| --- Log opened Wed Jul 27 00:00:22 2016 |
| 00:05 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has quit [Ping timeout: 121 seconds] |
| 00:09 | <&[R]> | Alek: JS strings are mutable |
| 00:16 | <@Alek> | meaning? |
| 00:17 | <&McMartin> | That you can change them such that the sequence I gave above with arrays also works |
| 00:17 | <&[R]> | Yes |
| 00:56 | | catalyst [catalyst@Nightstar-bt5k4h.81.in-addr.arpa] has quit [[NS] Quit: Leaving] |
| 01:02 | | catadroid [catadroid@Nightstar-b2e25b.dab.02.net] has joined #code |
| 01:02 | | Derakon [chriswei@Nightstar-5mvs4e.ca.comcast.net] has joined #code |
| 01:02 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 01:10 | | gizmore [kvirc@Nightstar-l2ftk6.dip0.t-ipconnect.de] has quit [[NS] Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] |
| 01:19 | <&ToxicFrog> | Wait, they are? |
| 01:19 | <&ToxicFrog> | aaaaaaaaaa |
| 01:19 | < catadroid> | ? |
| 01:19 | <&McMartin> | JS strings being mutable |
| 01:19 | <&ToxicFrog> | catadroid: JS strings being mutable |
| 01:19 | <&McMartin> | Which among other things would make them unsafe as dictionary keys |
| 01:19 | < catadroid> | Ah right |
| 01:20 | <&ToxicFrog> | I think of mutable strings as an idiosyncracy of C/++ these days |
| 01:20 | <&McMartin> | That said |
| 01:20 | <&McMartin> | I'm looking at a reference and all things that look like mutators say "returns a new string that" |
| 01:20 | <&McMartin> | So I'm going ot have to ask for a citation on this one before I believe it |
| 01:21 | < catadroid> | Yeah, it would surprise me if they were |
| 01:21 | < catadroid> | And yet at the same time it would not |
| 01:21 | <&McMartin> | JS is kind of in that unfortunate space, yes |
| 01:21 | <&McMartin> | Java offers several kinds of mutable strings, but java.lang.String itself is immutable |
| 01:27 | <&[R]> | You can modify a string as an array. |
| 01:27 | <&[R]> | Same as C |
| 01:28 | <&McMartin> | Can you give a sample for that? That implies APIs that I do not see in my documentation. |
| 01:30 | | mac [macdjord@Nightstar-r9vt2h.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 01:33 | <&ToxicFrog> | McMartin, [R]: in my testing, subscript assignment works in the sense that it does not throw, but it also doesn't change the contents of the string. |
| 01:35 | | Turaiel[Offline] is now known as Turaiel |
| 01:42 | <&[R]> | s = 'foo'; s[0] = 'F'; @mcmartin |
| 01:45 | | himi [sjjf@Nightstar-dm0.2ni.203.150.IP] has joined #code |
| 01:45 | | mode/#code [+o himi] by ChanServ |
| 02:29 | <@Alek> | thaaaaat doesn't work in my sandbox. |
| --- Log closed Wed Jul 27 02:36:22 2016 |
| --- Log opened Wed Jul 27 02:41:31 2016 |
| 02:41 | | TheWatcher [chris@GlobalOperator.Nightstar.Net] has joined #code |
| 02:41 | | Irssi: #code: Total of 37 nicks [32 ops, 0 halfops, 0 voices, 5 normal] |
| 02:41 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 02:41 | | mode/#code [+o TheWatcher] by ChanServ |
| 02:41 | | mode/#code [+ao Reiver Reiver] by ChanServ |
| 02:42 | | Irssi: Join to #code was synced in 51 secs |
| 02:44 | <&[R]> | Odd i swear that used to work. |
| 02:46 | <&ToxicFrog> | I can easily believe that it worked (but shouldn't have) in certain browsers/browser versions in the past |
| 02:46 | | Derakon [chriswei@Nightstar-5mvs4e.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 02:47 | <&ToxicFrog> | Considering how often the approach to JS standards conformance has been "eh, fuck it" |
| 02:48 | <&[R]> | Yeah |
| 02:50 | | macdjord [macdjord@Nightstar-r9vt2h.mc.videotron.ca] has joined #code |
| 02:50 | | mode/#code [+o macdjord] by ChanServ |
| --- Log closed Wed Jul 27 03:04:53 2016 |
| --- Log opened Wed Jul 27 03:05:05 2016 |
| 03:05 | | TheWatcher [chris@GlobalOperator.Nightstar.Net] has joined #code |
| 03:05 | | Irssi: #code: Total of 38 nicks [33 ops, 0 halfops, 0 voices, 5 normal] |
| 03:05 | | mode/#code [+o TheWatcher] by ChanServ |
| 03:05 | | Irssi: Join to #code was synced in 55 secs |
| 03:35 | | mac [macdjord@Nightstar-r9vt2h.mc.videotron.ca] has joined #code |
