00:01:41 <alpounet> seriously
00:01:43 <alpounet> still no idea
00:02:26 <stepcut> I am practicing violin before it gets too late
00:04:16 <alpounet> ok
00:29:11 <stepcut> ok.. so the purpose of the buildbot is testing and safety. We just throw away the binaries it generates. One thing that makes it unique is that it is dedicated to building and testing cabal packages (using cabal install)
00:30:43 <alpounet> yeah
00:33:14 <stepcut> so things related to protection might work (like guardian or taster) or things related to testing and record keeping (like auditor)
00:33:55 <stepcut> it's also about quality assurance and early detection
00:34:14 <stepcut> we want to know about problems as soon as possible -- which is why test as soon as possible
00:35:33 <alpounet> so maybe a kind of inspector
00:35:34 <alpounet> detective
00:35:56 <stepcut> yeah
00:36:12 <stepcut> though it doesn't really figure anything out or solve problems..
00:36:25 <alpounet> it just figures them out yeah
00:38:02 <stepcut> it's more like the egg tester in charlie and the chocolate factory..
00:38:18 <stepcut> but 'veruca salt' is already taken :p
00:38:39 <alpounet> hah
00:40:52 <alpounet> wait
00:40:55 <alpounet> gonna ask my gf
00:41:06 <stepcut> that's what I usually do..
00:41:37 <stepcut> saint peter
00:41:43 <stepcut> :p
00:51:00 <alpounet> hah
00:52:35 <alpounet> ok stepcut my gf is on it
00:52:39 <alpounet> doing some research etc
00:53:02 <alpounet> had to explain a bit about breaking changes, etc hah
00:53:13 <stepcut> :)
00:54:57 <alpounet> i found hades
00:55:22 <alpounet> but well no, it leads to hell so yeah maybe not :p
00:58:14 <stepcut> :P
01:12:11 <alpounet> heed ?
01:12:13 <alpounet> heed bot ?
01:12:35 <alpounet> heeding bot ?
01:13:40 <alpounet> hectic ?
01:14:43 <stepcut> hectic: Relating to or denoting a regularly recurrent fever typically accompanying tuberculosis.
01:16:32 <alpounet> i kinda like heed though
01:22:02 <stepcut> we should put it on the list of potential names
01:22:12 <stepcut> though. so far it *is* the list :p
01:22:33 <alpounet> ok hm
01:22:44 <alpounet> can i edit the wiki page ?
01:23:51 <stepcut> yes
01:23:57 <stepcut> I hope so anyway
01:24:47 <alpounet> hm
01:24:52 <alpounet> i only can "enter a comment"
01:26:15 <stepcut> are you logged in ?
01:26:31 <alpounet> yeah
01:26:49 <stepcut> let me check some stuff
01:27:48 <stepcut> under the wiki tab, does it say, "new page" search edit delete ?
01:28:10 <stepcut> above the page content
01:28:12 <stepcut> http://code.google.com/p/happstack/wiki/BuildBot
01:28:41 <stepcut> (I couldn't find the edit button at first myself.. trying to figure out if it is really missing for you, or if you just haven't found it yet)
01:29:13 <stepcut> looks like it only allows 'search' but not 'edit or delete' if I log in as a different user
01:29:27 <alpounet> it just says "Search"
01:29:52 <alpounet> no edit, delete
01:30:12 <stepcut> let me add you as a member
01:31:22 <stepcut> try now
01:32:51 <alpounet> yeah, i see them now, thanks
01:33:17 <stepcut> that's a bit annoying :-/
01:33:26 <stepcut> the new website will eventually fix that though
01:34:05 <alpounet> happstack powered? :]
01:36:20 <stepcut> yes
01:37:56 <alpounet> with wiki, blog, etc?
