Perl 6 - the future is here, just unevenly distributed

IRC log for #rosettacode, 2011-03-12

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

All times shown according to UTC.

Time Nick Message
00:07 FireFly left #rosettacode
00:42 MigoMipo left #rosettacode
03:13 kpreid joined #rosettacode
06:26 mwn3d_phone left #rosettacode
06:48 mwn3d_phone joined #rosettacode
09:43 MigoMipo joined #rosettacode
12:21 FireFly joined #rosettacode
13:33 MigoMipo_ joined #rosettacode
13:35 MigoMipo__ joined #rosettacode
13:36 MigoMipo left #rosettacode
13:39 MigoMipo_ left #rosettacode
13:57 MigoMipo__ left #rosettacode
13:59 MigoMipo joined #rosettacode
15:44 shortcircuit http://rosettacode.org/mw/index.php?title=​Rosetta_Code:Village_Pump/Suggest_a_progra​mming_task&diff=103238&oldid=prev
15:44 fedaykin "Rosetta Code:Village Pump/Suggest a programming task - Rosetta Code" http://rldn.net/FsCo
15:44 shortcircuit [[RCPiet]], anyone? Sounds immenently doable.
15:44 fedaykin http://rosettacode.org/wiki/RCPiet (Doesn't exist.)
15:55 Hypftier It's quite complex, though. Especially determining which codel block to execute next in non-trivial images :)
15:59 Alegend[RCo] joined #rosettacode
16:00 Alegend[RCo] Hello!
16:01 Alegend[RCo] IRC sure is quiet today.
16:43 mwn3d_phone Activity here comes in bursts
16:43 mwn3d_phone I've gone whole days without seeing anything go on here
16:44 mwn3d_phone And then I've missed whole conversations cause they go off my scrollback
16:45 moah joined #rosettacode
16:46 moah hello, anybody around?
16:48 moah what to do when there are basically two tasks for the exact same problem, but worded slightly differently?
16:49 moah http://rosettacode.org/wiki/Multiple_regression and http://rosettacode.org/wiki/Polynomial_regression basically do the same, but differe only in the example.
16:49 fedaykin "Multiple regression - Rosetta Code" "Polynomial regression - Rosetta Code"
16:51 moah and also there are solutions for the multiple regression, where people have already discussed that the task was worded poorly, so the solutions differ, but no solution has been reached.
16:51 moah what to do with that?
16:56 moah left #rosettacode
17:02 mwn3d_phone I'm gonna try to make a task to see who is eliminated from the NHL playoffs. I'll have to gather up the tiebreaking rules and make a solution myself but that will be fun.
17:22 mwn3d_phone left #rosettacode
17:26 mwn3d_phone joined #rosettacode
17:44 parsleyfirefly left #rosettacode
17:51 mwn3d_phone left #rosettacode
17:52 Alegend[RCo] left #rosettacode
18:00 mwn3d_phone joined #rosettacode
18:04 parsleyfirefly joined #rosettacode
18:08 * TimToady wonders how removing the comment on 'range(1,11) # 1..10' can be construed as "responding to criticism" in any positive way :)
18:08 TimToady it's a prime example of one way that Python forces you to say something manifestly unnatural...
18:09 TimToady but the talk was not primarily intended to attack Python, but to discuss linguistics
18:10 shortcircuit TimToady: I don't think it's a positive response to the criticism. I think he may have taken it personally.
18:11 shortcircuit Personally, I think the correct solution to case like that is to improve the documentation in and around the example, in an explanatory fashion.
18:11 TimToady well, it wasn't intended that way, fershure
18:11 shortcircuit TimToady: *nod*
18:12 TimToady I just thought it said something interesting about psychology that there was a (temporarily) perceived need for the comment
18:12 TimToady and that it illustrated the main point of the talk
18:12 shortcircuit TimToady: I did like your evolve sequence going from the Perl example to the Perl 6 example. I think that's a very good way to explain P6.
18:12 TimToady that languages differ more in what you must say, not in what you can say
18:13 shortcircuit True enough.
18:14 shortcircuit I do think that increased verbosity and explicitness in some languages aids their use in particular use cases.
18:14 TimToady to the first approximation, Perl 6 is competing with itself, not with any other language
18:14 shortcircuit In other languages, syntactic and symbolic compactness lends them to other use cases.
18:15 TimToady sure, and I like a language with a large dynamic range so you can do either :)