| 03:35 | | mode/#code [+o mac] by ChanServ |
| 03:37 | | macdjord [macdjord@Nightstar-r9vt2h.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 03:57 | | Derakon [chriswei@Nightstar-5mvs4e.ca.comcast.net] has joined #code |
| 03:57 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 04:07 | | abudhabi [abudhabi@Nightstar-7nkq9k.de] has quit [Ping timeout: 121 seconds] |
| 04:13 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 04:13 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 04:14 | | abudhabi [abudhabi@Nightstar-7nkq9k.de] has joined #code |
| 04:15 | | mode/#code [+o abudhabi] by ChanServ |
| 04:19 | | catadroid` [catadroid@Nightstar-r5617l.dab.02.net] has joined #code |
| 04:21 | | catadroid [catadroid@Nightstar-b2e25b.dab.02.net] has quit [Ping timeout: 121 seconds] |
| 04:23 | | abudhabi [abudhabi@Nightstar-7nkq9k.de] has quit [Ping timeout: 121 seconds] |
| 04:35 | | abudhabi [abudhabi@Nightstar-7nkq9k.de] has joined #code |
| 04:35 | | mode/#code [+o abudhabi] by ChanServ |
| 05:04 | | Derakon is now known as Derakon[AFK] |
| 05:49 | | Turaiel is now known as Turaiel[Offline] |
| 06:47 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed] |
| 07:00 | | Kindamoody[zZz] is now known as Kindamoody |
| 07:44 | | Kindamoody is now known as Kindamoody|afk |
| 07:50 | | celticminstrel [celticminst@Nightstar-nhhr58.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 07:55 | | himi [sjjf@Nightstar-dm0.2ni.203.150.IP] has quit [Ping timeout: 121 seconds] |
| 08:22 | | catadroid` [catadroid@Nightstar-r5617l.dab.02.net] has quit [[NS] Quit: Bye] |
| 08:22 | | ErikMesoy [Erik@Nightstar-hq72t5.customer.cdi.no] has quit [Ping timeout: 121 seconds] |
| 09:45 | <@simon> | any windows users here? |
| 09:45 | <@simon> | what's a painfree way to copy files from one directory to another as a scheduled task? |
| 09:54 | <@TheWatcher> | Write a .bat file that does `xcopy /s/e/y c:\from z:\to`, open the task scheduler, create basic task, give it a name, set up the trigger you want (daily, weekly, etc), set the time, ensure "start a program" is selected, select your batch file, click finish. |
| 09:58 | <@simon> | thanks, TheWatcher! |
| 10:00 | <@TheWatcher> | (/s tells it to do a recursive copy, /e includes empty directories, /y supresses confirmation for overwrites) |
| 10:06 | <@simon> | TheWatcher, right. so, uhm, when I've craeted a basic task, where does it go? it doesn't seem to be listed under "Active Tasks" or in the list to the left (2008 Server) |
| 10:07 | <@simon> | ah, it's in "Task Scheduler Library" |
| 10:08 | <@simon> | hm, the .bat file works, but clicking "Run" in the task scheduler doesn't. |
| 10:10 | <@simon> | ah, privileges. |
| 10:10 | <@simon> | okay, great. |
| 10:11 | <@simon> | and it's actually scheduled when it's sitting in Task Scheduler Library? I guess I'll find out. :) |
| 10:14 | | catadroid [catadroid@Nightstar-r5617l.dab.02.net] has joined #code |
| 10:14 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has joined #code |
| 10:14 | | mode/#code [+o himi] by ChanServ |
| 10:15 | < catadroid> | Oh Herb Sutter, I do love you so, but my heart is now functional... |
| 10:49 | | catadroid` [catadroid@Nightstar-qdd4sv.dab.02.net] has joined #code |
| 10:50 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Connection closed] |
| 10:51 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 10:51 | | mode/#code [+ao VirusJTG VirusJTG] by ChanServ |
| 10:51 | | catadroid [catadroid@Nightstar-r5617l.dab.02.net] has quit [Ping timeout: 121 seconds] |
| 10:58 | | catadroid` [catadroid@Nightstar-qdd4sv.dab.02.net] has quit [[NS] Quit: Bye] |
| 10:58 | | catadroid [catadroid@Nightstar-qdd4sv.dab.02.net] has joined #code |
| 11:28 | < catadroid> | Oh. So yes, it turns out that the problem I had programming in Haskell for so long is because I fundamentally didn't understand the execution model |
| 11:34 | < catadroid> | My mind is being turned inside out again |
| 11:34 | < catadroid> | Oh well |
