Cryto! 22 March 2014

00:40:32 leaking (leaking@cryto-1309AB55.dfri.se) has joined #crytocc
01:06:51 <dpk> joepie91: ping
01:23:24 leaking has quit (User quit:  Page closed)
01:28:05 muamary2891 has quit (Ping timeout)
01:28:25 sheltond8222 (sheltond82@BE02B62F.1F65062D.6C8646.IP) has joined #crytocc
01:47:02 An0nEnd3r has quit (Connection reset by peer)
02:03:31 An0nEnd3r (An0nEnd3r@B184722E.61681D90.6347A1D1.IP) has joined #crytocc
03:05:03 rmrfbuck (rmrfbuck@cryto-7266665C.us-west-1.compute.amazonaws.com) has joined #crytocc
03:06:26 rmrfbuck has quit (User quit:  Connection closed)
05:07:49 Thor has quit (User quit:  Quitte)
05:18:49 <pzuraq> yo
05:47:49 An0nEnd3r has quit (Connection reset by peer)
05:48:47 An0nEnd3r (An0nEnd3r@AC98F370.61681D90.6347A1D1.IP) has joined #crytocc
09:26:46 IDC916 (IDC916@5BA091AF.881AB5A7.8FB3DC63.IP) has joined #crytocc
09:30:07 drugged (drugged@cryto-FAFB75BA.torproject.afo-tm.org) has joined #crytocc
09:33:00 drugged has quit (User quit:  Page closed)
09:37:41 IDC916 has quit (User quit:  Page closed)
09:37:50 IDC491 (IDC491@cryto-3E1E0419.anonymizer.ccc.de) has joined #crytocc
09:48:36 An0nEnd3r has quit (Connection reset by peer)
09:55:05 An0nEnd3r (An0nEnd3r@598035F6.F8BE82A8.1CFE3EC9.IP) has joined #crytocc
09:55:26 IDC491 has quit (Ping timeout)
10:31:52 pzuraq has quit (Input/output error)
11:00:48 guest1974 (guest1974@672F3E79.C6EA6E41.86E5D318.IP) has joined #crytocc
11:00:56 <guest1974> hello
11:01:31 <guest1974> I am still wondering if someone might have some free time to help
11:23:04 iceTwy (iceTwy@iceTwy.users.cryto) has joined #crytocc
11:34:38 Cryto488 (Cryto488@BB9135B8.DCC86F23.AB59D1E1.IP) has joined #crytocc
11:46:38 Cryto488 has quit (Ping timeout)
11:53:48 HiveResearch has quit (Ping timeout)
12:05:03 RATherder (RATherder@cryto-594DD758.us-west-1.compute.amazonaws.com) has joined #crytocc
12:06:24 RATherder has quit (User quit:  Connection closed)
12:09:00 guest1974 has quit (Ping timeout)
12:21:03 iceTwy has quit (User quit:  WeeChat 0.4.3)
13:03:02 anonnews673 (anonnews67@cryto-F8D38F71.rev.sfr.net) has joined #crytocc
13:03:07 <anonnews673> salut
13:07:13 anonnews673 has quit (Ping timeout)
13:36:08 user (user@cryto-91A00DD0.torservers.net) has joined #crytocc
13:36:25 <user> hy @ all
13:37:01 <user> did someone knows a way to use apt-get also through tor network?
13:41:11 <user> noone active?
13:49:21 user has quit (User quit:  Page closed)
13:49:39 An0nEnd3r has quit (Connection reset by peer)
13:50:43 An0nEnd3r (An0nEnd3r@598035F6.F8BE82A8.1CFE3EC9.IP) has joined #crytocc
14:48:03 <iceTwy_znc> er
14:48:06 <iceTwy_znc> http://www.lrb.co.uk/blog/2014/03/14/nick-richardson/translating-lorem-ipsum/
14:48:08 <iceTwy_znc> .title
14:58:34 <DrWhat> STOP PMING ME OR I WILL SERIOUSLY FUCK YOUR LIFE!!!!