01:39:24 <stepcut> of course
01:50:10 <alpounet> nice
02:04:10 <alpounet> stepcut, ok, so something around "heed" is the best we have tonight
02:04:16 <alpounet> i added a relevant section to the wiki page
02:04:28 <stepcut> nice
02:11:17 <alpounet> <Axman6> ok, this is something that hackage needs to do. whenever you upload a package, it keeps track of what your package depends on. when a dep changes, it sends you an email saying you need to make sure your code is up to date
02:11:17 <alpounet> <Axman6> yes i can see problems with this of course, but it'd be a start
02:11:17 <alpounet> <ivanm> that's what lispy's build-bot proposal was for
02:11:17 <alpounet> <ivanm> and what packdeps is partially aimed at
02:11:39 <alpounet> potential users! heh
02:12:58 <stepcut> that is what hackage2 is for ;)
02:13:36 <stepcut> no features ever get added to hackage 1 because it is a bunch of perl scripts and the future is hackage 2.. but then hackage 2 got stalled for a long time
02:14:44 <alpounet> yeah, but a few people started working on where hackage2 was left
02:14:47 <alpounet> that's promising
02:14:59 <stepcut> yup
02:15:23 <stepcut> I have worked on it.. but only to upgrade it to happstack 6 and acid-state
02:15:29 <stepcut> not any new features
02:16:23 <alpounet> gonna update it to happstack 7 ?
02:16:42 <alpounet> or will you wait for happstack 8 with the new smoking fast backend? :p
02:17:04 <stepcut> it's already up-to-date for Happstack 7
02:17:53 <stepcut> during this dev cycle, we just released everything as it was developed instead of saving up all the changes for a big release
02:18:04 <stepcut> and there were less breaking changes
02:18:28 <stepcut> the big breaking change for Happstack 7 was changing from happstack-state to acid-state, and I already did that for hackage 2
02:18:54 <stepcut> Happstack 8 will definitely break a few things. But in a good way. And easy to update. So I'll submit patches for those things as they happen
02:19:09 <alpounet> what's planned for happstack 8 ?
02:19:17 <stepcut> everything!
02:19:19 <stepcut> :p
02:19:54 <alpounet> happstack-coffee-machine
02:19:55 <alpounet> yyay
02:20:12 <stepcut> the big things are: finishing web-routes + plugins. Re-examining the backend. And re-examining the 'dynamic' routing combinators (dir, path, etc)
02:21:26 <stepcut> would also be nice to see improvements to HSX and acid-state (multimaster/sharding, a better IxSet, etc)
02:21:40 <alpounet> are you gonna use HiggsSet?
02:21:49 <stepcut> maybe
02:22:09 <stepcut> lpeterse is going to do a writeup about when it is better and when IxSet is better
02:22:18 <stepcut> we may end up having both, but improving the type-safety of IxSet
02:22:29 <alpounet> ok
02:22:40 <stepcut> also, would be nice to explore in-memory compression of those collection types
02:23:10 <stepcut> but getting happstack-server rock solid is the most important thing
02:23:51 <stepcut> it's pretty good now.. but it could be faster, avoid lazy IO, and have more tests/other forms of proof that it is correct
02:24:20 <alpounet> ok i see
02:24:28 <stepcut> I try not to think about 8 too much, because then I will want to work on that instead of finishing 7 :p
02:25:11 <alpounet> hah
02:25:44 <stepcut> testing that everything still works on Win32: boring. playing with Pipes: exciting!
02:26:02 <stepcut> one reason for wanting to have a buildbot that does all the testing automatically ;)
02:29:05 <stepcut> also have to write-up all the GSoC proposals for Happstack
02:30:27 <alpounet> hah
02:30:33 <alpounet> i would have liked to do that as a GSoC
02:30:43 <alpounet> if it didn't imply working 3 months full-time ...
