The web in a box - a next generation web framework for the Perl programming language

IRC log for #mojo, 2016-09-14

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

All times shown according to UTC.

Time Nick Message
00:30 Pierre joined #mojo
00:51 tchaves joined #mojo
00:52 jberger https://twitter.com/joelaberger/status/775859253118693381
01:31 Pierre joined #mojo
01:40 Pierre joined #mojo
01:44 salva joined #mojo
01:55 Pierre joined #mojo
01:57 itaipu joined #mojo
02:07 Pierre joined #mojo
02:14 asarch joined #mojo
02:24 kaare_ joined #mojo
02:24 noganex_ joined #mojo
02:25 Adurah joined #mojo
02:29 zivester joined #mojo
03:00 Pierre joined #mojo
03:18 Pierre joined #mojo
03:42 Pierre joined #mojo
04:02 Pierre joined #mojo
04:29 Pierre joined #mojo
05:15 che-quest joined #mojo
05:22 che-quest jberger: ok, never   http://irclog.perlgeek.de/mojo/2016-09-13#i_13198236
05:33 dod joined #mojo
05:38 dod joined #mojo
05:53 dod joined #mojo
05:55 inokenty-w joined #mojo
06:02 Adura_ joined #mojo
06:04 fontardion joined #mojo
06:04 Pierre joined #mojo
06:18 ashimema joined #mojo
06:31 mbudde joined #mojo
06:31 CW joined #mojo
06:33 Vandal joined #mojo
06:51 marcus I'm a happy plenv user on both osx and linux
07:00 anon joined #mojo
07:00 AndrewIsh joined #mojo
07:02 mishanti1 I've switched to plenv myself. I do miss the flexibility of perlbrew, but I like that plenv is a bit cleaner and faster. Feels less invasive.
07:16 marcus mishanti1: What do you miss ?
07:24 mishanti1 marcus: Mostly the ease of switching between perl-versions and libs for each project. I have tried plenv-contrib and it goes some way towards offering that functionality, but it just feels too cludgey.
07:29 marcus mm
08:17 arthas joined #mojo
08:20 Lee joined #mojo
08:26 trone joined #mojo
08:33 osfabibisi joined #mojo
08:36 cuechan joined #mojo
08:40 Lee joined #mojo
09:02 kwa_ I'm really happy with plenv. Don't really need to switch libs for a project. I usually develop for a specific perl version, and let the CI tools test on other Perl versions.
09:03 kwa_ mishanti1: What's your main reason for needing to quickly switch libs for a project out of interest?
09:08 mishanti1 kwa_: Mainly because I frequently work on issues on multiple branches of projects, in which the dependencies between branches (perl even) might differ quite a bit.
09:19 new2perl joined #mojo
09:21 new2perl hi everyone is morbo/hypnotoad going to handle persistence for me automagically? I want to know if prepare_cached() is going to work well (https://www.reddit.com/r/perl/comments/51wvtp/)
09:22 Lee prepare_cached is only going to help in a loop or something
09:23 new2perl so not every time I call a route where the same query happens, but with different params?
09:24 new2perl so the guy posting this (https://www.reddit.com/r/perl/comments/51wvtp/mojoliciouslite_best_way_to_return_json_made_from/d7g4hxk) is mistaken?
09:24 Lee well if your connection is persistent then it will *probably* help
09:25 Lee "probably help" = profile it perhaps
09:25 new2perl how can I tell if my connection is persistent? is this something morbo/hypnotoad does
09:25 bpmedley joined #mojo
09:25 Lee no, it depends on how and where you are creating the DBI object/connction
09:25 kwa_ new2perl: You probably don't need to worry too much about it at this stage.
09:26 Lee yep, premature optimisation yadda yadda yadda
09:26 kwa_ The point when caching or not caching a query becomes an issue, you'd probably have other things to worry about.
09:26 kwa_ Exactly. There's no harm in not using it.