14:58:45 <DrWhat> joepie91 ban IR601
14:59:33 <DrWhat> Anyone been on STS before?
15:09:14 iceTwy (iceTwy@iceTwy.users.cryto) has joined #crytocc
16:14:45 iceTwy has quit (Ping timeout)
17:31:22 pzuraq (pzuraq@cryto-BAE76FBA.hsd1.ca.comcast.net) has joined #crytocc
17:41:01 HiveResearch (HiveResear@7E9D6C07.9F32FB1E.57B9DF22.IP) has joined #crytocc
17:50:26 An0nEnd3r has quit (Connection reset by peer)
17:52:51 <joepie91> sigh
17:52:54 <joepie91> dpk: pong
17:52:57 <joepie91> ish
17:53:01 <joepie91> DrWhat: why?
17:55:29 botpie91 (botpie91@botpie91.users.cryto) has joined #crytocc
17:56:39 An0nEnd3r (An0nEnd3r@63EA280E.F8BE82A8.1CFE3EC9.IP) has joined #crytocc
18:12:41 <pzuraq> wassup joepie91
18:14:10 <joepie91> pzuraq: OHAI!
18:14:13 <joepie91> well
18:14:17 <joepie91> I have new internets :D
18:14:18 <joepie91> dronten (serverius): http://www.speedtest.net/result/3386880625.png
18:14:19 <joepie91> haarlem (leaseweb): http://www.speedtest.net/result/3386882626.png
18:14:19 <joepie91> rotterdam (i3d): http://www.speedtest.net/result/3386884334.png
18:14:19 <joepie91> amsterdam (softlayer): http://www.speedtest.net/result/3386886382.png
18:14:19 <joepie91> amsterdam (nforce): http://www.speedtest.net/result/3386888377.png
18:14:24 <joepie91> speedtest during peak hours (21:00)
18:14:30 <joepie91> aside from that, it's kinda-weekend
18:14:33 <pzuraq> nice
18:22:40 iceTwy (iceTwy@iceTwy.users.cryto) has joined #crytocc
18:30:10 <pzuraq> I'm trying to figure out how to keep local and server data in sync
18:30:14 <pzuraq> is hard :/
18:49:30 <joepie91> pzuraq: what stack?
18:50:00 <pzuraq> joepie91: Ember js on the client, Rails on the server
18:50:24 <pzuraq> I think I'm going to leave the onus on the user, give them a refresh button
18:50:34 <pzuraq> collisions are unlikely at this company
18:51:09 <joepie91> lol
18:51:12 <joepie91> but yeah, sync is hard
18:51:20 <joepie91> it's one of those things that seems dead easy
18:51:26 <joepie91> until you try to actually implement it
18:51:26 <joepie91> lol
18:53:03 <pzuraq> totally
18:53:23 <pzuraq> I really would like to use something like firebase
18:53:44 <pzuraq> but I dislike having an entire segment of my app stack being closed
18:53:45 <pzuraq> https://www.firebase.com/
18:54:02 <pzuraq> and other realtime frameworks simply aren't mature
18:54:07 <pzuraq> meteor, derby
18:54:16 <pzuraq> dunno if they ever will be
18:54:33 <pzuraq> but yeah, realtime Single Source of Truth is the way to go
18:54:41 <pzuraq> if you have the resources ;)
19:02:49 <joepie91> pzuraq: what kind of application are you trying to write?
19:08:44 <Zekka> joepie91 -- Have I bragged about my current projects in your direction yet?