02:30:47 <stepcut> :)
02:31:21 <stepcut> there is a bunch of low-hanging fruit for acid-state related projects still.. but things where 3 month of fulltime work could make a huge difference
02:35:42 <alpounet> yeah, but i think i prefer doing a more casual job for the summer
02:36:01 <alpounet> working less than 3 months, like 2 and keeping one to travel a bit, do some fun stuffs etc
02:37:47 <stepcut> what could be more fun than creating some of the coolest new technology on the planet!
02:37:58 <stepcut> time for contortion practice, bbiab.
02:43:31 <alpounet> well, for a starter, i need the last two weeks of august!
03:16:17 <alpounet> off to bed, good night!
03:16:54 <stepcut> night!
11:46:26 <alpounet> stepcut, i'm still liking heed or heed-bot today
15:23:35 <Lemmih> This is odd. My computer says my USB drive is plugged in but I can't find it.
15:38:30 <srhb> Lemmih: What OS?
16:28:45 <KorriX1> hello
16:29:00 <KorriX1> how to change Html renderer into happstack ?
16:29:26 <KorriX1> or how to rewrite correctly instance ToMessage Html ?
16:30:43 <donri> I don't think there's a sane way to replace an existing instance like that; you'd have to wrap it in a newtype or something such
16:33:14 <stepcut> KorriX1: what is wrong with the existing instance
16:33:37 <KorriX1> i want "pretty" blaze renderer
16:34:01 <KorriX1> Text.Blaze.Renderer.Pretty
16:34:18 <stepcut> ah
16:35:01 <donri> yea I do that in development (using CPP conditionals) but I just have my own "respond" function that does some other things too
16:35:05 <donri> newtype would probably be cleaner
16:35:06 <stepcut> as donri suggests, the only way is to use a newtype wrapper like, newtype PrettyHtml = PrettyHtml { toBlazeHtml :: Html }
16:35:17 <stepcut> and then create an, instance ToMessage PrettyHtml
16:35:34 <donri> why a record newtype though, just pattern match :)
16:35:40 <stepcut> but then you have to explicitly do, toResponse $ PrettyHtml $ <blaze html code>
16:36:05 <KorriX1> okej
16:36:08 <KorriX1> thenaks
16:36:10 <KorriX1> *thanks
16:36:52 <stepcut> one thing I do not like about the ToMessage class is that it forces you to pick a single representation for a type -- which is not always appropriate
16:37:09 <donri> OTOH that sorta makes sense, esp. since we have newtype
16:37:11 <stepcut> another option would be to make a function like:: prettyHtml :: Html -> Response, and just use that
16:38:02 <donri> but the typeclass let's you pass around a PrettyHtml and have it convert to response where needed, independent of prettification
16:38:18 <stepcut> prettyHtml html = toResponseBS (B.pack "text/html;charset=UTF-8") (prettyRenderer html)
17:31:39 <stepcut> central scrutinizer
17:33:48 <stepcut> overseer, custodian, steward, warden
17:34:18 <stepcut> sentinel, shepard
17:35:08 <stepcut> watchdog
17:35:50 <stepcut> watchkeeper
17:37:49 <donri> caretaker? star trek!
17:38:53 <stepcut> which star trek? I have only seen the original and NG
17:39:19 <donri> voyager
17:39:31 <stepcut> ah
17:41:56 <donri> .oO( "vigor"; from synonym for "wellbuilt" )
17:42:24 <stepcut> hmm
17:45:12 <donri> hey while we're on star trek; "duras" (klingon house), because similarity to "durable"
17:45:27 <stepcut> heh
17:46:33 <donri> duress, because that's what it puts the software under
17:47:07 <stepcut> yeah
17:47:10 <stepcut> i like duress
17:51:51 <stepcut> testomatic
17:52:02 <stepcut> cabaltron
17:52:06 <donri> testosterone!
17:52:15 <donri> testicular cancer?
