
15:52 Mithaldu maybe it's a german thing
15:52 Mithaldu but i expect honesty from the people whose software i use
15:53 jberger its his own interview and would read it as trying to move past history to just get on with things
15:53 jberger we was there, now he isn't
15:54 jberger phrase that as you will
15:54 jberger I wasn't a part of the situation, mst and I have discussed at length and he is satisfied, so I am too
15:54 jberger mst spends plent of time in #mojo actually
15:55 Mithaldu he also spends plenty of time in #dancer
15:55 jberger sure
15:55 jberger I'm just saying that he doesn't seem to bear any ill-will
15:55 * jberger should spend more time in #dancer
15:55 Mithaldu honestly, you're looking at things happening and drawing entirely skewed conclusions
15:55 jberger ETOOMANYCHANNELS
15:56 Mithaldu he spends a lot of time in dancer because people use it and because it is, mildly put, not a very competent software
15:56 Mithaldu it's entirely possible to consider a thing not fit for use, but still provide help to those who decide to use it because said help benefits the perl community
15:57 Mithaldu same reason why i gave dancer2 a bunch of patches for portability despite considering its code to be terrifying
15:57 jberger Mithaldu: I'm saying I like the people who work on it and would like to engage with them more
15:57 jberger Mithaldu++ # helping
15:58 Mithaldu if mojolicious had issues on windows i'd be providing patches too :P
15:58 Mithaldu luckily they have plenty competence for that
15:58 jberger but it doesn't because love him or hate him, sri writes good code
15:59 Mithaldu you get to claim that once i can change the serializer of a mojo session freely
16:00 jberger again, Mithaldu I'm ok if you have your reservations because of historical reasons, but if the major players have moved past it, then I don't see why you have to keep stirring it up
16:00 jberger a good framework/codebase with a murky history or a "terrifying" one
16:01 Mithaldu now you're conflating what i'm saying about mojo and dancer
16:01 jberger Mithaldu: write a subclass of Mojo::Sessions and use it here: http://mojolicio.us/perldoc/Mojolicious#sessions
16:01 jberger I would be surprised if someone else hasn't done that already
16:02 jberger oops, Mojolicious::Sessions
16:02 Mithaldu actually, after reading the code that's not even necessary anymore
16:03 Mithaldu also, as i said above
16:03 Mithaldu i expect honesty
16:04 Mithaldu maybe that's not a requirement for you, but someone who can't stand by what they said or did is the minimum for software i'm expected to recommend to other people with a calm heart
16:04 Mithaldu s/can't/can/
16:04 jberger its a puff piece article! do you really want him to say "yeah, everything sucked and people were pissed and blah blah" every time Catalyst comes up?
16:04 Mithaldu consider that my day job is making things work for other people
16:05 Mithaldu if i tell someone "use that" and they end up having problems with "that", then i get flak
16:05 Mithaldu and that is not a theoretical point, that is my daily reality
16:05 jberger Mithaldu: so you must never use any of schmorp's stuff either right?
16:06 Mithaldu if i do, then only with a giant warning
16:06 jberger his history is much darker than sri's
16:06 Mithaldu it's simultaneously much lighter
16:06 Mithaldu schmorp is an asshole, but he's completely open and honest about it
16:06 Mithaldu also, as to your other question: a simple answer is "let's stick to talking about software"
16:06 jberger Mithaldu: I think you have a personal qualm with sri, I think it clouds your vision
16:07 jberger but you aren't!
16:07 jberger you want him to talk about history all the time
16:07 Mithaldu no?
