<    March 2017    >
Su Mo Tu We Th Fr Sa  
          1  2  3  4  
 5  6  7  8  9 10 11  
12 13 14 15 16 17 18  
19 20 21 22 23 _2_4 25  
26 27 28 29 30 31
00:00 rschmukler joined
00:01 cdg joined
00:07 mozzarella joined
00:12 markh joined
00:13 MotherFlojo joined
00:36 cemilowski joined
00:41 vendethiel joined
00:43 tofutoon joined
00:53 codestorm joined
00:54 nighty joined
00:55 modlin joined
01:03 potatosalad joined
01:06 bvcosta joined
01:07 technikhil[m] joined
01:07 beatpanic[m] joined
01:07 M107262[m] joined
01:07 M-Quora joined
01:07 bolton joined
01:07 M-manveru joined
01:07 karambirMatrix joined
01:08 DLSteve joined
01:08 refriedchicken joined
01:10 mindflayer[m] joined
01:10 dp[m] joined
01:10 OAK0[m] joined
01:15 notriddle joined
01:15 M-nickgal joined
01:16 uhoreg joined
01:16 malteof[m] joined
01:16 nolan_d joined
01:16 seequ_ joined
01:16 prose[m] joined
01:16 TheGillies joined
01:16 timmo joined
01:17 cdg joined
01:18 rkazak joined
01:32 io_bora joined
01:35 rschmukler joined
01:41 chriswarren joined
01:49 chriswarren joined
01:51 rschmukler joined
01:58 chriswarren joined
02:04 rschmukler joined
02:05 jerel joined
02:05 willyk joined
02:07 chriswarren joined
02:12 chriswarren joined
02:12 w-t-h_ joined
02:12 myronmarston joined
02:14 rschmukler joined
02:21 hotpancakes joined
02:25 <hotpancakes> I'm not sure if this is kosher or not, but I forgot to add a field when I ran `mix phx.gen.html`, so I just ran it again with the additional field appended. I opted to overwrite all files, but the context boundary module now has duplicate functions in it. Is this expected behavior?
02:30 gvaughn joined
02:31 icecreamcohen joined
02:33 MotherFlojo joined
02:37 cdg_ joined
02:41 tomterl joined
02:45 hotpanca_ joined
02:45 spirithuang joined
02:47 cdg joined
02:51 rschmukler joined
02:54 spirithuang left
03:02 markholmes joined
03:10 hotpancakes joined
03:13 Aftans joined
03:13 harfangk joined
03:14 systemfault joined
03:15 FIFOd joined
03:16 willyk joined
03:16 efu left
03:17 efu joined
03:19 rschmukler joined
03:20 gvaughn joined
03:21 vendethiel joined
03:24 r00tl3ss joined
03:27 proflonghair joined
03:29 raycoll joined
03:30 <Aftans> I was wondering why everyone seems to authenticate the websocket channel with some seperate signature scheme such as Phoenix.Token or JWT.
03:30 <Aftans> Phoenix already maintains a session using a signature scheme. The most simple case I would imagine is checking for the sent session cookie in the "Connection: Upgrade" request.
03:30 <Aftans> I read an answer from José Valim here: https://stackoverflow.com/questions/34446712/in-elixir-phoenix-how-to-get-session-in-other-module-as-a-module-use-phoenix-c
03:30 <Aftans> But I don't really see why the "Connection: Upgrade" request would be more insecure than any other HTTP client request. Plug.Conn is indeed not available in Phoenix.Channel, but the server does receive the cookie value at that point.
03:30 <Aftans> Does the connection/2 method of Phoenix/Channel always and only get triggered by an "Connection: Upgrade" client request?