17:52:24 <stepcut> heh
17:54:56 <donri> testify
17:55:05 <stepcut> ooo
17:55:16 <stepcut> I still like 'saint peter'
17:55:21 <donri> wat
17:55:26 <stepcut> testify is pretty good though
17:55:36 <stepcut> integrity
17:55:37 <donri> actually "wat" is a potential name: when it goes wrong you say "wat"
17:56:31 <stepcut> judgement
17:57:02 <donri> or just "judge"
17:57:12 <donri> "the judge says this build failed"
17:57:45 <donri> "that patch was judged a failure" :D
17:58:26 <stepcut> umpire
18:04:23 <stepcut> ascertain
18:07:41 <stepcut> gold star
18:08:12 <stepcut> gavel
18:08:59 <stepcut> diligence
18:09:50 <stepcut> discretion
18:11:05 <stepcut> frisk
18:11:23 <donri> obamacare!
18:11:48 <stepcut> perlustrate
18:14:56 <stepcut> I kind of like frisk
18:17:05 <stepcut> ransack
18:17:21 <donri> shakedown
18:18:14 <stepcut> ooo
18:20:26 <donri> buildstack
18:21:32 <donri> bakery
18:24:16 <stepcut> except we through away the generated binaries
18:24:25 <stepcut> we only care if they built
18:24:47 <stepcut> but we don't use the built binaries for anything
18:25:48 <stepcut> so, we want to test that everything builds and report errors. An early warning system that something has gone wrong
18:26:16 <stepcut> and report when the packages on hackage have broken, or when they need to be updated due to newer build dependencies
18:28:21 <stepcut> so it is the reporting and watching that is the important part
18:28:56 <stepcut> the building only happens for the purpose of reporting
18:37:18 <stepcut> it's a system with gathers information and generates reports and alerts so that we can detect problems as soon as possible
18:54:39 <alpounet> ACTION still likes heer
20:02:23 <donri> ACTION finally manages to make jmacro quasi quotations syntax highlighted as javascript in vim
20:03:51 <donri> stepcut: do you have any examples of typechecking jmacro?
20:04:02 <donri> it's rather badly documented
20:04:35 <stepcut> donri: nope. I have never even tried.
20:05:23 <donri> all i can figure is it seems to be something you need to do at runtime and with a "context" (variable types perhaps?)
20:05:58 <donri> which isn't quite as nice as compile-time but i guess something i can add to my test suite
20:09:07 <stepcut> well, since you can splice in code generated at runtime.. it would be hard to check that at compile time I think
20:12:46 <donri> i wonder if it's by design that it doesn't render == as ===
20:13:39 <stepcut> no idea
20:13:45 <stepcut> you could ask
20:16:50 <donri> yea i'm considering emailing the author :)
20:17:34 <stepcut> i have emailed him and twisted his arm into making changes a few times :p
20:18:36 <donri> hehe
20:19:34 <stepcut> could  be nice to add coffeescript support
20:24:57 <donri> not sure about that, coffeescript is even more insane than plain javascript
20:25:06 <stepcut> :)
20:25:26 <donri> http://lucumr.pocoo.org/2011/12/22/implicit-scoping-in-coffeescript/
20:27:23 <donri> jmacro's extensions already get you quite far in terms of improved readability
20:28:31 <stepcut> yeah
20:28:46 <stepcut> jmacro needs to have a pure javascript mode, and an extended javascript mode IMO
20:29:17 <donri> YUI().use "node" \y { ... }  vs  YUI().use("node", function(y) { ... })
20:30:40 <donri> it's nice that JS has anonymous functions/closures but they're so wordy!
20:30:54 <donri> i particularly dislike all the }); lines you end up with
20:32:57 <stepcut> yeah
20:33:23 <stepcut> what I really want to see, though, is a sane way of sharing data structures between Haskell and javascript
20:34:09 <stepcut> right now I have to: (1) define a Haskell data structure (2) manually specify how to convert that to and from JSON (3) manually specify how to turn the JSON into a javascript data structure
20:34:12 <donri> http://i.imgur.com/5fPAk.png -- much nicer with jmacro extensions and haskell-style formatting :)
20:34:16 <stepcut> and then I have to keep everything in sync
20:34:32 <stepcut> heh
20:35:29 <stepcut> but what I really really want to happen is a VM in the browser.. though interfacing with OO interfaces  would  still be a problem even then
20:35:39 <donri> antiquoting only works for JSON?