09:27 new2perl i'm new to all this, and learning as I go. Thanks for the help. My dbh is setup vi aapp->attr(dbh => sub { then my oracle settings followed by a DBI->connect; return $dbh;
09:28 kwa_ new2perl: The query would be cached in the worker. So long as you're using lazy initialisation properly, and not connecting in the startup().
09:28 new2perl for context I'm trying to use twitters typeahead to query a large Oracle table.
09:28 Lee new2perl: if you are curious about profiling then checkout Mojolicious::Plugin::NYTProf
09:29 new2perl can you link to something about lazy initialisation? I don't know what that means. Yes. I will try and figure out the NYTProf profiler
09:32 nic new2perl: "lazy X" means a strategy of don't do X till it's proved you need to
09:32 nic lazy evaluation
09:32 purl lazy evaluation is asking someone else if it has that feature
09:33 irqq joined #mojo
09:33 nic it's loosely related to just-in-time
09:34 sri well, you do have to jump through some hoops to performance optimize DBI
09:35 sri wouldn't be surprised if most of the optimizations done in Mojo::Pg also applied for oracle
09:36 new2perl thanks guys, I'll try and profile this and check the outcome. thanks for all your help everyone
09:39 sri and it's quite a bit, from making sure subprocesses don't kill the parent connection, over transparently cached database and statement handles, to making sure there can be multiple active statement handles for the same query at the same time
09:39 sri and tons of other async optimizations
09:40 sri doesn't look like DBD::Oracle can do async though
09:41 Vandal joined #mojo
11:01 geira joined #mojo
11:05 juikuen joined #mojo
11:07 Jonis -w 4
11:07 * coolo doesn't understand the purpose of the weaken call in http://mojolicious.org/perldoc/Mojolicious/Guides/Cookbook#Streaming-multipart-uploads
11:07 * coolo read the documentation of the function, but it's all pretty Bahnhof to me ;(
11:08 coolo i.e. do I always need to call it on $tx or only if I subscribe to events on it?
11:12 geira_ joined #mojo
11:13 batman coolo: you need to create an issue on github. i've totally forgot about assetpack
11:14 coolo batman: you mean totally? have you seen men dressed in black lately? :)
11:15 batman coolo: i don't follow :(
11:15 sri coolo: because the closure using $tx is stored in an object that is stored somewhere inside $tx
11:15 coolo batman: nevermind - just look here: http://www.conny-geiger.de/wordpress/wp-content/uploads/2014/02/Blitzdings.gif
11:16 sri you'd have a circular reference without the weaken call
11:16 coolo sri: ... in that specific example?
11:16 sri yes
11:17 sri perl can't garbage collect $tx without the weaken call
11:17 * coolo is asking because of https://github.com/os-autoinst/openQA/pull/871/files#r78727527
11:18 coolo batman: I just checked - there is no such warning about assetpack in our journal for the last days. So don't waste your time on it - might have been a sun storm
11:18 sri it will work without the weaken call, but you'll be leaking memory
11:19 batman coolo: i'm pretty sure i can reproduce it. i would greatly appreciate if you created the issue
11:19 coolo deal!
11:20 batman thanks :)
11:25 coolo it's gone ;(
11:26 batman what's gone?
11:26 purl gone is probably a strong wrong
11:26 coolo I can't find it - I even tried https://irclog.perlgeek.de/mojo/search to no avail
11:26 coolo batman: the error, I can't find the log that hit it
11:26 batman ok :/
11:28 coolo sorry, but I can't do what you asked
11:32 batman <coolo> batman: hi, did you see "Can't locate object method "start_range" via package "Mojolicious::Plugin::AssetPack::Asset" at /usr/lib/perl5/vendor_perl/5.18.2/Mojolicious/Static.pm" before?