19:08:53 <joepie91> Zekka: not yet :P
19:09:10 <Zekka> I feel obligated, especially since I've been using Github pretty religiously lately (as opposed to letting code rot in my programming/ folder)
19:09:35 <Zekka> I'm on a few, but this is what's been taking up most of my time for about the last five days: https://github.com/Zekka/sho
19:09:45 <Zekka> (I'm on break today)
19:10:25 <Zekka> It's got first-class functions, a pretty decent interface to inject Haskell code, and, obviously, a working parser/compiler
19:10:46 <joepie91> fancy :P
19:10:50 <Zekka> Oh yeah, it also has exceptions which work pretty much how you'd expect
19:11:06 <Zekka> The code quality's a little poor and needs an overhaul
19:11:38 <Zekka> Since you probably don't have Haskell installed, here's the output of the current test file: http://pastebin.com/g3b2tibS
19:11:52 <Zekka> You'll notice the stack trace isn't perfect -- it includes a lot of block definitions for some reason
19:12:09 <Zekka> that's because there's a compile-time transformation that causes starting a block to occasionally create additional stack frames
19:12:18 <Zekka> I'll be fixing that later, although maybe not today
19:14:37 <Zekka> I also implemented IRC and a simple bot in Haskell because DZ was rewriting his C# bot: https://github.com/Zekka/gekko
19:15:00 <Zekka> and other than that, I put the source code for my current game project online: https://github.com/Zekka/mobilize
19:16:25 <joepie91> Zekka: I should add that I don't really speak haskell at all
19:17:06 <Zekka> I didn't know if you did but I'm not too surprised by that
19:17:46 <Zekka> I might bug Schyler later about refactoring -- he worked on GHC, he'd probably be able to help
19:17:52 <Zekka> although not without doing some refactoring myself first
19:18:18 <Zekka> well, he didn't just work on GHC -- I think he worked on DDC and possibly some other projets
19:18:24 <joepie91> mobilize seems interesting
19:18:27 <joepie91> GHC? DDC?
19:19:00 <Zekka> http://www.haskell.org/haskellwiki/DDC <- Disciplined Disciple Compiler, and https://www.haskell.org/ghc/ <- Glasgow Haskell Compiler
19:19:34 <joepie91> ah :P
19:19:48 <Zekka> From what I hear the GHC codebase is a mess
19:21:46 <Zekka> If, by the way, any name comes to mind for a small JSlike scripting language that's better than "Sho", let me know
19:22:36 <Zekka> I should probably stop calling it 'small' because somehow it's about 1100 lines
19:26:20 <joepie91> Zekka: given that one of my servers is named "Shi"...
19:26:21 <joepie91> :p
19:26:22 OpRussia (OpRussia@CD00346F.75307B2D.20AABD02.IP) has joined #crytocc
19:26:28 <joepie91> .welcome OpRussia
19:26:29 <botpie91> OpRussia: welcome to #crytocc! Please be aware that this channel is publicly logged, and make sure to read the rules in the channel topic. You may hide messages from the public logs by prefixing them with [off].
19:26:54 <OpRussia> haha i am not new to irc i know this
19:27:01 <OpRussia> I am he person who runs OpRussia
19:27:05 <OpRussia> On AnonOps irc
19:27:21 <OpRussia> and all it socal media accounts
19:27:33 <Zekka> joepie - It's just a contraction of the name I originally used for an earlier design
19:28:41 <joepie91> aha
19:30:16 <OpRussia> I need to speak with macbeth
19:30:19 <OpRussia> hey u seen them
19:30:43 <Zekka> Doesn't macbeth usually hang out here?
19:30:52 <Zekka> I don't think I've actually seen him lately
19:31:14 <OpRussia> Mac is caring out attack on russia site that are being leaked and he ddosing them
19:31:29 <OpRussia> I am oprussia leader and cam to speak with his  skid
19:31:48 <joepie91> ...................................
19:31:54 <joepie91> OpRussia
19:32:00 <OpRussia> OpRussia
19:32:02 <joepie91> I literally just told you to read the rules
19:32:02 <OpRussia> yes
19:32:14 <joepie91> the rules explicitly say "no Anonymous"
19:32:16 <joepie91> as well as "no cracking"
19:32:18 <joepie91> and "no drama"
19:32:19 <Zekka> I'd considered mentioning the 'no anonymous' thing but it seemed redundant
19:32:23 <joepie91> you're breaking two-and-a-half of them now
19:32:27 <OpRussia> I dont read the rules i came to speak with mac
19:32:33 <OpRussia> I am not cracking or doing anything
19:32:38 <OpRussia> Just came to speak with him
19:32:40 <OpRussia> or her
19:33:10 <joepie91> seriously
19:33:14 <joepie91> people..