20:37:17 <donri> have you seen chrisdone's "ji" library
20:38:16 <stepcut> vaguely
20:38:36 <stepcut> I am aware that it exists, and that it works, in part, by maintaining 1 server-side thread per client ?
20:38:37 <donri> it's basically "reverse RPC"
20:38:51 <donri> set up a websockets or similar connection and control the DOM from the server-side
20:38:54 <donri> more or less.
20:38:56 <stepcut> ah
20:39:03 <stepcut> neat
20:39:14 <stepcut> guess we need to add websockets to happstack
20:39:31 <stepcut> though I think there is a library already that does most of the work
20:39:44 <donri> innovatively called "websockets" if i recall
20:39:47 <donri> @hackage websockets
20:39:47 <lambdabot> http://hackage.haskell.org/package/websockets
20:58:34 <alpounet> stepcut, so? heed? testify?
21:00:11 <stepcut> alpounet: undecided still :-/
21:00:35 <alpounet> these are the two that i liked
21:01:30 <stepcut> the essence of the tool is "it's a system with gathers information and generates reports and alerts so that we can detect problems as soon as possible". I am trying to think of job or something that is similar in nature
21:01:57 <donri> internal affairs
21:02:11 <alpounet> alarm?
21:02:15 <alpounet> halarm? :p
21:02:16 <stepcut> testify is a great name.. but seems to hint too much at being specific to unit tests
21:03:01 <donri> minority report
21:03:11 <stepcut> what sort of person would gather information and create reports to detect problems?
21:03:13 <donri> detecting crime before it happens ;)
21:03:16 <donri> aka. before release :)
21:03:22 <stepcut> reportor and auditor are close.. but not quite  right
21:03:32 <donri> secret agent
21:03:48 <stepcut> yeah
21:03:52 <stepcut> but it's not secret in this case
21:04:01 <alpounet> hagent
21:04:04 <donri> true dat
21:04:21 <donri> closest thing to a non-secret agent is a reporter/journalist
21:04:49 <donri> investigator?
21:05:01 <donri> an investigation can be open and public
21:05:38 <stepcut> investigator implies too much inference and deductive reasoning though.. like there is a problem and something is going to solve it
21:05:50 <alpounet> yeah, same problem as with detective etc
21:06:02 <donri> not necessarily, but i guess popularlty
21:06:25 <stepcut> journalist is closer because it implies collecting and presenting information.. but something is not quite right about that either
21:06:31 <alpounet> yeah
21:06:31 <donri> i guess we want something closer to "standard procedure"?
21:06:52 <donri> like, tests for qualification [for driver's license for example]
21:06:53 <alpounet> coming to you next week, the Haskell Standard Procedure
21:06:54 <stepcut> i'm trying to think if there is a military position or something that would be relevant
21:06:55 <alpounet> tatataaaaan
21:07:19 <alpounet> stepcut, a radar? :P
21:07:29 <stepcut> alpounet: sort of
21:07:37 <donri> scout?
21:07:47 <stepcut> surveillance
21:08:01 <donri> i like scout
21:08:12 <donri> "the scout system reports bla bla"
21:08:31 <donri> "reconnaissance"
21:08:36 <stepcut> yeah
21:08:56 <donri> then you can call the beta version "boy scout" ;)
21:09:26 <alpounet> hah
21:10:15 <donri> scout is a verb too
21:10:18 <stepcut> reconnoiter
21:10:19 <alpounet> ACTION would actually like to do that
21:10:21 <donri> "i've scouted a problem" after running a build
21:11:04 <stepcut> scout could work
21:12:10 <donri> "A person sent out to gain and bring in tidings; especially, one employed in war to gain information about the enemy and ground." where bugs are the enemy ;)
21:12:32 <stepcut> an act or instance of reconnoitering, inspecting, observing, etc.