03:34 MotherFlojo joined
03:36 raycoll joined
03:39 sfbw joined
03:45 <Aftans> Upon re-reading, José probably refers to Websockets not respecting the Same Origin Policy, which can be mediated by checking the Origin header (which Phoenix does by default with check_origin)
03:49 Guest89756 joined
03:50 MotherFlojo joined
04:00 myronmarston joined
04:07 mozzarella joined
04:08 gvaughn_ joined
04:08 racycle_ joined
04:09 modlin joined
04:12 codestorm777 joined
04:16 myronmarston joined
04:16 w-t-h_ joined
04:48 fowlduck joined
04:55 craigp joined
04:56 wtetzner joined
05:00 potatosalad joined
05:01 gvaughn joined
05:01 gjaldon__ joined
05:02 gvaughn joined
05:04 Nezteb joined
05:04 _Nezteb joined
05:14 Nezteb_ joined
05:23 squalloser joined
05:33 wtetzner joined
05:34 MotherFlojo joined
05:36 zabirauf joined
05:38 kommen joined
05:39 hotpancakes joined
05:50 rschmukler joined
05:50 fowlduck joined
05:55 hotpanca_ joined
05:55 montanonic joined
06:03 icanhazbroccoli joined
06:05 dec0n joined
06:11 alexarnon joined
06:26 willyk left
06:30 montanonic joined
06:35 icanhazbroccoli joined
06:37 Tica2 joined
06:42 sfbw joined
06:43 raycoll joined
06:51 fowlduck joined
06:59 MaSven joined
06:59 potatosalad joined
07:01 oetjenj joined
07:05 codestorm joined
07:10 dignifiedquire joined
07:18 kommen joined
07:21 kommen_ joined
07:21 kommen_ joined
07:26 nanohest joined
07:28 dch_ joined
07:38 cemilowski joined
07:42 sfbw joined
07:42 hotpancakes joined
07:52 fowlduck joined
07:53 apotry joined
07:55 tuacker joined
07:59 codestorm joined
07:59 Cohedrin joined
08:00 Guest67__ joined
08:01 josevalim joined
08:04 refriedchicken joined
08:06 acscherp joined
08:07 mattyw joined
08:16 ZippoWeb joined
08:18 hotpancakes joined
08:19 proteusguy joined
08:20 haennar joined
08:21 zabirauf joined
08:24 cemilowski joined
08:28 beatpanic[m] joined
08:29 refriedchicken joined
08:31 nd__ joined
08:39 PaReeOhNos joined
08:44 tofutoon left
08:48 squalloser joined
08:49 inoas joined
08:51 flaviodesousa joined
08:52 fowlduck joined
08:53 codestorm joined
08:56 imack joined
08:56 <the_voice_> Is there a need for sticky sessions when using phoenix cookie based sessions?
08:57 MotherFlojo joined
09:02 hbraun joined
09:02 lexmag joined
09:05 M107262[m] joined
09:05 M-manveru joined
09:05 bolton joined
09:05 TheGillies joined
09:05 prose[m] joined
09:05 seequ_ joined
09:05 karambirMatrix joined
09:05 nolan_d joined
09:05 malteof[m] joined
09:05 notriddle joined
09:05 timmo joined
09:05 dp[m] joined
09:05 mindflayer[m] joined
09:05 OAK0[m] joined
09:05 uhoreg joined
09:05 technikhil[m] joined
09:05 M-Quora joined
09:05 M-nickgal joined
09:11 zabirauf joined
09:12 rschmukler joined
09:13 the_voice_ left
09:14 the_voice_ joined
09:14 PaReeOhNos joined
09:22 sumi joined
09:22 thiagoalessio joined
09:25 refriedchicken66 joined
09:29 codestorm joined
09:40 achauvin joined
09:42 lexmag joined
09:51 codestorm joined
09:53 codestorm777 joined
09:53 fowlduck joined
10:05 rschmukler joined
10:12 proteusguy joined
10:15 dnorris joined
10:23 marr joined
10:25 vmoravec joined
10:31 mattyw joined
10:31 haennar joined
10:40 qatanah joined
10:44 mattyw joined
10:45 dnorris joined
10:46 codestorm joined
10:52 cemilowski joined
10:53 gmcabrita joined
10:54 fowlduck joined
10:57 MotherFlojo joined
10:58 sixstone joined
11:11 Guedes joined
11:11 Guedes joined
11:15 pbrooko joined
11:16 sumi joined
11:28 simonoram joined
11:28 cristianR_ joined
11:33 cemilowski joined
11:40 codestorm joined
11:44 bvcosta joined
11:49 <OliverMT> bitwalker: ping?