11:32 batman there you are ^
11:32 batman and "fyi, I have mojo 7.05 and assetpack 1.17 in production"
11:33 * sri often wishes the irclog search feature was better
11:33 sri one of the weaknesses of irccloud is that you can't search your infinite backlog
11:34 sri left #mojo
11:34 sri joined #mojo
11:35 sri hahaha, and i kicked myself by clocking on the ircclog [archive] button
11:35 sri *clicking
11:35 batman sri: grep -r coolo ~/.local/share/convos/jhthorsen@cpan.org/irc-magnet/2016/09/ #:)
11:35 sri which actually makes you leave a channel and archive your backlog -.-
11:35 batman i need to add a "share log" button to convos... no reason why others can't read those
11:35 itaipu joined #mojo
11:35 sri heh, yea i guess that's a big advantage for convos
11:36 batman sri: have you tried installing it?
11:36 sri umm, irccloud button of course
11:36 sri my typo game today is on point again
11:36 sri batman: nope, i have irccloud for a year now ;p
11:36 * jberger yawns
11:37 sri but i guess i should try it
11:37 coolo batman: https://github.com/jhthorsen/mojolicious-plugin-assetpack/issues/104 then :)
11:38 sri jberger: didn't you promise to file websocket finish event bugs?
11:39 batman coolo++
11:48 che-quest joined #mojo
11:55 che-quest hi! how to repeate UA transaction inside on finish: $ua->on(start=>sub {my ($ua, $tx) = @_; $tx->once(finish=>sub {my $tx = shift; <here ????>});})
11:56 che-quest i need change proxy and repeate but got Premature connection close
11:56 sri transactions are not reusable
11:57 che-quest oh, thank your
11:57 sri btw. the soundtracks of the deus ex games are great background music for programming
11:58 sri still need to find some good headphones
12:00 che-quest sorry again, can I copy old tx req to new tx?
12:00 sri no
12:01 che-quest ok
12:14 anton joined #mojo
12:15 itaipu joined #mojo
12:19 gizmomathboy joined #mojo
12:24 jberger sri: i would love to file them, if only i could reproduce
12:26 jberger As with lots of my recent stuff, I'm dealing with deeply complex behavior and I'm not sure exactly what is causing the problems.
12:29 Pierre joined #mojo
12:31 sri well, i would have had some time to take a look today ;p
12:33 jberger I am happy to open a tracking bug report
12:34 jberger I don't like to file a bug until I'm sure it isn't my problem
12:34 sri not much of a point
12:34 sri if you can't replicate it
12:35 jberger It Heisenbugs us
12:35 jberger Mostly around when we reset our QA environment
12:36 jberger Which includes a smash close of haproxy
12:38 sri and you're sure it's not just that the connections need to time out?
12:38 sri if the peer just vanishes, tcp has no way of knowing if the connection is alive or dead
12:38 jberger Wouldn't they still be in the ioloop stream lookup though?
12:39 cuechan joined #mojo
12:40 jberger the error I get is that it can't call ->write on an undefined value here: https://github.com/kraih/mojo/blob/master/lib/Mojo/Server/Daemon.pm#L232
12:44 Adura joined #mojo
12:52 sri not enough information
12:52 jberger which is why I haven't opened a bug :-P
12:53 sri sure, more time for overwatch today ;p
12:53 jberger enjoy
12:55 jberger what I do know is that I have a finish handler set (on the multiplexer, but that is fired by the websocket) which SHOULD be cleaning up the notification relay from postgres to finished websocket clients
12:56 jberger so my suspicion is that that event isn't firing
12:56 jberger but I can't prove that yet
13:00 ramortegui joined #mojo
13:24 bwf joined #mojo
13:25 zivester joined #mojo
13:25 fontardion joined #mojo
13:27 sri argh
13:28 sri the way he uses the term utf8 here to mean many things really bugs me https://github.com/kraih/mojo/pull/997
13:29 sri wish he would try to understand the problem first :(
13:33 sri at this point i feel rather comfortable considering STDERR/STDOUT in binary mode the best practice
14:10 tchaves joined #mojo
14:10 bm_ just wanted to say I really like the mojolicious projekt. Trying bring it to the company in which iam working. We're still using mason 1 and some leaders wanted to change to java :( :( :(
14:12 vicash bm_: changing to java means cheaper programmers from india
14:12 bm_ but iam cheap too :D
14:14 bm_ think the main reason is, that there are many people coming directly from university and they know java. only java.