16:07 Mithaldu i guess an even better answer would be
16:07 jberger he can't very well deny that he was a part of Catalyst, so the euphemism is that he left, because it doesn't need to come up
16:07 Mithaldu "we wanted different things from the project and they asked me to leave"
16:08 Mithaldu but even i can understand why he wouldn't want to say that
16:08 jberger and then he did what? he left
16:08 Mithaldu now you're talking politics word-turning
16:08 jberger again, I think you are reading malice into a comment that was a throwaway to move on to talking about the present
16:09 jberger Mithaldu: I WOULDN'T TALK ABOUT THIS AT ALL
16:09 jberger it wasn't an issue for me and you are talking about "honesty"
16:09 jberger nm, I'm not interested
16:09 jberger o/
16:09 Mithaldu jberger: i think you need to calm down :)
16:10 jberger Mithaldu: please just keep it to yourself ok
16:10 jberger I don't need you to like sri
16:10 jberger but I really think that the project is a good one for bringing node-hackers to perl
16:10 Mithaldu look man, i was having a calm little chat about a thing that happened
16:11 jberger its a good way to show that Perl can be modern outside of the echo chamber
16:11 Mithaldu i think you're interpreting much too much emotion into what i'm saying
16:11 jberger Mithaldu: no, you were injecting history in a place it didn't need to be
16:11 Mithaldu in fact, i'm actually impressed that he fixed ::Sessions
16:11 Mithaldu jberger: he did that interview quite recently
16:11 Mithaldu but then again
16:12 Mithaldu i tried to explain my reasons for why i consider it relevant
16:12 Mithaldu and you don't get them
16:12 Mithaldu maybe because you work in a completely different industry as i do
16:12 * jberger rereads
16:12 Mithaldu but that's fien :)
16:12 Mithaldu also keep this in mind:
16:13 Mithaldu a small criticism doesn't mean i think he's hitler
16:13 jberger "jberger: fwiw, sri recently gave me another reason to distrust him" # that is pretty inflamatory in my mind
16:13 Mithaldu nah, that's just an observation
16:13 Mithaldu like
16:13 Mithaldu the outcome i'd expect to come from something to that is along the lines of
16:14 jberger everyone agrees that the Catalyst split was bad, but most parties are happy with the end result, so can't we just let a throwaway euphenism in an interview just be that?
16:14 Mithaldu "huh, what happened?" ... "ah, that's not perfectly excellent, maybe next time i am interacting with him i'll pay attention to whether something like that happens again and try and do something to minimize it"
16:15 Mithaldu the end result was "we don't talk about the matter"
16:15 Mithaldu that's what everyone was happy about
16:16 Mithaldu actually, here, let me rephrase my mind concisely
16:17 Mithaldu i think sri still needs to improve on a personal level, and i think it'll help him, mojo, and the perl community, if the people who want to be around him and communicate with him regularly are aware of that and help him with that
16:18 jberger I can agree with that, and I actually think for the most part he is
16:18 Mithaldu he is?
16:18 jberger he still has some tendency to deal with derp loudly, but we try to reign that in
16:18 jberger improving
16:18 Mithaldu :)
16:18 Mithaldu yes, i'm honestly seeing that too
16:18 Mithaldu and that's great to hear
16:19 jberger actually, one of my biggest things in the mojo community is keeping him busy, he seems to be most pleasant when he's got a project
16:19 Mithaldu likely
16:19 jberger part of why I'm glad the focus has shifted to Pg from Mongo, he has plenty to do
16:20 Mithaldu postgres?
16:20 Mithaldu mojo had a focus on mongo?