19:33:19 OpRussia (OpRussia@CD00346F.75307B2D.20AABD02.IP) has joined #crytocc
19:33:45 <joepie91> that was your first and last warning
19:34:59 <OpRussia> I am will ur shitty irc haha wow  tell macbeth oprussia looking for them
19:35:06 OpRussia has parted #crytocc ()
19:35:19 <joepie91> saves me a ban slot
19:35:36 <Zekka> I can't tell if he's threatening Macbeth or if he's working with him
19:35:44 <joepie91> Zekka: threatening as far as I can tell
19:36:00 <joepie91> but honestly I don't particularly care
19:36:11 <joepie91> if he'd behaved, I would've pointed him at the 'tell' command
19:36:30 <joepie91> but given his attitude, I really don't want the drama to overflow into this channel
20:26:23 <joepie91> https://laverna.cc/
20:35:47 <DrWhat> this another one of your pojects joepie91?
20:36:11 <pzuraq> joepie91: Why do so many crazies come here?
20:36:27 <pzuraq> :(
20:37:18 <joepie91> DrWhat: no
20:37:23 <joepie91> pzuraq: I have no idea.. :/
20:56:45 <pzuraq> ppl b crzy
21:51:12 An0nEnd3r has quit (Connection reset by peer)
22:02:50 iceTwy_znc has quit (User quit:  ZNC - http://znc.in)
22:05:03 THX5555 (THX5555@cryto-6EFDBCAA.us-west-1.compute.amazonaws.com) has joined #crytocc
22:05:53 An0nEnd3r (An0nEnd3r@F4E02264.AAE2E054.8BD853D0.IP) has joined #crytocc
22:06:08 iceTwy_znc (iceTwy_znc@iceTwyznc-07274.users.cryto) has joined #crytocc
22:06:14 <pzuraq> soooo
22:06:18 <pzuraq> I figured out
22:06:24 THX5555 has quit (User quit:  Connection closed)
22:06:26 <pzuraq> that js promises are monads
22:11:07 <Zekka> pzuraq - Isn't that common knowledge?
22:12:49 <pzuraq> Zekka: Nope, lot of people use JS without any idea what it is doing
22:13:10 <joepie91> pzuraq: isn't that pretty much the default modus operandi of javascript lol
22:13:13 <Zekka> Do promises provide return by the way?
22:13:16 <joepie91> "I have no idea what this does but it works"
22:13:28 <pzuraq> I didn't put two and two together because haskell is where I learned monads and it's very different
22:13:39 <pzuraq> joepie91: pretty much :p
22:13:47 <Zekka> I don't remember the commonjs promise API too well but I remember that the chaining mechanism is basically binding.
22:13:53 <pzuraq> when I first started using js I thought $ was part of the language
22:13:54 <Zekka> er, that is, binding as in >>=ing
22:14:26 <pzuraq> Zekka: I also didn't get monads that well :/
22:14:30 <pzuraq> shit was confusing!
22:14:37 <Zekka> Out of curiosity, how much do you use Haskell? I'm currently working on a Haskell project and while I'm still completing a refactor, I might ask for organizational/design advice later
22:14:44 <pzuraq> Haskell is a weeeeeeiiiiiiirrrrrrd language
22:14:55 <pzuraq> yeah don't ask me for advice
22:15:29 <pzuraq> I would like to learn it and Lisp/Clojure at some point, but I am no where near groking many of the core concepts
22:15:44 <pzuraq> though now that I see the similarities to JS I may get better at that...
22:20:07 iceTwy_znc has quit (User quit:  ZNC - http://znc.in)
22:20:38 <pzuraq> joepie91: waht do you think of hack?