11:51 efu joined
11:53 rschmukler joined
11:55 fowlduck joined
11:56 willyk joined
12:01 nighty joined
12:04 cristianR_ joined
12:05 zv joined
12:19 cemilowski joined
12:19 pbrooko_ joined
12:25 rschmukler joined
12:34 codestorm joined
12:39 <whodidthis> i was ready to throw my bluetooth headphones in the garbage after watching "elixir adoption at scale" talk, but looks like the noise was from the video
12:40 <whodidthis> who knew you could have jump scares on conference videos
12:55 fowlduck joined
12:58 <ciawal> whodidthis:
12:58 MotherFlojo joined
12:59 gvaughn joined
13:00 <OliverMT> haha, that conference keeps delivering complaints about audio for months
13:01 gvaughn joined
13:03 gvaughn_ joined
13:09 slackbotgz joined
13:09 <Nicd-> audio problems are bad to the speaker as well :( I held my first real presentation yesterday and the mic kept cutting off when I looked left
13:09 <Nicd-> it was really distracting
13:10 <Nicd-> but yay, Elixir presentation and I didn't die! :) thanks for the people here that gave their suggestions
13:11 cristianR_ joined
13:11 haennar joined
13:12 <whodidthis> i'm talking about the lonestar one, this one has quite the punch to it and is not recommended if using headphones/earplugs
13:17 gvaughn joined
13:20 haennar joined
13:22 dch_ joined
13:22 PaReeOhNos joined
13:28 codestorm joined
13:31 jeregrine joined
13:34 <asonge> is there a way to run a test in complete isolation?
13:34 <asonge> i am trying to think of a way to write a test for timing attacks against authentication
13:35 <asonge> (do 100 runs of each type of auth, take the average in some way, make sure one doesn't exceed the other by too much)
13:40 <nyaray> Nicd-: will the slides be up anywhere?
13:40 rschmukler joined
13:47 cdg joined
13:47 cristianR_ joined
13:49 simonoram joined
13:49 <Nicd-> nyaray: well they're not worth much on their own: https://docs.google.com/presentation/d/1BxPmksfgVsS7ela-jS3tYI3W2PFb08vJVFkC8vMwj5M/edit?usp=sharing
13:49 <Nicd-> hope there's nothing terribly incorrect there…
13:52 <nyaray> failed slide graveyard :D
13:52 <nyaray> I'll steal that concept...
13:55 <Nicd-> I spent too much time on them to just throw them away...
13:55 rschmukler joined
13:55 Or1on joined
13:56 fowlduck joined
13:58 stephen_m joined
13:58 robozahn joined
13:59 alex88 joined
14:04 ramortegui joined
14:08 <celyr> Let it crash lol
14:09 alexarnon joined
14:20 AckZ joined
14:22 codestorm joined
14:23 gvaughn joined
14:26 sfbw joined
14:27 spinningtopsofdo joined
14:28 rschmukler joined
14:29 BennyP101 joined
14:30 BennyP101 left
14:31 chrismccord joined
14:32 Helmclever joined
14:37 Schalla joined
14:43 rschmukler joined
14:45 bvcosta joined
14:50 hbraun joined
14:52 sfbw joined
14:56 smferris joined
14:57 fowlduck joined
14:57 rschmukler joined
14:59 MotherFlojo joined
15:00 myronmarston joined
15:09 sixstone joined
15:10 vircung joined
15:11 dch_ joined
15:15 simonoram joined
15:16 codestorm joined
15:17 racycle_ joined
15:23 PaReeOhNos joined
15:25 chriswarren joined
15:30 cemilowski joined
15:32 myronmarston joined
15:34 <iFire> I'm wondering if someone can help me integrate this code with guardian. https://github.com/smpallen99/ex_admin/issues/117#issuecomment-232102281
15:34 <iFire> not sure how to provide the get_authenticated_user(conn) with guardian