16:20 jberger Mango
16:21 Mithaldu now i'm very confused
16:21 jberger non-blocking interface to mongodb using Mojo::IOLoop and a full implementation of the wire protocol
16:21 jberger the problem is that the company behind mongodb keep moving the goalposts for wire protocol (and not announcing or even documenting the changes)
16:22 jberger so he got tired of chasing and gave the project to a volunteerr
16:22 jberger now Mojo::Pg is the focus for database layer
16:22 Mithaldu that makes me happy
16:23 jberger not that it is in any way "blessed" or preferred in Mojo, they are just related in that they make use of the Mojo::IOLoop abstraction
16:23 Mithaldu all i know is, mongo shouldn't be used :P
16:23 jberger ahaha, yeah, we all found things to dislike
16:23 Mithaldu and short of oracle, pg is the best solution
16:24 jberger IIRC some company paid him to write Mango (and open source it), and because it was there we started to use it
16:24 Mithaldu probably some berlin company
16:25 jberger or Oslo, there is a lot of mojo in norway thanks to marcus and batman
16:25 jberger and sjn
16:25 jberger it was really nice to have a fully non-blocking db layer
16:26 jberger Mojo::Pg is only mostly non-blocking due to limitations of DBI
16:26 Mithaldu i'm not even sure how that's useful
16:26 Mithaldu unless you can start fetching rows while the query is still going
16:27 jberger multiple queries simultaneously, especially if they are both long-running
16:27 jberger the transport is blocking
16:27 Mithaldu huh, ok
16:28 jberger its not as useful as Mango was from a non-blocking perspective
16:28 Mithaldu i was gonna say, i don't do something like that often
16:28 jberger there is a proposal for Pg to get a native REST API, which would make it very easy to do fully non-blocking using Mojo::UserAgent
16:29 jberger but that is still on the todo list for the Pg folks
16:29 Mithaldu hahaha, cute
16:30 Mithaldu also, to your earlier question
16:31 Mithaldu what schmorp soft (don't forget the c) do you think i'd be using?
16:31 jberger https://wiki.postgresql.org/wiki/HTTP_API
16:31 jberger AnyEvent
16:31 jberger JSON::XS
16:31 jberger Coro perhaps
16:31 jberger hard to avoid for most people
16:32 jberger I was just making a point
16:32 jberger but I'm not actually interested in the result
16:32 Mithaldu anyevent: a while back i liked it for the sane ai (and only used it once for a parallel downloader), meanwhile i've switched to io::async, because leonerd is really nice
16:32 jberger I use schmorp stuff too when I have to
16:32 jberger LeoNerd++
16:32 Mithaldu next up is: https://metacpan.org/pod/JSON::MaybeXS
16:33 jberger there is an interface between IO::Async and Mojo::IOLoop somewhere
16:33 Mithaldu btw, here's the thing i'm doing with io::async: https://github.com/wchristian/Microidium
16:33 jberger IIRC it does things in sort-of the opposite direction (these things make my brain hurt sometimes)
16:33 Mithaldu not anywhere near production-ready, but working surprisingly well
16:33 * jberger looks
16:34 Mithaldu i really need to put a screenshot on there
16:34 jberger yeah
16:34 Mithaldu and coro i used for exactly one project:
16:34 Mithaldu a dwarf fortress 3d viewer 5+ years ago
16:35 jberger I wrote a Coro based python-like generator with it
16:35 jberger I still kinda like it, but I don't actually use it for anything
16:35 jberger Perl just rarely needs true generators
16:36 jberger https://metacpan.org/pod/Generator::Object
16:36 Mithaldu crap, i forgot i was revamping the OpenGL in microidium and now everything is grass textures
16:37 jberger here is the IO::Async / Mojo::IOLoop thing I remembered: https://metacpan.org/pod/IO::Async::Loop::Mojo
16:37 jberger I would probably prefer to have Mojo::Reactor::IOAsync, but I haven't had the motivation
16:37 jberger then again, I really would like it ....
16:37 * jberger adds to todo list
16:38 jberger Mithaldu: does that mean I should wait before trying it?
16:38 Mithaldu trying microidim?
16:38 Mithaldu +u
16:39 jberger yeah
16:39 jberger I was just about to
16:39 Mithaldu nah, master is fine for messing about
16:39 Mithaldu my opengl branch is still local
16:39 jberger k
16:39 Mithaldu just make sure you have all the deps the readme mentions
16:39 Mithaldu https://dl.dropboxusercontent.​com/u/10190786/microidium.png
16:40 * jberger is on a mac (and still new at it)
16:40 Mithaldu fuck
16:40 Mithaldu it doesn't mention any
16:40 jberger will that present any issues?