21:12:39 <stepcut> a person sent out to obtain information.
21:12:44 <alpounet> i love that!
21:13:38 <stepcut> i wonder if there is someway we can personalize it so that it is easier to search for
21:13:41 <stepcut> like, skout
21:13:47 <stepcut> but somehow relevant
21:13:54 <donri> "Always prepared" ... to release!
21:14:18 <donri> oh, guess it's "be prepared" in english
21:14:18 <alpounet> hascout
21:14:25 <stepcut> :-/
21:14:29 <alpounet> hah
21:14:40 <stepcut> scobal
21:14:45 <stepcut> :-(
21:14:57 <alpounet> naah
21:15:05 <donri> shout uuh
21:15:09 <stepcut> scoutomatic
21:15:16 <stepcut> scoutron
21:15:42 <donri> meh, not hard to search for "haskell scout" or whatever
21:15:45 <alpounet> \scout / lambdascout
21:15:53 <alpounet> naaa
21:15:53 <donri> oh i like that one
21:15:55 <stepcut> I think it is ok to have a playful name for this particular project
21:16:22 <alpounet> once published, we just would have to have the best page rank for the "scout" keyword
21:16:32 <donri> scoutstack! *inventive*
21:18:16 <stepcut> I would definintely prefer a term that currently has close to 0 hits on google
21:18:30 <stepcut> so, some perversion of scout would be good
21:18:46 <alpounet> lambdascout has almost none
21:18:56 <donri> scoutonaut
21:19:19 <stepcut> heh
21:20:03 <donri> scoutbot, roboscout
21:20:48 <stepcut> i like roboscout
21:21:16 <stepcut> though there is already something called roboscout
21:22:00 <alpounet> and there's a twitter account for "scoutbot"
21:22:20 <stepcut> surely we can do better  !
21:22:38 <stepcut> I am going to make some lunch and think up more variations
21:22:42 <donri> scoutess, because geeks insist on calling bots female
21:22:53 <stepcut> ooo
21:23:11 <donri> 26k results on google :(
21:23:20 <alpounet> that could still do it
21:23:38 <alpounet> don't feel any of the first results will stay there once we get smth out
21:23:41 <stepcut> yeahÂ… nothing too high ranking
21:24:03 <stepcut> scouticus
21:24:12 <stepcut> I am scouticus
21:24:34 <donri> of borg?
21:26:00 <stepcut> heh
21:26:15 <stepcut> I am liking scoutess so far
21:28:02 <alpounet> $ scoutess happstack-server
21:32:19 <stepcut> ok
21:32:22 <stepcut> now we can make a repo
21:33:17 <alpounet> should i create it on patch-tag, github, darcsden, c.h.o?
21:33:23 <stepcut> are you willing to use the most awesomest DVCS in the  world (darcs) or would you prefer that pile of crap that Linus foisted on the world after a day of work (git)
21:33:51 <stepcut> everything else happstack related is on patch-tag in darcs
21:34:09 <stepcut> and I have no idea how to use git ;)
21:34:18 <stepcut> but I did create a github happstack organization
21:35:05 <alpounet> i guess it would be stupid to have scoutess somewhere else than patch-tag
21:35:13 <stepcut> yeah
21:35:35 <donri> especially if you're going to set up github mirroring
21:35:39 <stepcut> yeah
21:35:39 <donri> anyway
21:36:12 <stepcut> when I redo the site, I am thinking of moving the official darcs repos to, src.happstack.com and then mirroring them at github
21:36:26 <alpounet> stepcut, so, i create the repo?
21:37:01 <donri> is that going to be a patch-tag instance or just plain fileserve?