15:39 achauvin joined
15:47 lexmag joined
15:53 dimitarvp joined
15:57 rschmukler joined
15:58 fowlduck joined
16:01 pcre joined
16:01 <drewolson> ok this phoenix 1.3 keynote is great
16:04 icecreamcohen joined
16:06 davidw joined
16:07 myronmarston joined
16:11 codestorm joined
16:11 jerel joined
16:12 Gringill joined
16:16 robozahn joined
16:17 simonoram joined
16:20 codestorm joined
16:20 codestorm joined
16:22 sfbw joined
16:24 marcdel joined
16:33 potatosalad joined
16:34 Helmclever left
16:37 meh` joined
16:37 alexarnon joined
16:40 drewolson joined
16:44 tuacker joined
16:45 achauvin joined
16:45 sumi joined
16:48 raycoll joined
16:50 duggiefresh joined
16:52 PaReeOhNos joined
16:54 r00tl3ss joined
16:55 racycle_ joined
16:57 hashpuppy joined
16:57 gvaughn joined
16:59 fowlduck joined
17:00 MotherFlojo joined
17:03 robozahn joined
17:03 fowlduck joined
17:04 amclain joined
17:05 refriedchicken joined
17:05 agnostikfront joined
17:07 jeregrine joined
17:07 chrismccord joined
17:11 rschmukler joined
17:11 ZippoWeb joined
17:14 drincruz joined
17:20 jeznet2 joined
17:20 markholmes joined
17:28 jeffweiss joined
17:29 josevalim_ joined
17:35 alexarnon joined
17:35 nextloop joined
17:36 edwinvdg_ joined
17:36 codestorm joined
17:37 wycats_ joined
17:37 Hates__ joined
17:37 hansihe_ joined
17:37 teamj_ joined
17:38 caw__ joined
17:39 arpunk3 joined
17:39 Antiarc_ joined
17:39 Shados_ joined
17:39 adgtl- joined
17:39 acscherp joined
17:40 balo_ joined
17:40 msantos_ joined
17:40 whodidth1s joined
17:40 liru joined
17:40 ljarvis_ joined
17:41 Liquid_X_ joined
17:41 felideon_ joined
17:41 coatezy_ joined
17:42 SouvikB_ joined
17:42 TheGillies1 joined
17:42 leons_ joined
17:42 andman_ joined
17:42 larshesel_ joined
17:42 pigoz1 joined
17:42 SirFunk_ joined
17:42 dp[m]1 joined
17:42 agit0 joined
17:42 wallyd joined
17:42 halorgium_ joined
17:43 robotmayo_ joined
17:43 carterparks joined
17:43 unga joined
17:43 Xikeon joined
17:43 rom15041 joined
17:43 ikopico joined
17:43 Siyo joined
17:43 Siyo joined
17:45 nii joined
17:47 jbhannah joined
17:48 fredsir joined
17:49 OliverMT joined
17:50 Vukan joined
17:51 Vukan left
17:51 Zor joined
17:55 nextloop joined
17:56 Cohedrin joined
17:57 markh joined
17:58 MotherFlojo joined
17:59 chriswar_ joined
18:02 alexarnon joined
18:04 nvd joined
18:05 wsieroci joined
18:11 acscherp joined
18:16 zidoh joined
18:19 <zidoh> i've installed elixir on ubuntu using the erlang-solutions apt list, but it's stuck at 1.3.1.. any apt list that provides the latest releases?
18:19 SvenMa joined
18:22 <ivan> zidoh: I see elixir 1.4.1 in the e-s dists for xenial
18:24 sschneider joined
18:24 <sschneider> is there a way to debug why a node connect fails?
18:25 <sschneider> i can open a tcp connecting to the listening epmd port okay and my cookies are the same so I'm not sure why my connection isn't working
18:28 <zidoh> ivan: probably because i have an older version then
18:28 <zidoh> i'll try upgrading and see, thanks!
18:29 jerel joined
18:29 ZippoWeb joined
18:30 codestorm joined
18:32 raycoll joined
18:33 montanonic joined
18:34 grzs joined
18:34 _flow__ joined
18:37 _flow__ joined
18:38 elixir09823 joined
18:38 <elixir09823> Where is the default phoenix log file saved to?