| 11:40 | | * TheWatcher replaces catadroid's mind with a tesseract on a turntable |
| 11:41 | < catadroid> | Replaces? |
| 11:44 | <@TheWatcher> | Heh :) |
| 11:44 | | * Pi waves |
| 11:45 | <@Pi> | catadroid: Want to elaborate on that? Sounds like an interesting insight. :) |
| 11:46 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code |
| 11:46 | | mode/#code [+o Emmy] by ChanServ |
| 12:06 | < catadroid> | Pi: before your explanation of what IO actions really are, it seemed impossible to me that Haskell could actually do anything at all that interacted with the outside world. Now that I understand the concepts behind lisp macros, the key thing that changed that view is that IO actions are irreducible literals and that evaluation is not execution |
| 12:06 | <@Pi> | Ah, cool. |
| 12:07 | <@Pi> | Yup. :) |
| 12:07 | < catadroid> | So now in my head, a Haskell program evaluation produces a list of actions, executing those actions being a single instance of a program run |
| 12:07 | <@Pi> | Yup! |
| 12:07 | < catadroid> | Essentially my model now is that Haskell code is a giant macro and it reduces to a single IO action that may contain composite actions that are sequenced |
| 12:08 | <@Pi> | Right. |
| 12:08 | < catadroid> | That's actually incredibly powerful |
| 12:08 | <@Pi> | (A macro only in the sense of being non-strictly evaluated, of course.) |
| 12:08 | < catadroid> | I did not realise how alien side effects actually are when you think about what they do |
| 12:08 | <@Pi> | Yeah, they really are. |
| 12:08 | < catadroid> | Sure, macro in a LISP sense |
| 12:09 | < catadroid> | So the macro itself may perform 'compile time' evaluation itself, but it produces more choice to be evaluated / executed |
| 12:09 | < catadroid> | And the conflation of evaluation and execution is something that I believe can hurt understanding of code, now |
| 12:10 | <@Pi> | For sure. |
| 12:10 | <@Pi> | Separating them is amazingly clarifying. |
| 12:10 | <@Pi> | And simplifying. |
| 12:10 | < catadroid> | Aye. That's the step that's done the mind blowing |
| 12:10 | <@Pi> | This is what I mean when I say I think in Haskell when I work in other languages. |
| 12:10 | <@Pi> | Once you see that structure you can't unsee it. |
| 12:10 | < catadroid> | And also the concept that I've never seen explained |
| 12:10 | < catadroid> | Until you did |
| 12:11 | <@Pi> | So I usually manage my effects as I would in Haskell, even though the language typically don't give you the assistence that Haskell does. |
| 12:11 | < catadroid> | Because one of the key things basically every programmer learns is that functions may also be actions |
| 12:11 | <@Pi> | And that results in much cleaner and simpler code, most of the time. |
| 12:11 | < catadroid> | And actually, they're separate concepts |
| 12:12 | < catadroid> | Presumably, this means that Haskell libraries can provide new black box actions? |
| 12:12 | <@Pi> | So many design patterns and effective development strategies in imperative and OO languages come down to "approach what you would do in Haskell (or similar)" :) |
| 12:12 | < catadroid> | Which is how you can interact with, eg, graphics hardware |
| 12:12 | <@Pi> | Yeah. |
| 12:12 | < catadroid> | I still am not sure I can forgive you :p |
| 12:12 | <@Pi> | For example, if you wrap C libraries with the FFI, you can expose those as IO actions |
| 12:13 | < catadroid> | (although I'm being facetious, that's actually really cool) |
| 12:13 | < catadroid> | I love gaining this kind of insight |
| 12:13 | <@Pi> | GHC's own primitive IO actions are also just low-level Haskell code at the end of the day. |
| 12:14 | <@Pi> | And you can open them up and muck about with the internals if you really want to. |
| 12:14 | < catadroid> | I want to play more with it now, though |
| 12:14 | <@Pi> | (It's all really low-level, though, and needs familiarity with GHC internals, of course) |
| 12:14 | < catadroid> | Which is annoying, because I was enjoying playing with clojure |