21:37:39 <alpounet> an usual patch-tag project i guess
21:37:53 <donri> i mean src.happstack.com
21:39:38 <alpounet> oh, ok
21:40:03 <stepcut> donri: it will be neither
21:40:46 <stepcut> it will be what patch-tag should be :p
21:41:12 <stepcut> it'll make more sense when you see it all
21:41:32 <donri> :)
21:41:44 <donri> how bout issue tracking
21:41:58 <stepcut> that too
21:42:03 <donri> cool
21:42:11 <donri> just don't aim too high or it'll never get done!
21:42:22 <stepcut> by self-hosting the source, we can then integrate scout reports into the history listing and stuff
21:42:36 <donri> ah yea
21:42:49 <stepcut> well, the site will be done incrementally
21:42:53 <donri> sure
21:42:59 <alpounet> stepcut, repo created + i added you there
21:43:01 <stepcut> to get the first rev out I just need to finish the theme mostly
21:43:04 <donri> launch the MVP early ;)
21:44:09 <stepcut> alpounet: now run cabal init :p
21:44:21 <stepcut> I have to eat and then go to the park
21:44:38 <stepcut> but I can hack out a framework tonight I think
21:44:49 <stepcut> where I create all them modules and add some types
21:46:41 <stepcut> if you are ready to start coding, I can give you some suggestions for things to start on
21:52:19 <alpounet> stepcut, BSD3, righT?
21:52:28 <stepcut> yes please
21:52:36 <stepcut> this project is going to be pretty neat I think
21:55:01 <alpounet> that's gonna be fun yes
21:55:13 <alpounet> ok cabal init-ed the project
21:55:16 <stepcut> nice
21:55:22 <alpounet> (and pushed)
21:56:17 <stepcut> so, now we need to figure out how to do a bunch of things
21:56:44 <stepcut> for example, is there a library version of 'cabal install' or do we need to call the executable
21:56:59 <stepcut> and, should we use, cabal-dev, virtualenv, or something else (hsenv?)
21:57:17 <alpounet> hm
22:00:18 <stepcut> and we have to figure out how to generate the index file that cabal looks at (which is probably easy)
22:01:38 <alpounet> i'm investigating the cabal-dev / virtualenv / hsenv question
22:02:01 <stepcut> excellent!
22:02:57 <stepcut> the key requirements that I am looking for is the ability to ignore the global package conf so that we can build against a clean haskell platform only base.. and the ability to support multiple versions of ghc
22:05:44 <stepcut> s/I/we/
22:07:28 <donri> simple matter of environment variables in both cases i think
22:08:51 <stepcut> cool
22:09:43 <alpounet> and they all handle that anyway
22:14:50 <alpounet> hm
22:15:05 <alpounet> hsenv doesn't seem to be on hackage
22:17:15 <alpounet> maybe i should email its maintainer
22:17:24 <alpounet> 'cause that's supposed to be the successor of virtualenv
22:20:23 <stepcut> is it in a public repo somewhere?
22:22:20 <donri> https://github.com/Paczesiowa/hsenv maybe
22:24:03 <alpounet> yep
22:24:09 <alpounet> the project url is a hackage url
22:24:13 <alpounet> but it isn't valid
22:25:31 <stepcut> hmm
22:26:56 <alpounet> so is there any maturity issue here?
22:27:04 <alpounet> should we rather rely on virtualenv or cabal-dev?
22:27:56 <donri> it's virthualenv with a h ;)
22:30:04 <alpounet> oh yeah sorry
22:47:16 <stepcut> you tell me
22:50:26 <stepcut> I have no idea what the difference is or why to pick one over the other..
22:52:42 <alpounet> i'm investigating :P
22:53:41 <stepcut> I'm going to the park, bbiab
23:11:14 <alpounet> haha "If you'd like to hear more, consider sending your Github profile and favorite Haskell one-liner to: missionhackerhouse@gmail.com"