22:21:22 <joepie91> pzuraq: haven't looked into it too much yet
22:21:37 <pzuraq> php with static typing
22:21:39 <pzuraq> could be cool
22:24:48 <joepie91> pzuraq: I'm not exactly a fan of static typing
22:25:54 <pzuraq> joepie91: I swing both ways on it. Great for large projects, less so for small teams that need to move quickly.
22:26:07 iceTwy_znc (iceTwy_znc@iceTwyznc-07274.users.cryto) has joined #crytocc
22:26:14 <joepie91> pzuraq; I don't really see the point of it at all, really
22:26:20 <joepie91> other than maaaaaybe for optimization purposes
22:26:24 <pzuraq> I'd shudder to think about writing an OS in an untyped language from the size of the codebase alone...
22:26:34 <pzuraq> catch errors early
22:26:48 <Zekka> joepie91 - Have you looked into Haskell? It's got a really great static typing system
22:26:51 <pzuraq> ensure all team members are on the "same page"
22:27:01 <joepie91> pzuraq: that doesn't really sufficiently answer my question
22:27:05 <joepie91> or well, my not-quite-a-question
22:27:12 <Zekka> the language can reason about your code for you (look at Control.Lens for a great example of this), and it catches most errors at compiletime
22:27:38 <joepie91> in my experience, in any well-written piece of software, the amount of typing-related errors is miniscule compared to other bugs
22:27:55 <Zekka> You would be surprised how many errors in a Haskell program become typing-related errors
22:27:58 <Zekka> (It's almost all of them)
22:28:19 <Zekka> Does your code ever crash because of an unexpected null? In Haskell that's a typechecker error
22:28:19 <joepie91> Zekka: perhaps that's the case in Haskell, I don't know
22:28:23 <pzuraq> yes, but the larger the project gets the more typing related errors get involved
22:28:33 <joepie91> Zekka; no
22:28:37 <Zekka> Do you forget to handle an exception? The exception is now part of the type.
22:28:38 <pzuraq> particularly at the low levels
22:28:53 <joepie91> pzuraq: again, this does not match my experiences
22:29:02 iceTwy_znc has quit (User quit:  ZNC - http://znc.in)
22:29:09 <joepie91> I find that the amount of typing-related errors is much more strongly related to the (non-)experience of the developer(s)
22:29:11 <pzuraq> again, would shudder to work with pointers without a type system...
22:29:11 <joepie91> than to the size of a codebase
22:29:36 <iceTwy> Could anyone visit my websites (iceb.in, icetwy.re and blog.icetwy.re) and tell me if SSL is working properly?
22:29:36 <joepie91> pzuraq: PHP/Hack do not have pointers
22:29:43 <iceTwy> pretty please
22:29:44 <pzuraq> size of project, not codebase. More people, and more people working asynchronously
22:29:53 <joepie91> pzuraq: does not change my point
22:29:55 <Zekka> joepie91 - Where do your bugs come from?
22:30:07 <pzuraq> joepie91: Not arguing for PHP/Hack, arguing for typing in general.