16:40 Mithaldu hell if i know
16:40 Mithaldu but uh, sec
16:41 Mithaldu actually, since i used SDL it should be fine
16:41 Mithaldu though installing the latest SDL might be tricky
16:41 Mithaldu i thought i was already using my bleeding edge OpenGL
16:42 jberger my new $job wanted to buy me a MacBook Pro, I already had a linux lappy, but if they were buying  :-)
16:43 Mithaldu macbooks seem to be a good choice if all one wants is a decent unix box
16:43 jberger I'm still getting used to it
16:44 jberger the keyboard is just different enough
16:44 jberger and homebrew isn't quite apt
16:44 * Mithaldu doesn't bother with laptop keyboards
16:44 Mithaldu i have a usb one
16:44 jberger Mithaldu: that was basically my response, at least when I'm really working
16:44 jberger which is when I notice
16:46 jberger Mithaldu: can I just use the runtime deps list from the META.json?
16:46 Mithaldu add to those:
16:46 Mithaldu IO::Async SDL::Mixer Sereal
16:47 Mithaldu wooo
16:47 Mithaldu https://github.com/wchristian/Microidium
16:48 Mithaldu and fresh meta..json pushed
16:49 Mithaldu also, pushed my opengl branch if you'd like to laugh at the stupidity in that
16:49 jberger perl -Mojo -E 'say for keys j(b(shift)->slurp)->{prereqs}{runtime}{requires}' META.json | cpanm
16:49 Mithaldu nice :)
16:50 Mithaldu i forgot that in modern perls keys can work on hashrefs
16:50 jberger hmmm Alien::SDL failed
16:50 Mithaldu i think that one's only for windows
16:51 jberger yeah, I try not to do it that often, actually I really dislike the hashref thing, it has caused me troubles
16:51 Mithaldu try installing sdl manually?
16:51 jberger I hope it will go away with postderef
16:51 jberger I will, homebrew to the rescue
16:53 jberger yeah, that worked fine
16:53 jberger I assumed as much
16:53 jberger Alien::Base has made quite a bit of progress, fwiw
16:54 jberger I have given it over to a consortium of interested users
16:54 jberger I just keep a watchful eye
16:54 jberger mostly I just make sure that they don't resort to "lets have this install X system-wide" which is always a temptation
16:55 Mithaldu hahaha
16:55 Mithaldu i really need to look at that sometime
16:55 Mithaldu it'll probably help me get opengl stuff to be nicer
16:55 jberger my golden rule is that installing a CPAN module should NEVER changes the system outside of the current Perl (on the current user)
16:56 jberger chm would like it too I'm sure
16:56 jberger hey we finally are back to #pdl :-)
16:57 Mithaldu well he's going a different way, since last i heard he's working on making opengl.pm work with glew
16:57 Mithaldu i just want to make stuff work so i can do stupid things :P
16:58 jberger do I need to run both the server and the app?
16:58 Mithaldu nope
16:58 Mithaldu the app is standalone
16:58 Mithaldu and if you start it twice, then the server, then connect, they'll network
16:58 Mithaldu note:
16:58 Mithaldu a client will run on AI until you hit a key
16:58 jberger SDL::Constants was not installed via the one-liner above
16:59 jberger fyi
16:59 Mithaldu then it'll wait for a minute i think before retaking control
16:59 Mithaldu did you fetch my latest push before doing that?