18:50 johnhamelink joined
18:51 w-t-h_ joined
18:54 <benwilson512> elixir09823: phoenix doesn't do any specific log handling
18:54 <benwilson512> it just uses elixir's built in Logger
18:54 <benwilson512> if you're using releases these just get put in `log`
18:54 <benwilson512> you can run the release in the foreground if you want it inside something like systemd or upstart to handle the loge
18:54 cemilowski1 joined
18:54 <benwilson512> sschneider: cookies?
18:55 <benwilson512> oh wait you said that
18:55 <benwilson512> you need more than epmd port
18:55 <benwilson512> you need 1 additional port per node from a range
18:56 PaReeOhNos joined
19:00 gvaughn joined
19:00 <elixir09823> benwilson512: thanks. where does elixir's logger store development logs? (if any)
19:01 <benwilson512> it does not log to file by default, just stdout
19:02 wsieroci joined
19:02 PaReeOhNos joined
19:03 <sschneider> benwilson512: do you know what the default range is?
19:05 davidw joined
19:06 refriedchicken joined
19:08 sahrizv joined
19:09 <sahrizv> Hi folks!! Does anyone know of any documentation on consul-ex from UndeadLabs? https://github.com/undeadlabs/consul-ex
19:10 <sahrizv> Even a blog post would do. I am willing to read the code but want to build a mental model of how this lib works.
19:11 <Gringill> sschneider: EPMD uses 4369 by default. You can adjust the port range via inet_dist_listen_min and inet_dist_listen_max which are mentioned here: http://erlang.org/doc/man/kernel_app.html
19:11 <Gringill> Relevant reading: http://learnyousomeerlang.com/distribunomicon
19:12 <sschneider> thanks
19:12 alexarnon joined
19:13 MotherFlojo joined
19:17 <zidoh> ivan: upgrading to xenial worked, thanks :)
19:21 <OliverMT> nox: ping
19:22 craiglyons joined
19:22 <nox> OliverMT: Pong.
19:22 <OliverMT> if I was distributing video from an elixir app (capture remotely, basically 1:n fanout of video), but I wanted to serve multiple codecs because of transcoding etc.. would you emit the frames directly through rustler or have rust just send packets on localhost?
19:23 elixir9 joined
19:23 <OliverMT> I havent really googled for transcoding in rust, but I'd imagine those very starting to crop up
19:24 <nox> I know exactly nothing about video stuff.
19:24 <OliverMT> I wonder if packet in packet out -> erlang layer -> fanout is the best solution, instead of saucing in NIFs
19:24 <OliverMT> no, I mean more of a conceptual layer here
19:24 <OliverMT> if I do packet stuff there are buttloads of solutions to use, gstreamer main candidate
19:24 <elixir9> (phoenix question): when I create two separate HTML resources within the same context, the auto-generated context test is only created for the first resource. Is this expected behavior?
19:24 <OliverMT> but it would be so cool to just spin up a transcoder "inside" the mesh
19:25 codestorm joined
19:25 <elixir9> elixir9: I'm using 1.3.0.rc
19:25 <elixir9> Oops, just replied to myself :/
19:25 <OliverMT> OliverMT: no worries about that
19:26 hotpancakes joined
19:30 <OliverMT> nox: I know the video stuff
19:30 <OliverMT> its the rustler and nif stuff I dont know :D
19:30 bilal80 joined
19:31 <craiglyons> How can I run observer on a test suite? Specifically, I'm wanting to trace function input/output.
19:32 <hansihe> OliverMT: I might be able to help if you have any questions as well, feel free to ask
19:32 <OliverMT> I am just fishing for ideas
19:32 <OliverMT> we already scoped up a solution for this using wowza
19:32 <OliverMT> but it feels weird to use wowza when we can just solve the transcoding then do the fanout in erlang
19:32 <OliverMT> the beam is pretty good at shuffling packets around
19:33 marr joined
19:34 sahrizv joined
19:36 refriedchicken66 joined
19:40 jeffweiss joined
19:42 yial2 joined
19:43 <yial2> upgrade to elixir 1.4.2, getting this error - variable "deps" does not exist and is being expanded to "deps()", please use parentheses to remove the ambiguity or change the variable name mix.exs:13
19:43 <yial2> any ideas about this warning?