22:30:17 <joepie91> Zekka: they are usually syntax issues
22:30:22 <joepie91> in a rare case, an unhandled exception
22:30:27 <joepie91> which, while you MIGHT make it a type thing
22:30:30 <joepie91> isn't inherently a type thing
22:30:35 <Zekka> Really? That doesn't describe my experience at all
22:30:55 <Zekka> Any good language will crash on syntax issues -- really, dealing with exceptional behavior is the only case where Haskell's type system shines there
22:31:12 <pzuraq> actually now that I think about it tho majority of my C related bugs could be solved with a stronger typing system
22:31:24 <joepie91> pzuraq: I'm sure that there are certain languages/environments where static typing is necessary to work well with other parts of that environment - but at that point, the environment was MADE for static typing, and that requirement may very well have been introduced as a consequence of static typing being available
22:31:33 <Zekka> I guess if you want a brief summary of why I like Haskell's typechecker so much, I'd say that about two thirds of the time my code works on the first successful compile
22:31:40 <Zekka> I can't say that about any other language
22:31:44 <pzuraq> most of them come from me casting shit willy-nilly and accidentally forgetting what type I'm working with
22:32:14 <joepie91> pzuraq: how do you forget what type you're working with
22:32:15 <pzuraq> once ALL of my pointer math was off because I was multiplying everything by sizeof(int) when I didn't need to
22:32:15 <joepie91> :|
22:32:16 <pzuraq> lol
22:32:31 <pzuraq> joepie91: Cause I'm working directly on memory, like managing
22:32:31 <pzuraq> it
22:32:39 <pzuraq> all it is is a stream of bits
22:33:04 <joepie91> pzuraq: that goes into the "necessity of static typing" bit
22:33:12 <joepie91> but doesn't explain why it'd be a desirable feature in itself
22:33:14 <joepie91> if you get what I mean
22:34:31 <pzuraq> I think it's helpful when you are A. Working on a large codebase with B. a large team of coders who may not even be located near each other and C. are not necessarily quality coders.
22:34:55 <joepie91> the only thing in there that I can possibly see being valid is C
22:35:05 <joepie91> and that is a problem with the quality of the developers, not the tools they work with...
22:35:21 <joepie91> static typing would basically just end up being a roll of duct tape
22:35:27 <joepie91> to prevent things from leaking too badly
22:35:28 <Zekka> To be fair, if you pick a language like Haskell you automatically eliminate horrible coders
22:35:33 <Zekka> if you pick a language like PHP you inite them
22:35:35 <Zekka> invite them*
22:35:49 <pzuraq> joepie91: I would say they are necessary and sufficient, i.e. all three are required and without any one the usefulness of a typing system drops dramatically
22:36:07 iceTwy_znc (iceTwy_znc@iceTwyznc-07274.users.cryto) has joined #crytocc
22:36:11 <joepie91> pzuraq: I still fail to see points A and B altogether
22:36:15 iceTwy_znc has quit (User quit:  ZNC - http://znc.in)
22:36:29 <joepie91> how is a large codebase / team relevant for static typing?
22:36:53 <pzuraq> well, if you don't have A, then even though you may have C but it's easy to check all the code for errors and so you don't have to worry
22:37:07 <joepie91> what?
22:37:15 <pzuraq> if you don't have B, then you can check up on all the coders in your team and even train them
22:37:51 <joepie91> how is having a large team an excuse for not having quality developers?
22:38:02 <joepie91> or rather
22:38:04 <joepie91> how is it a valid excuse
22:38:22 <pzuraq> but if your codebase is too large to keep track of (i.e. a kernel) your devs are all of the planet (kernel dev team) and you have people from all over the planet contributing all the time (kernel) then a type system does a lot of the work for you.
22:38:36 <joepie91> pzuraq: again, duct tape
22:38:44 <pzuraq> ugh...
22:38:51 <pzuraq> ok the biggest reason for a type system
22:39:04 <pzuraq> for code that is meant to last and evolve forever
22:39:22 <pzuraq> you aren't going to have the same team forever
22:39:37 <pzuraq> you cannot guarantee the quality of coders working on it.
22:39:43 <pzuraq> what can you do?
22:39:47 <pzuraq> use a typed language.
22:39:52 <joepie91> <pzuraq>you cannot guarantee the quality of coders working on it.
22:39:58 <joepie91> then you are doing something inherently very very wrong
22:40:04 <pzuraq> ...dying?
22:40:06 <joepie91> and again, a static typing system is basically duct tape then
22:40:13 <pzuraq> Linus Torvalds is going to die
22:40:20 <joepie91> so?
22:40:23 <pzuraq> and we will keep working on Linux, forever
22:40:26 <iceTwy> right, okay
22:40:28 <joepie91> so?