| 12:14 | <@Pi> | But you don't need to know that to use IO actions: they're just a type and an API, and every Haskell implementation can implement it differently. |
| 12:14 | < catadroid> | Not that I can't do both, of course |
| 12:15 | < catadroid> | Yeah, they're essentially machine code with annotations |
| 12:15 | < catadroid> | You learn the verbs of your system the same way you learn numeric values |
| 12:16 | < catadroid> | Just in this case, that's much more explicit |
| 12:17 | <@Pi> | :) |
| 12:17 | <@Pi> | If you like this, I can't wait to introduce you to the really mind-blowing bits that come later. |
| 12:20 | < catadroid> | That feels patronising, but I think that's just my lack of self confidence |
| 12:23 | <@Pi> | I don't mean to be patronising, sorry. |
| 12:23 | <@Pi> | Haskell is just really mind-blowing, and it never really stops doing that. |
| 12:23 | <@Pi> | There's so much built in it. |
| 12:24 | <@Pi> | That's why I'm so passionate about it! |
| 12:29 | <@simon> | there's just so many abstractions you can always learn new ones. |
| 12:32 | | macdjord [macdjord@Nightstar-r9vt2h.mc.videotron.ca] has joined #code |
| 12:32 | | mode/#code [+o macdjord] by ChanServ |
| 12:33 | <@simon> | there's this guy, OCharles, who made a series of christmas calendars called 24 days of (GHC Extensions|Hackage) where he introduced a new feature every day throughout christmas. |
| 12:34 | | mac [macdjord@Nightstar-r9vt2h.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 12:34 | <@simon> | he did it for three years. |
| 12:35 | <@simon> | since Haskell has become such a hotbed of PLT, you'll find state-of-the-art abstractions implemented and ready for use in production. I think this must've been how Lisp felt like years ago. |
| 12:35 | <@simon> | community-wise. |
| 12:37 | <@Pi> | Yeah. It really is the world's premier PLT engineering lab, really. |
| 12:38 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Connection closed] |
| 12:38 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 12:38 | | mode/#code [+ao VirusJTG VirusJTG] by ChanServ |
| 12:39 | <@simon> | here's a pretty cool one: https://ocharles.org.uk/blog/guest-posts/2014-12-06-rebindable-syntax.html |
| 12:43 | <@simon> | I think it's nice that you can learn Haskell with basic knowledge of the type system, or with this category theoretical perspective, and you can learn one from the other or vice versa. |
| 12:44 | <@simon> | "Another thing you might be wondering is if we can replace >>= with =>>, and return with extract (from Control.Comonad) and get some meaningful expression from it. The short answer is: Not really." |
| 12:46 | < catadroid> | I currently feel like I'm worthless and know nothing, which is depressing. And entirely my own issue to deal with |
| 12:46 | <@simon> | I taught a Haskell course for four years, and I still feel so stupid when I'm on #haskell at freenode. |
| 12:46 | <@simon> | s/taught/TA'ed/ |
| 12:46 | < catadroid> | And I want to learn new things, I just feel so much like the time I've spent focusing so hard on C++ is wasted due to how... basic it feels in terms of expressive power |
| 12:48 | <@Tamber> | "*ding* You have reached the first stage of enlightenment." |
| 12:49 | <@simon> | I'm sure it's not wasted. it probably just takes a while for the parallels to come clear. (I don't know C++, but knowing Haskell makes me a better C# programmer for certain.) |
| 12:49 | < catadroid> | It's not wasted at all |
| 12:50 | < catadroid> | This isn't really objective truth speaking, it's the effects of depression |
| 12:50 | <@simon> | ok |
| 12:50 | < catadroid> | So it's not even really related to programing at all, just that's what I'm looking at this minute |
| 12:50 | < catadroid> | Sorry, this isn't really the channel for this introspection |
| 12:50 | < catadroid> | Now I've identified it I'll be quiet |
| 12:50 | <@simon> | :) |
| 12:52 | <@simon> | I've settled with math as a minor because it made me feel too stupid. now I just read it at my own rate and find more enjoyment from it. |