14:15 Adura joined #mojo
14:18 vicash bm_: universities still teaching java ? i thought they switched to python which is why every thing is written in python now
14:19 bm_ in northern germany they do.
14:23 sri the one closest to me teaches java, c/c++ and ruby
14:30 rshadow joined #mojo
14:30 bd is Eiffel still a thing?
14:30 rshadow hi all
14:31 rshadow I have a question on latest commit
14:31 rshadow https://github.com/kraih/mojo/commit/8fdf806ff56520bd79dfe0fe8fc24315918bfd37
14:31 bm_ oo concepts java only. os programing or algorithms in c. (university of applied sciences kiel) are they now teaching ruby for webapps?
14:32 rshadow Why comments for tests necessary to display in binary format instead of utf8?
14:33 itaipu joined #mojo
14:33 sri rshadow: that question makes no sense
14:35 bm_ maybe data is still utf8 encoded and doesnt need a second encode
14:35 bm_ => :bytes
14:36 sri be aware that utf8 and UTF-8 are not the same thing
14:36 sri mixing up terms makes talking about unicode in perl much harder
14:38 rshadow I want to know what you think: output in the console should always be in UTF-8 or not?
14:39 sri i think keeping STDOUT/STDERR in binary mode is a best practice
14:39 howitdo joined #mojo
14:42 rshadow Excuse me. Of course, I wanted to ask about the output of the tests.
14:43 rshadow Output from tests in the console should always be in UTF-8 or not?
14:47 sri as far as i know TAP doesn't specify an encoding
14:48 sri although, the latest spec seems to support inline YAML, and the YAML spec does specify UTF-8/16/32, so it might be safe to assume one of those is required for TAP as well
14:54 sri honestly, i don't think even the folks working on the Test::Simple dist can give an authoritative answer, considering they do silly stuff like changing global handles to UTF-8 (with a plugin in Test2 was it?)
14:54 sri this seems like mostly uncharted territory full of cargo culted hacks, where real best practices still need to be established in the community
14:56 sri i mean, most Test modules are originally from the ascii era
14:58 sri all things considered, choosing UTF-8 as the universal encoding and not messing with Perl globals appears like the only sensible choice to me
14:59 jberger ^^ agreed
15:00 Grinnz exodist would be the guy to talk to about Test2, but doesn't seem to be in here anymore?
15:01 zivester joined #mojo
15:02 Lee joined #mojo
15:02 zivester joined #mojo
15:03 rshadow In my project I turn on unicode everywhere:
15:03 rshadow use utf8; # for perl
15:03 rshadow use open qw(:std :utf8); # for STDOUT ...
15:03 rshadow pg_enable_utf8      => 1, # for postgres DBI
15:04 rshadow All runs in UTF-8, and also output to the console.
15:05 rshadow И я не знаю что теперь делать.
15:05 rshadow And I do not know what to do now.
15:06 rshadow For CI course there is no difference.
15:06 sri well, when you're changing Perl globals you have to expect that some 3rd party code won't work anymore
15:08 sri you can't expect all code to introspect the PerlIO layers on STDOUT/STDERR to figure out what kind of data you might like
15:08 Grinnz "use utf8;" is fine because it only affects the current scope of the source code. the postgres setting is fine because it only affects that connection you created
15:08 Grinnz for STDOUT/STDERR on the other hand those are global, there is no way to only set a utf8 layer for your code
15:08 bm_ why 3rd party code will have problems? I know the issues with 'use encoding', but i find rshadows solutions seems fine for me.
15:09 sri because you're changing Perl globals shared by all code
15:09 sri best practice is to keep all your changes localized
15:12 rshadow perl -e 'use utf8; print "♥";'
15:12 bm_ i thought 'use utf8' tells perl parser that the current source is utf-8 encoded
15:12 rshadow get: Wide character in print at -e line 1.