22:40:37 <iceTwy> why does cron hate date
22:40:48 <joepie91> it's cron, cron hates everything
22:40:51 <pzuraq> can't guarantee that quality devs will be doing the work
22:40:54 <iceTwy> derp
22:41:20 <pzuraq> but even if they aren't, Linux will still evolve because of the quantity of users
22:41:21 <iceTwy> OOOOOOOOH
22:41:21 <joepie91> pzuraq: see, the problem with your reasoning is that it's loaded with assumptions and jumps in logic
22:41:25 <joepie91> based on "common knowledge"
22:41:25 <iceTwy> you need to escape it
22:42:04 <joepie91> pzuraq: for example, you claim that "you can't guarantee quality devs will be doing the work when Torvalds dies"
22:42:08 <joepie91> but fail to explain why that would be the case
22:42:17 <pzuraq> Torvalds can't because he is dead
22:42:18 <joepie91> and fail to analyze possible methods to make sure that that DOES happen
22:42:23 <joepie91> sp?
22:42:25 <joepie91> so? *
22:42:29 <joepie91> is Torvalds the only sane person on this globe?
22:42:33 <joepie91> I would damn well hope not
22:43:00 <pzuraq> So if he wants to make a decision, at the very beginning that will guarantee that even if shitty devs attack his code they will not ruin it, what can he do?
22:43:15 <joepie91> pzuraq: you're missing the point
22:43:36 <joepie91> pzuraq: haven't you noticed how Bitcoin developers are STILL very careful with implementations in the core code, and how they are still just as diligent as they were before?
22:43:40 <joepie91> despite Satoshi going poof?
22:44:00 <joepie91> it's a perfect example that shows why your theory isn't correct
22:44:10 <pzuraq> yeah, and I think that quality devs will be working on Linux
22:44:20 <joepie91> Satoshi jumped ship for $whatever_reason, but that doesn't mean that somehow all the diligence left the building
22:44:22 <pzuraq> I'm not talking realistic, I'm talking worst case scenario
22:44:26 <joepie91> because he found a suitable person to take over the reins
22:44:43 <pzuraq> a typing system is like a guard rail, it prevents the worst case.
22:44:49 <joepie91> pzuraq: if the development of Linux is taken over by incompetent developers, do you _really_ think a static typing system would save anything?
22:44:56 <joepie91> you'd be best off running for the hills
22:45:00 <joepie91> because if they don't fuck up there
22:45:02 <joepie91> they will fuck up elsewhere
22:45:04 <iceTwy> alright
22:45:08 <iceTwy> now it fucking works.
22:45:08 <joepie91> there's a reason I'm calling it duct tape
22:45:16 <joepie91> because it kinda sorta stops the worst leaking
22:45:18 <joepie91> but isn't infallible
22:45:20 <pzuraq> of course, this allows shit giants like Oracle to pump out terrible code
22:45:20 <joepie91> isn't a real fix
22:45:22 <joepie91> isn't nicely done
22:45:27 <joepie91> and has risk of tearing apart at any point
22:46:08 iceTwy_znc (iceTwy_znc@iceTwyznc-07274.users.cryto) has joined #crytocc
22:46:09 <joepie91> pzuraq: the whole "static typing is useful for large codebases" is, as far as I can tell, just one of those "widely assumed bits of wisdom" that doesn't actually have a base in reality
22:46:15 <joepie91> and that everybody diligently repeats
22:46:20 <joepie91> without really having thought about it very well
22:46:31 <pzuraq> joepie91: Ok, I can agree with that description. But I'd point out that with a really good system, duck tape doesn't HARM the system in any way. So at best, it helps.
22:46:37 <pzuraq> at worst, it helps.
22:46:48 <joepie91> pzuraq: sure, until you need to unscrew that lightbulb you duct-taped to the socket
22:46:56 <joepie91> then you'll be screwing around with the duct tape, trying to get it off cleanly
22:46:59 <joepie91> it sticking to your hands
22:47:00 <joepie91> being in the way
22:47:02 <joepie91> and slowing you down
22:47:10 <joepie91> which, guess what, is exactly what static typing systems have a tendency of doing
22:47:22 <Zekka> (I still like Haskell's type system)
22:48:00 <pzuraq> joepie91: Consider getting a pull request for some part of your app you haven't touched in months. You don't want to go over the entire request to make sure it is 100 working, searching for small bugs like accidentally coercing a type.