| 12:52 | < catadroid> | I do actually feel like I could write a short introduction to how to use Haskell for programmers that would undo the damage done by people attempting to explain monads |
| 12:53 | <@simon> | you probably could. |
| 12:53 | < catadroid> | Although I want to get a lot more practice with the basic concepts before I would attempt that |
| 12:53 | <@simon> | just never use the word monads. :P |
| 12:53 | < catadroid> | Quite |
| 12:54 | <@simon> | have you read "Abstraction, intuition and 'the monad tutorial fallacy'"? it's a blog post: https://byorgey.wordpress.com/2009/01/12/abstraction-intuition-and-the-monad-tut orial-fallacy/ |
| 12:55 | <@simon> | s/'the /the '/ |
| 12:57 | < catadroid> | Yes, in fact I internalised its ideas as my default teaching method |
| 13:01 | <@simon> | nice. I miss teaching. |
| 13:02 | <@simon> | I stopped because I'd been TA'ing for so long I started to doubt if I could even write code any more. |
| 13:08 | | Turaiel[Offline] is now known as Turaiel |
| 13:12 | < catadroid> | It's more mentoring really |
| 13:33 | <@Pi> | Amen about the above. |
| 13:33 | <@Pi> | That's why I'm so adament about saying "IO action", not IO monad. |
| 13:33 | <@Pi> | *Especially* in context where the monadic functionality isn't even being used. |
| 13:33 | <@Pi> | (Like the entire discussion / tutorial yesterday) |
| 13:34 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 13:34 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 13:36 | <@TheWatcher> | simon: Actually, teaching has actually helped to improve my code. It's made me actually look at some things I'd learnt years ago that proved to be incorrect, as well as made me learn and appreciate some of the subtlties of low-level stuff in c and c++ |
| 13:36 | | * TheWatcher eyes those actuallys, has some of the shot |
| 13:37 | <@Pi> | Teaching is incredibly rewarding. :) |
| 13:37 | <@Pi> | I've hung out in places like #haskell-beginners for years. |
| 13:37 | <@Pi> | And #lambdanow more recently. |
| 13:37 | <@Pi> | (Both Freenode) |
| 13:37 | <@TheWatcher> | Yeah, it is. |
| 14:19 | | Turaiel [Brandon@Nightstar-7mqsi0.mi.comcast.net] has left #code ["Leaving"] |
| 14:21 | <@simon> | hmm, guy interviewed at job was a string theoretician |
| 14:22 | < catadroid> | Was he mutable? |
| 14:22 | <@simon> | apparently he had never touched programming. |
| 14:22 | <@simon> | it's funny how really smart people think they don't need any practical skills to apply for jobs :P |
| 14:23 | <@TheWatcher> | What was he applying for? |
| 14:23 | <@simon> | quantitative analyst |
| 14:23 | <@simon> | there's a lot of programming though |
| 14:24 | <@simon> | neural networks, lots of tweaking of statistical modelling. so very practical, really. |
| 14:24 | <@TheWatcher> | Well, it's just like maths, right?~ |
| 14:27 | < catadroid> | It's very similar to law, I hear |
| 14:28 | <@TheWatcher> | (I've actually had a physicist say that to me about programming...) |
| 14:28 | < catadroid> | Isn't it a lot like martial arts? |
| 14:28 | < catadroid> | Or Mexican food? I heard there are burritos! |
| 14:28 | <@TheWatcher> | Mmm, burritos. |
| 15:03 | <&ToxicFrog> | Pi: just wanted to thank you for all the Haskell burbling, it's explained a lot of things that I didn't really get before. |
| 15:06 | < catadroid> | Easy as Pi I suppose o: |
| 15:06 | < catadroid> | Just needed to circle around and come at it from a different angle |
| 15:06 | | * ToxicFrog ded |
| 15:08 | <@Pi> | ToxicFrog: My pleasure! |
| 15:08 | <@Pi> | I could burble Haskell all day long. :3 |
| 15:08 | < catadroid> | That's good x) |
| 15:09 | < catadroid> | Because there's a lot I would love to learn about it |
| 15:09 | <@Pi> | Well, you can always prompt me! |
| 15:09 | <&ToxicFrog> | My burbling tends to be lua or clojure flavoured; haskell is one of those languages I've poked at repeatedly over the years but have never been productive in or really felt like I understood. |
| 15:16 | | * TheWatcher args, stabs the central timetabling team until they stick to their own godsdamned rules |