19:45 <OliverMT> yes
19:45 <OliverMT> look at line 13 and change it to deps()
19:46 <OliverMT> previously if you wrote deps it would try to call local function deps if there was no variable
19:47 <OliverMT> this is very unsafe, as a variable introduced called deps would take over the function call
19:47 <OliverMT> add () and there is no ambuigity
19:49 <yial2> [OliverMT] Thank You!
19:49 <OliverMT> np :)
19:53 josevalim joined
19:56 hotpancakes joined
19:58 alexarnon joined
19:58 codestorm joined
19:59 kyrix joined
20:01 kommen joined
20:03 alexarnon joined
20:04 <ivan> ericmj: should I update my PR to remove the -> alignment? confused by the approval
20:08 myronmarston joined
20:09 <josevalim> ivan: it would be best, yes :)
20:09 <josevalim> ivan: let me know so i can finally merge it :)
20:09 <ivan> thanks
20:11 alexarnon joined
20:12 hotpancakes joined
20:16 <ivan> josevalim: alright it is unaligned
20:18 <sschneider> has anyone used erlport with python?
20:19 kommen joined
20:19 <josevalim> ivan: merged <3
20:19 alexarnon joined
20:20 vmoravec joined
20:20 icanhazbroccoli joined
20:21 <ivan> hurray
20:22 codestorm joined
20:23 markh joined
20:30 travis-ci joined
20:30 <travis-ci> elixir-lang/elixir#13702 (master - 3ba85ad : Ivan Kozik): The build passed.
20:30 <travis-ci> Change view : https://github.com/elixir-lang/elixir/compare/824b26842052...3ba85ad570df
20:30 <travis-ci> Build details : https://travis-ci.org/elixir-lang/elixir/builds/207492697
20:30 travis-ci left
20:32 jparise joined
20:35 alexarnon joined
20:35 rschmukler joined
20:39 chrismccord joined
20:40 hotpancakes joined
20:41 lacour joined
20:42 chriswarren joined
20:43 montanonic joined
20:44 travis-ci joined
20:44 <travis-ci> elixir-lang/elixir#13703 (master - 8cee6fc : Eksperimental): The build passed.
20:44 <travis-ci> Change view : https://github.com/elixir-lang/elixir/compare/3ba85ad570df...8cee6fc20c3f
20:44 <travis-ci> Build details : https://travis-ci.org/elixir-lang/elixir/builds/207497182
20:44 travis-ci left
20:46 <sschneider> can you define rebar plugins in mix.exs?
20:48 rschmukler joined
20:50 raycoll joined
20:53 davidw joined
20:54 hotpancakes joined
20:55 boojinks joined
20:57 jorendorff joined
20:59 <jorendorff> I have a question about the speed of code compiled with Code.eval_quoted()
21:00 <jorendorff> Basically, I have a compiler (AST -> Elixir fn) and an interpreter... and the compiled fn is slower. How is that possible?
21:01 kommen joined
21:01 <jorendorff> core of compiler: https://github.com/NashFP/rdbms/blob/compiler/jorendorff%2Belixir/lib/tiny_rdbms.ex#L405-L434
21:02 <jorendorff> interpreter: https://github.com/NashFP/rdbms/blob/compiler/jorendorff%2Belixir/lib/tiny_rdbms.ex#L456-L479
21:03 simonoram joined
21:06 MotherFlojo joined
21:08 <fishcakez> jorendorff: eval_quoted will use the shell (iex) interpreter and does not compile
21:09 <fishcakez> It will always be slow
21:09 <jorendorff> fishcakez: Thanks, that makes sense. Is there any convenient way to compile and load code?