22:48:20 <pzuraq> with a typed language, if it compiles at least there aren't type bugs.
22:48:31 <pzuraq> there may be others, but some of the work is done for you.
22:49:04 <pzuraq> this is from some rando who never touched the project before, no idea if he's a quality dev or not.
22:49:54 <joepie91> <pzuraq>joepie91: Consider getting a pull request for some part of your app you haven't touched in months. You don't want to go over the entire request to make sure it is 100 working, searching for small bugs like accidentally coercing a type.
22:49:55 <joepie91> um
22:50:04 <joepie91> how is it acceptable to not properly review pull requests
22:50:44 <pzuraq> ok, then you do review it but you are human so you accidentally miss the type error
22:50:54 <pzuraq> typing system catches it
22:51:04 <joepie91> then congrats, you caught one tiny bug
22:51:10 <joepie91> at the cost of slowing down your entire dev process permanently
22:51:53 <pzuraq> Again, I think this is a case by case thing man
22:52:13 <pzuraq> ALL I am saying is that I don't think you can unilaterally state that type systems are bad, mmkay
22:52:15 <pzuraq> it
22:52:27 <pzuraq> *might* be a bit more complicated than that
22:53:12 <joepie91> pzuraq: I still haven't seen any compelling reasons for static typing
22:53:29 <joepie91> other than "well it works as duct tape when you screw up other shit that shouldn't have been screwed up in the first place"
22:53:30 iceTwy_znc has quit (User quit:  ZNC - http://znc.in)
22:53:38 <joepie91> which is a very poor case for a core language feature
22:54:44 <pzuraq> that's like saying safety mechanisms in a factory are a bad idea because if workers don't screw up they won't get hurt
22:56:07 iceTwy_znc (iceTwy_znc@iceTwyznc-07274.users.cryto) has joined #crytocc
23:01:18 <joepie91> pzuraq: not really, no
23:03:35 <DrWhat> 10(151914:153310) 4/8!4\ATTENTION4/8!4\15 joepie91 0,4 KIcK in #crytocc on Cryto-IRC
23:03:35 <DrWhat> Woot
23:06:24 <pzuraq> joepie91: Out of curiosity, what language would you want to write a kernel in if you had to choose?
23:07:43 inttalk has quit (Input/output error)
23:11:23 iceTwy_znc has quit (User quit:  ZNC - http://znc.in)
23:11:38 iceTwy_znc (iceTwy_znc@iceTwyznc-07274.users.cryto) has joined #crytocc
23:16:42 Thor (numz@cryto-DDD48406.telostor.ca) has joined #crytocc
23:17:13 <joepie91> <DrWhat>(19:33) /!\ATTENTION/!\ joepie91  KIcK in #crytocc on Cryto-IRC
23:17:13 <joepie91> bit late
23:17:27 <joepie91> <pzuraq>joepie91: Out of curiosity, what language would you want to write a kernel in if you had to choose?
23:17:34 <joepie91> kernel for?
23:17:42 <pzuraq> an operating system
23:20:06 <joepie91> pzuraq: assuming arbitrary language choice (so not taking into account whether it is currently technically possible), I'd be interested in perhaps Go, but Python would also be quite nice if they got their docs in order
23:20:31 <pzuraq> mmm, would be fun to use Go or Python
23:20:42 <pzuraq> I'm really liking the look of Rust
23:21:08 <pzuraq> also I think an untyped version of C, where everything was just a certain number of bytes, that might be cool
23:21:59 <pzuraq> though that would really be hard when making structs...
23:22:31 <pzuraq> that probably why types were made for C in the first place, just to figure out allocation and dereferencing