| 15:20 | <@TheWatcher> | I had the perfect code to determine academic year dates! It handled Easter correctly, it didn't blink about working out when christmas break started and finished! Its calculated dates matched the last 25 years; as many years as I could get data for. Until this bloody year, where they shifted everything by a week for no reason |
| 15:29 | < Pink> | hah |
| 15:30 | < Pink> | You know what they say about idiot proofing versus the universe's ability to evolve better idiots. |
| 15:42 | | catadroid [catadroid@Nightstar-qdd4sv.dab.02.net] has quit [The TLS connection was non-properly terminated.] |
| 15:42 | | catadroid [catadroid@Nightstar-qdd4sv.dab.02.net] has joined #code |
| 16:21 | | catadroid [catadroid@Nightstar-qdd4sv.dab.02.net] has quit [[NS] Quit: Bye] |
| 16:28 | | catadroid [catadroid@Nightstar-qdd4sv.dab.02.net] has joined #code |
| 16:28 | < catadroid> | Hm. I need to play with the idea of IO actions for a bit |
| 16:29 | | abudhabi [abudhabi@Nightstar-7nkq9k.de] has quit [Connection closed] |
| 16:30 | | abudhabi [abudhabi@Nightstar-7nkq9k.de] has joined #code |
| 16:30 | | mode/#code [+o abudhabi] by ChanServ |
| 16:56 | <@Pi> | catadroid: Let me know if you have any questions! |
| 17:20 | | catadroid` [catadroid@Nightstar-pb0.jtr.132.82.IP] has joined #code |
| 17:23 | | catadroid [catadroid@Nightstar-qdd4sv.dab.02.net] has quit [Ping timeout: 121 seconds] |
| 17:30 | | catalyst [catalyst@Nightstar-bt5k4h.81.in-addr.arpa] has joined #code |
| 17:42 | | celticminstrel [celticminst@Nightstar-nhhr58.dsl.bell.ca] has joined #code |
| 17:42 | | mode/#code [+o celticminstrel] by ChanServ |
| 18:03 | <&McMartin> | re: Haskell feeling now like how Lisp used to feel - both started out as notations for publishing papers about functional programming, so this would make sense |
| 18:04 | <@simon> | http://sigtbd.csail.mit.edu/pubs/veryconference-paper10.pdf -- the dating problem is solved! |
| 18:04 | <@simon> | but there's something I don't understand: |
| 18:04 | <@simon> | "Alice generates two residues x and y modulo N. x consists of 0 (with some suitable padding scheme applied), and y consists of Alice's response (0 if her response is No, 1 if her response is Yes)." |
| 18:04 | <&McMartin> | That said, I've either sadly lost the capacity for wonder or found it in other contexts, because the way I finally got a grip on Monad and related types was "oh, Haskell is batteries included" |
| 18:05 | <@simon> | McMartin, what does batteries included mean? |
| 18:05 | <&McMartin> | Crudely, "rich standard library" |
| 18:06 | <@simon> | I know there's a distribution of OCaml called "Batteries Included". |
| 18:06 | <@simon> | ah, right |
| 18:07 | <&McMartin> | Less crudely, a standard library that is rich and generic enough that you'll be mixing it in with your own code seamlessly. |
| 18:09 | <&McMartin> | Also, I had already done work with lazy-evaluated languages, so the specific bit that's been the topic here wasn't new on its face - it was instead effective use of it that was where I struggled. |
| 18:09 | <&McMartin> | And I was expecting to learn something transformational with it at the same level as when I learned Lisp lo these many years ago |
| 18:09 | <&McMartin> | Which is kind of the "here is a new, large, thing that once you've learned it, changes everything" |
| 18:10 | <&McMartin> | But it seems the experience I should have been looking for with Haskell was more like, well, https://www.youtube.com/watch?v=uCgnWqoP4MM |
| 18:11 | <&McMartin> | (Also, part of my experience with these things tends to include a step along the lines of "Okay, so, you've gone on your journey, what have you brought back" and the answer there seems to be "you should be able to transform your data structures to and from lists in a structurally-preserving way, and here are even more things you can do with lists") |