15:12 Grinnz yes
15:13 Grinnz rshadow: you need to encode it to UTF-8 for output.
15:14 Grinnz sri: you can use Test2::Bundle::Extended with the -no_utf8 => 1 option to avoid loading the plugin that sets layers, btw
15:15 Grinnz it also sets the encoding for formatters, whatever that means...
15:15 sri yaml i imagine
15:15 rshadow So I have to do encode throughout the project? And I can not use `use open qw(td :utf8);` not to break the other modules?
15:16 rshadow It `s so sad.
15:17 ribasushi PERL_UNICODE=SAD kind of sad? :)
15:18 jberger rshadow: mostly mojo uses syswrite to avoid perlio layers
15:18 jberger but for Test::Builder it can't do that
15:18 genio perl -C -Mutf8 -E 'say "♥";'    # you can print wide characters to STDOUT using the -C flag.  perldoc perlrun
15:18 genio just a side note to the earlier thing
15:20 sri Grinnz: wait, setting STDOUT to utf8 was the default in Test2? :O
15:20 Grinnz for that bundle
15:20 sri that is terrifying!
15:20 sri so you can't actually mix Test2 with Test::Mojo
15:21 sri or any other old school Test module
15:21 sri because those would double encode
15:22 sri so we have to recommend not to use Test2, kinda sad
15:25 sri is Test2 even meant to be used directly for testing, or just to implement higher level modules like Test::More?
15:25 sri have to admit i've not really looked into it
15:25 itaipu joined #mojo
15:26 Grinnz the bundles are the higher level modules
15:26 Grinnz Test2 itself is the builder
15:27 sri ok, so we recommend against using Test::Bundle::*
15:28 Grinnz https://metacpan.org/pod/Test2 is the backend for both Test::More and https://metacpan.org/pod/Test2::Suite
15:28 sri because they might pull in Test2::Plugin::UTF8
15:28 sri what a mess
15:28 purl PLEASE DO NOT MAKE A MESS IN HERE
15:28 sri :)
15:29 Grinnz Test2::Bundle::More notably does not include that plugin, and is mostly drop-in for Test::More
15:30 sri i mean, you can have a bunch of tests using Test2::Bundle::More, and then someone pulls in a Test modules that uses Test2::Bundle::Extended and boom
15:30 Grinnz nothing would use Test2::Bundle::Extended except the test file
15:30 sri umm, it changes STDOUT
15:31 bm_ is my understanding now correct? before decode incoming data and encode right before output, better use utf8::upgraded data only and output :bytes?
15:31 sri or so you mean nobody would make Test modules based on it?
15:31 sri i mean, look at what we've done with Test::Mojo
15:31 sri someone is gonna do it
15:32 Grinnz something like Test::Mojo could probably be better done in Test2 by using tools and plugins directly
15:32 sri people are lazy though ;p
15:33 jberger if it makes the test tools consistently take characters and then specify an output encoding then that is fine isn't it?
15:33 jberger it only happens once you convert the test library to Test2
15:36 sri i guess anyone who agrees with Test2::Plugin::UTF8 also should disagree with what Test::Mojo does
15:37 Grinnz i think there should be some discussion with Exodist at least about the purpose of the UTF8 plugin, and whether it would be better encoding its output to UTF8 as needed
15:38 * sri nods
15:38 Grinnz i'll open a bug report but right now i gotta go :P
15:40 diabolical joined #mojo
15:41 diabolical Hi all, I had the idea of making my app delete all its source files during startup, to prevent hackers from having any chance of stealing my source code. I will lose the ability to do restarts, but I can just re-create the docker container instead. And I will need to make sure not to load anything dynamically at runtime. Is this a crazy idea? Anyth
15:41 diabolical ing else to consider?
15:44 sri lots of code is lazy loaded
15:44 sri templates will be tough to preload
15:45 diabolical It's just an API backend. No templates.