21:09 <fishcakez> Rather you'd need to use defmodule ... end
21:09 <fishcakez> Then return the fun from a function call to that module
21:10 <jorendorff> The code itself is dynamically generated. I need something eval-like
21:10 <jorendorff> if that makes sense
21:11 <fishcakez> Only modules are compiled code
21:12 MotherFlojo joined
21:12 cdg joined
21:12 rschmukler joined
21:14 montanonic joined
21:14 cemilowski joined
21:16 codestorm joined
21:17 hotpancakes joined
21:20 Guedes joined
21:20 Guedes joined
21:23 byte512 joined
21:23 duggiefresh joined
21:27 <OliverMT> benwilson512 how do you guys test stuff that works on s3? do you mock out the s3 service?
21:29 <jorendorff> fishcakez: Thanks. I passed a defmodule form to eval_quoted and it's much faster.
21:29 <jorendorff> However it warns each time I do it that I'm redefining the module.
21:29 <jorendorff> warning: redefining module Tmp (current version defined in memory)
21:31 <benwilson512> OliverMT: yeah all our code generally uses a FileStore module
21:31 <benwilson512> and it has pluggable backends that we swap out. dev and test use a file system backend, prod uses s3
21:31 <fishcakez> jorendorff: yeah it will
21:33 <jorendorff> fishcakez: Is there a way to suppress the warning or unload the old module before compiling a new one?
21:34 <fishcakez> jorendorff: if there isn't a function in Code, then see :code.purge /delete
21:35 bttf joined
21:37 w-t-h_ joined
21:38 <jorendorff> i see.
21:46 montanonic joined
21:55 nd__ joined
21:55 ipengineer joined
22:03 <OliverMT> josevalim: genstage to do 1:n fanout for video streams, yay or nay?
22:03 <OliverMT> New viewer = new consumer that holds a tcp ref
22:03 <josevalim> nay for taking decisions like that :P
22:03 <OliverMT> I mean in general
22:03 <OliverMT> Im just toying with ideas
22:04 <josevalim> there is not enough information for me to give you a meaningful answer of any sort :)
22:05 <OliverMT> That its video is irrelevant
22:05 <OliverMT> You have a source of packets
22:05 <fishcakez> Someone served video over rabbitmq so you can do anything 😀
22:05 <OliverMT> And N subscribers who should get a copy each
22:06 <OliverMT> So simple 1:n packet fanout
22:10 codestorm joined
22:13 <benwilson512> would you want genstage for that?
22:13 <OliverMT> I am asking
22:13 <OliverMT> :D
22:13 <benwilson512> I feel like video packets are something you don't want to be demand driven
22:13 <benwilson512> at least
22:13 <benwilson512> each user's demand would not necessarily be the same
22:13 <benwilson512> seems like it'd be something you just want to stream at a constant rate regardless
22:14 alexarnon joined
22:14 drincruz joined
22:15 montanonic joined
22:17 sfbw joined
22:17 kommen joined
22:18 alexarnon joined
22:20 hexus0 joined
22:22 <asonge> OliverMT: i wrote a streaming audio server once upon a time, you don't really want to complicate streaming with a flow/gen_stage model, mostly because you already have a buffer in the OS if you use TCP, and you still get the benefit of shared binaries if they're groups of video frames lumped together...dunno what protocols you're looking at using.
22:25 MotherFlojo joined
22:26 edmz_ joined
22:29 alexarnon joined
22:31 hotpancakes joined
22:34 montanonic joined
22:35 jeffweiss joined
22:37 wtetzner joined
22:42 codestorm joined
22:48 alexarnon joined
22:59 drincruz joined
23:03 tofutoon joined
23:04 codestorm joined
23:05 hexus0 joined
23:08 alexarnon joined
23:10 josevalim joined
23:21 MotherFlojo joined
23:24 rschmukler joined
23:26 hexus0 joined
23:27 sfbw joined
23:29 hexus0 joined
23:34 rschmukler joined
23:43 cemilowski joined
23:50 systemfault joined
23:50 Guedes joined
23:50 Guedes joined
23:52 jhack joined
23:52 jhack joined
23:55 rschmukler joined
23:59 codestorm joined