| 18:27 | <@simon> | I guess a lot of the cool features of Haskell focus on expression, e.g. embedding certain design patterns, algorithms and datastructures into an idiomatic, concise way of doing things. beginners in Haskell will write it like it's an ML without bells and whistles (lots of explicit recursion, lots of function parameters), and seasoned Haskell programmers will write things that are monoidal and pointfree and such |
| 18:27 | <@simon> | . |
| 18:29 | <@simon> | my experience with my co-workers who are already quite good at programming when they see Haskell for the first time is that they don't really see the point. |
| 18:29 | <@simon> | (quite good = better than me) |
| 18:29 | <&McMartin> | It says "point-free" right there |
| 18:29 | <&McMartin> | >.> |
| 18:30 | <@simon> | what? |
| 18:30 | <@simon> | oh. :) |
| 18:30 | <@simon> | the point is that there is no point! |
| 18:31 | <&McMartin> | OK, then, for reference, I picked up point-free almost instantly once it was introduced. "Right, I suppose that does fall out of currying and partial evaluation" |
| 18:31 | <@simon> | no really, I think more generic (mathy) design patterns => better standard library. I think that's really the winning argument in favor of Haskell, for me. |
| 18:31 | <@simon> | yeah, sure. |
| 18:31 | < catalyst> | Transducers feel like a pointfree kind of concept |
| 18:31 | <@simon> | yep |
| 18:32 | <&McMartin> | And the syntax being concise and elegant for certain patterns is true of both Haskell *and* ML, actually... |
| 18:32 | <&McMartin> | ... and is part of why I don't think there's a Tenth Law for it |
| 18:32 | < catalyst> | I still feel like LISP's syntax being literally an AST makes it feel qualitatively different |
| 18:32 | <&McMartin> | I think that's a conventional and valid opinion |
| 18:33 | <&McMartin> | That's why you need a separate "Template Haskell" and why it's kind of terrifying |
| 18:33 | <@simon> | are there even any other homoiconic languages worthy of mentioning? |
| 18:33 | | * catalyst feels like this got heated |
| 18:33 | <&McMartin> | oh my, etc |
| 18:34 | <@simon> | I mean besides Lisp/Scheme/etc. |
| 18:34 | <&McMartin> | It's tricky because as soon as you write one you *might as well* make it a Lisp |
| 18:34 | <@simon> | right :) |
| 18:34 | < catalyst> | assembly? makes me wonder if there's some sort of order variable involved |
| 18:35 | <@simon> | maybe a better question is, are there any other syntaxes for Lisp that are still homoiconic, but less full of parentheses? ;) |
| 18:35 | <@simon> | catalyst, hehe |
| 18:35 | <@Tamber> | I'm sure you could make a version that swaps the parens for curly-brackets~ :p |
| 18:35 | <&McMartin> | There are assemblers that take S-expressions as input, but they aren't homoiconic~ |
| 18:35 | <&McMartin> | Tamber: There's an infix syntax for Scheme |
| 18:36 | < catalyst> | But it's all integers, right? :3 |
| 18:36 | < catalyst> | I'm actually being kind of serious |
| 18:36 | <@Tamber> | It's all just voltage on a bit of wire, in the end, right? |
| 18:36 | <@Tamber> | :p |
| 18:36 | <&McMartin> | Aha |
| 18:36 | <&McMartin> | In that case |
| 18:36 | <&McMartin> | Not assembly |
| 18:36 | <&McMartin> | Machine code is in fact homiconic |
| 18:36 | <&McMartin> | *homoiconic |
| 18:36 | < catalyst> | ah, okay |
| 18:37 | < catalyst> | But if, say machine code is order 1, and LISP is order N (2?) then what's the order N+1 homoiconic language (if such a concept is relevant)? |
| 18:37 | <&McMartin> | I should be bowing out soon, but my initial reflexive reaction is "something with self-describing annotations" |
| 18:38 | <&McMartin> | Probably because we've been talking Haskell and I've been playing with Rust =P |
| 18:38 | <@simon> | McMartin, you mean a typed language? |
| 18:40 | <&McMartin> | Yes, but in the more expansive notion of type than the usual one used in coding. |
| 18:42 | | celticminstrel [celticminst@Nightstar-nhhr58.dsl.bell.ca] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] |
| 18:44 | | celticminstrel [celticminst@Nightstar-nhhr58.dsl.bell.ca] has joined #code |
| 18:44 | | mode/#code [+o celticminstrel] by ChanServ |
| 18:52 | | catadroid` [catadroid@Nightstar-pb0.jtr.132.82.IP] has quit [[NS] Quit: Bye] |