15:54 disputin joined #mojo
15:55 batman diabolical: the controllers are lazy loaded as well
15:59 dod joined #mojo
16:00 diabolical that's a problem
16:00 diabolical I'm going to experiment anyway
16:12 abra joined #mojo
16:15 nicomen https://github.com/kraih/mojo/wiki/Plugins-on-cpan - "Some plugins … have the namespace MojoX. This namespace was deprecated for a while …" - but not anymore?
16:15 jberger We have soft-un-deprecated that namespace
16:16 jberger it used to be sorta formally used, then we said "no just use Mojo:: but dont't step on namespaces we might want"
16:16 jberger but now, if you want to use MojoX:: I don't think anyone minds
16:17 nicomen I would like to publicize a subclass of Mojo::UserAgent that has caching and support for reading local files, what should I name it?
16:17 jberger well, propose a name and we can tell you if its ok
16:17 nicomen I see several MojoX::UserAgent (including a bare one which I don't see the point of)
16:17 jberger Mojo::UserAgent::DoesMyCoolThing is probably ok
16:18 nicomen MojoX::UserAgent::Cached::And::LocalFileSupport :-P
16:18 jberger as long as "DoesMyCoolThing" isn't a generic name we might need later
16:18 nicomen do you rally want me to use Mojo::UserA... ?
16:18 nicomen (no X)?
16:19 Atog beware of using Mojo::Docker ;)
16:19 nicomen lol
16:19 nicomen Docker::Mojo I think is worse
16:19 nicomen I'm starting to consider using a roles approach like you did jberger, but I am not so sure how to refactor it like that
16:19 nicomen would need to do "around" stuff for build_tx
16:23 saki joined #mojo
16:23 jberger a roles approach to adding functionality to Mojo::UserAgent would be nice to see
16:23 jberger it would take more design than Test::Mojo did
16:23 jberger as you say
16:24 nicomen could add around support to Mojo::Base *cough* ;D
16:25 jberger you are going to use Role::Tiny at least
16:26 jberger so if you use Role::Tiny and Class::Method::Modifiers (or whatever it is, check the Moo dependencies) then you get that
16:26 jberger or just use Moo if you need modifiers
16:27 nicomen I think one approach is to get it out, and it can be improved later
16:27 nicomen since it WFM for two years or so
16:27 nicomen but it is not elegant
16:27 jberger there is also https://metacpan.org/pod/Mojo::RoleTiny
16:28 nicomen I was going to ask about that
16:28 nicomen saw you even as a contributor
16:28 jberger it has been a while since I've looked at it tbh
16:29 nicomen _and_ Mojo::Role
16:29 jberger but I remember thinking is was a good idea
16:29 jberger Mojo::Role was the original name but it steps on the aforementioned "don't use generic names we might want"
16:29 jberger it was supposed to be deleted and have the namespace transferred to sri, doesn't look like that ever happend
16:30 nicomen hence why I'm thinking I should rather use MojoX...
16:32 jberger you can do anything you want with MojoX
16:33 jberger if you would like to use Mojo:: just ask us first
16:33 jberger general rule
16:33 purl general rule is probably `there is no general rule to determine *what* a function will do in scalar context if all you know is what it will do in list context'
16:33 jberger almost want to tell purl to forget that one since it will usually be unrelated, and yet ...