16:59 Mithaldu SDL::Constants is part of the SDL dist
16:59 jberger oh, damn
16:59 Mithaldu :v
16:59 jberger Alien::SDL didn't pass
17:00 * jberger missed it in the output
17:00 * jberger will actually have to read the log output now
17:01 Mithaldu haha
17:02 jberger cannot include SDL.h
17:02 jberger ick
17:03 jberger https://github.com/Homebrew/homebrew/issues/29487
17:03 jberger sigh
17:03 Mithaldu :(
17:04 Mithaldu reopen it? :P
17:05 jberger brew install sdl completes startlingly fast
17:05 Mithaldu you're used to tests
17:05 jberger maybe
17:06 jberger brew has typically been faster than cpan for that reason, but this is even more dramatic
17:08 jberger Oh well, I'll try to remember to check it out on my Linux laptop soon
17:12 jberger other places are suggesting specifying the path to the headers
17:12 jberger might try that later too
17:13 Mithaldu not quite sure if it'll work better on linux
17:13 Mithaldu but any feedback is useful feedback :)
17:14 jberger I've gotten sdl working on Linux before
17:15 Mithaldu :D
17:15 Mithaldu btw
17:16 Mithaldu it might also need some gfx card stuff, not sure about that
17:16 Mithaldu also
17:16 Mithaldu turn down the volume before trying
17:20 jberger haha, noted
17:20 Mithaldu i should try and make a video of how it works
17:24 jberger Mithaldu: so I think this is why there isn't a Mojo::Reactor::IOAsync: https://metacpan.org/pod/I​O::Async::Loop#DESCRIPTION
17:25 jberger it just uses IO::Poll too
17:25 Mithaldu i've no idea about these things
17:25 Mithaldu all i know is it works fine on linux and windows :)
17:25 jberger right, because IO::Poll does
17:25 Mithaldu and i trust leonerd to know his shit about event systems
17:26 jberger which is why Mojo::Reactor::Poll works fine on those platforms too
17:26 jberger Mithaldu: indeed
17:27 jberger what I'm saying is that there is almost no difference in how they work, so probably no one has thought it worth the effort to link them
17:27 jberger there would be some value in it though
17:27 jberger I will try to get to it at some point
17:28 Mithaldu i know fairly little about Reactor, so i couldn't opine
17:28 Mithaldu but i'm all for more io::async in things
17:32 jberger basically since they use the same loop logic, it wouldn't be hard to merge
17:32 jberger the only thing is to make sure that they use the same actual loop
17:34 jberger as it stands if you tried to use one from the other, the latter would block the former (whichever direction) just because they would each have their own "while(1) {}"
17:34 jberger which is essentially all a vanilla ioloop is
17:34 Mithaldu right
17:34 Mithaldu honestly, most of it is still over my head :)
17:35 jberger tbh, I learned a lot just from reading this: https://github.com/kraih/mojo/blo​b/master/lib/Mojo/Reactor/Poll.pm
17:36 jberger and mostly just the one_tick and start methods
17:36 jberger everything else falls out naturally
17:36 Mithaldu i mainly don't care to learn those details
17:36 Mithaldu i've plenty in my head with the design structure of an action game network system
17:36 jberger true enough
17:37 jberger anyway, it has developed into a pleasant chat, but I have other things coming up, so o/
17:37 Mithaldu wanna see a fun video where that is explained?
17:37 Mithaldu aight, later o/
17:37 jberger sure, I'll bookmark it at least
17:37 Mithaldu sec
17:37 Mithaldu http://www.gdcvault.com/play/101​4345/I-Shot-You-First-Networking
17:38 jberger I think I may have read a text version of something very similar (exactly similar?)
17:39 Mithaldu unlikely
17:39 Mithaldu you may have read a pdf on the starsiege tribes networking model
17:39 Mithaldu this is the evolution on that with description of modern real life use cases
17:40 jberger this: http://gameprogrammingpatterns.com/game-loop.html
17:40 jberger probably not the same people
17:40 jberger anyway, I will watch soon
17:40 jberger o/
17:40 Mithaldu that's a good read, but quite different :)
17:40 Mithaldu ciao
19:33 pdurbin jberger: ah, this makes sense to me (from the link you dropped earlier): "a thing called the mstpan which is basically my opinionated version of Task::Kensho"
19:34 pdurbin and in addition to Tornado (and Twisted), I recently came across this for Python: http://circuitsframework.com
22:20 mohawk joined #pdl

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