| 18:54 | | * catalyst is beginning to see type as orthogonal to data layout now, which is nice |
| 19:19 | <@celticminstrel> | Oh! Somehow I finally got the text to work. |
| 19:19 | <@celticminstrel> | Except it's upside down. |
| 19:19 | <@celticminstrel> | Presumably that just means I need to reorder the vertices or something. |
| 19:19 | <@celticminstrel> | Or the texture coordinates. |
| 19:20 | <~Vornicus> | YAY TEXT |
| 19:21 | <~Vornicus> | (does your text say "YAY TEXT"? If not, you're still doing it wrong) |
| 19:21 | <@celticminstrel> | Pffft. |
| 19:21 | <@celticminstrel> | Next step is to not use immediate mode, because that was removed in OpenGL 3 or something. |
| 19:22 | <@celticminstrel> | Though for some reason my other rendering also wasn't working when I ask SDL to use OpenGL 3. |
| 19:29 | <&McMartin> | The setup APIs are totally different, and SDL 1.x doesn't support OpenGL beyond 2.x at all. |
| 19:29 | <@celticminstrel> | I'm using SDL2. |
| 19:31 | <@celticminstrel> | And I'm using glVertexPointer with glDrawArrays to draw a vertex array. |
| 19:31 | <@celticminstrel> | (That's what I'm now going to convert the text rendering code to use.( |
| 19:31 | <@celticminstrel> | ^) |
| 19:32 | <@celticminstrel> | Though, if I recall correctly, when requesting GL3 even the clear colour doesn't work... |
| 19:33 | <@celticminstrel> | I recall incorrectly. |
| 19:33 | <@celticminstrel> | It's the only thing that works though. |
| 19:33 | <@celticminstrel> | But if I request GL3 compatibility I just get a black screen. (The clear colour is currently green.( |
| 19:36 | <@celticminstrel> | It seems like the reason my text didn't work is because the vertex colour was transparent. |
| 19:37 | <@celticminstrel> | If it's not transparent, it blends with the text colour... |
| 19:38 | <@celticminstrel> | I guess making it white is best...? |
| 19:38 | <&McMartin> | I think you blend the vertex colors with the texture colors |
| 19:38 | <&McMartin> | So if the texture OR the vertex is transparent, that means the face is |
| 19:39 | <@celticminstrel> | So if the vertex colour is white, will I end up with just the texture colour? |
| 19:40 | | celticminstrel [celticminst@Nightstar-nhhr58.dsl.bell.ca] has quit [Connection closed] |
| 19:40 | | celticminstrel [celticminst@Nightstar-nhhr58.dsl.bell.ca] has joined #code |
| 19:40 | | mode/#code [+o celticminstrel] by ChanServ |
| 19:40 | <@celticminstrel> | Did I miss anything? |
| 19:40 | <@celticminstrel> | And did my last message get through? |
| 19:41 | <~Vornicus> | <celticminstrel> So if the vertex colour is white, will I end up with just the texture colour? |
| 19:41 | <@celticminstrel> | Okay, good. |
| 19:42 | <&McMartin> | It did |
| 19:42 | <&McMartin> | And the answer is "that is my vague recollection of how it works" |
| 19:42 | <~Vornicus> | that's the standard way of doing it, yes. I know it used to be that one way to do lighting was to change the color of a particular vertex |
| 19:43 | <&McMartin> | That's actually what my old terrain demo did |
| 20:13 | <&McMartin> | (The rewrite does it in the pixel shader so it's more like Phong lighting than Goraud shading) |
| 20:13 | <&McMartin> | (Goraud (sp?) was the per-vertex one) |
| 20:15 | <@gnolam> | (Gouraud.) |
| 20:16 | <@gnolam> | (I remember when Gouraud shading was something you actually advertised. I feel old.) |
| 20:17 | <&McMartin> | I don't, but I do remember it for Phong |
| 20:17 | <&McMartin> | Back then Phong usually wasn't done in hardware, though |
| 20:30 | <@Pi> | http://www.inwap.com/mf/reboot/cast/img/phong.jpg ~ |
| 20:31 | < catalyst> | Dat Phong, Ph-Phong Phong Phong |
| 22:02 | <@Alek> | oh, is that from the reboot reboot? |
| 22:02 | <@Alek> | or the original? |
| 22:40 | | Kindamoody|afk is now known as Kindamoody |
| 23:14 | <&jerith> | Original, unless ther reboot reboot has a very similar aesthetic. |
| 23:49 | | celticminstrel is now known as celmin|away |
| --- Log closed Thu Jul 28 00:00:37 2016 |