16:34 nicomen hehe ok
16:37 nicomen this looked intriguing: http://search.cpan.org/~tomorrow/Momo-1.2/
16:59 PryMar56 joined #mojo
17:03 mcsnolte joined #mojo
17:03 Lee joined #mojo
17:19 abra_ joined #mojo
17:40 asan131 joined #mojo
17:43 blonewolfs joined #mojo
17:58 jamesaxl joined #mojo
18:02 itaipu joined #mojo
18:03 Lee joined #mojo
18:04 itaipu joined #mojo
18:09 Akron joined #mojo
18:12 irqq joined #mojo
18:18 batman joined #mojo
18:19 batman pretty awful that the fuse goes every time i vacuum :P
18:19 cuechan_ joined #mojo
18:20 Grinnz you might want to take note of which circuit you're using :P
18:20 Grinnz i like to use the kitchen outlets for vacuuming, they're more dedicated
18:21 perlpilot Or just swap out the 10 amp fuse for a 20 amp fuse
18:22 batman Grinnz: the apartment is too big. i need to buy an extension cord to use a fuse that doesn't take out my server
18:22 Grinnz ooh
18:22 batman perlpilot: that's illegal if i do it myself :/
18:23 batman i think just a "slower" fuse would do the trick
18:23 Grinnz yeah, would have to do that carefully
18:23 Grinnz slower fuse would risk a power surge to your server :/
18:23 Grinnz there's a reason these things work as they do :P
18:23 bd there is this futurama episode where fry and amy go on a ride in a convertible
18:24 batman haha. true :)
18:24 bd so what i am saying is, turn off the ac, tv, stereo, xbox and what not. then vacuum ;)
18:24 preaction jes, de Beta Romeo!
18:24 itaipu joined #mojo
18:25 batman bd: there's not much too turn off :/
18:25 batman thanks for the tip
18:26 sh14 joined #mojo
18:29 Grinnz sri: https://github.com/Test-More/Test2-Suite/issues/92
18:34 sri Grinnz++
18:35 sri hmm, lots of little changes on github today https://github.com/universe-2016
18:36 Grinnz neat
18:37 Grinnz oh wow, the projects tab is basically pivotal tracker
18:37 jberger that's awesome
18:41 Grinnz hmm, wonder what the integrations means for travis CI
18:42 Grinnz i wonder if the "github community forum" will include a feature request board :P
19:20 Lee joined #mojo
20:03 cuechan_ joined #mojo
20:17 dantti_laptop joined #mojo
21:14 cpan_mojo Mojo-IRC-0.38 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/Mojo-IRC-0.38
21:14 zeeq joined #mojo
21:15 disputin joined #mojo
21:22 batman joined #mojo
21:42 lluad joined #mojo
21:47 tchaves joined #mojo
22:41 * sri wonders why all laptop makers suddenly start using killer wireless chips
22:42 sri i remember seeing people complain about those in the cool razer laptops, and now dell adds them to the new xps too
23:01 Grinnz my new motherboard has killer ethernet, too
23:02 Grinnz it's a new atheros thing I guess
23:02 Grinnz they used to always have marvell or something
23:04 sri i feel like i know those killer chips from like 10 years ago, weren't those the ones bypassing the operating system tcp/ip stack?
23:05 Grinnz all i remember from back then is the ethkiller, which is a bit different :P
23:06 sri they have the sketchiest website for a company that's suddenly on every mainboard http://www.killernetworking.com
23:07 jberger yeah, I mean, google+?! who does that!
23:08 sri right
23:12 Grinnz heh
23:13 Grinnz i mean, it looks pretty similar to the websites for most laptop and motherboard manufacturers
23:13 Grinnz they don't invest a whole lot into web design I think :P
23:25 disputin joined #mojo
23:25 lluad They're the alienware or monster of nics, aren't they?
23:25 Lee joined #mojo
23:26 Grinnz doubt it, if they overcharged that much they wouldn't be used by all the manufacturers
23:27 lluad Well, they're just a Qualcomm brand, aren't they? Aimed at the sort of people who buy monster cables to plug into their alienware boxes.
23:28 lluad Generic silicon with a EXXXTREME! windows GUI.
23:30 Grinnz sri: looks like this is what you meant https://en.wikipedia.org/wiki/Killer_NIC
23:30 Grinnz and i do vaguely remember this now
23:31 Grinnz and yeah, Killer is a qualcomm/atheros brand
23:32 lluad http://techreport.com/review/29144/revisiting-the-killer-nic-eight-years-on
23:33 lluad "The Killer suite of software is only available for Windows. For the Linux users out there, the Killer NICs work with the existing alx Ethernet driver." Yeah, just preloaded windows bloatware.

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