18:15 shortcircuit I've got no argument with that.
18:15 TimToady well, there are always tradeoffs...
18:16 * shortcircuit ponders
18:16 shortcircuit In some cases, verbosity and explicitness helps reduce the likelihood of accidental bugs.
18:17 shortcircuit Take 'use strict', for example.
18:17 TimToady well, there are forms of complexity that contribute to the problem's solution (in whatever space you imagine it), and other forms of complexity that are gratuitous
18:18 TimToady we try to get rid of the latter without getting rid of the former
18:19 TimToady the reason a plane lifts its landing gear for most of the flight...
18:19 shortcircuit There are always pragmatic decisions at play.
18:19 TimToady a perfectly streamlined airplane still has some drag though
18:19 shortcircuit It may be simpler for a pilot if he didn't have to raise and lower the landing gear, but it costs a heck of a lot more in fuel if the gear's hanging out.
18:20 shortcircuit I was thinking about your pointing out having to import math functions in both Java and Python.
18:20 TimToady I did rather harp on that
18:20 shortcircuit In both cases, it can make some sense in reducing runtime requirements.
18:21 TimToady which can be reduced by lazy loading as well
18:21 TimToady the only real cost to having an operator is a name in the symbol table somewhere in the search path
18:21 TimToady well, and the grammar rule to parse it...
18:22 TimToady implementations don't have to get there until needed though
18:22 shortcircuit I was thinking also about how Java and Python differ in their usage of the imported function.
18:22 TimToady I don't profess to know the difference
18:22 shortcircuit Java requires you to specify the namespace for all imported symbols. Python doesn't.
18:22 shortcircuit (Math.sqrt() vs simply sqrt())
18:23 TimToady ah, Python imports as part of a search path, iirc
18:23 shortcircuit (Based on the code examples you used)
18:23 Hypftier shortcircuit: though you can use import static in Java in recent versions to make that a little easier.
18:23 TimToady which is sane, unless you need grammar mods
18:23 Hypftier But indeed, that's a problem that many more or less pure OO languages have -- there is no provision for non-class functions
18:24 TimToady Perl 6 is trying to be a highly mutable language (while giving you enough built-in stuff to discourage you from wanting to mutate it :)
18:24 shortcircuit Do any pure-ish OO languages have implicit link-object or app-object classes?
18:24 TimToady and Perl has always been operator-centric
18:25 Hypftier shortcircuit: »link-object or app-object classes«?
18:26 TimToady what Hypftier said...
18:26 shortcircuit Hypftier: "link object" -> Thinking about C and C++; you compile a .c file to an object file, which then gets linked. "app-object" -> Unless explicitly defined, declarations and code are assumed to be within an implicit class implementation.
18:27 TimToady I still don't understand the question...
18:27 shortcircuit It's based on a nebulous concept in my head, which probably doesn't help.
18:27 Hypftier Do you mean that you don't need to explicitly write a class around stuff you do?
18:27 shortcircuit Take a look at Java, for example.
18:28 TimToady or do you mean like in Ruby where things that look like functions are really methods of the current class?
18:28 shortcircuit I think both of those descriptions satisfy the concept I was thinking about, at some level.
18:29 TimToady they both have a lying-to-myself feel about them
18:29 TimToady I did enough of that in P5, and am trying to avoid it in P6 :)
18:29 shortcircuit Heh
18:30 shortcircuit There's a Java translation of Niecza, isn't there?
18:30 TimToady I once counted up, and of the 5 or 6 ways a lexer/parser can lie to itself, Perl 5 does 7 or 8 of them...
18:30 TimToady mberends++ is working on a Java emitter, I believe
18:30 TimToady or a JVM emitter, more likely
18:31 TimToady since you can do things at that level that are unnatural acts in Java
18:31 shortcircuit I just got an Android phone yesterday, and I want to try doing a "Hello World on Android in N languages" exercise.
18:32 Hypftier shortcircuit: but back to your question, I don't think so. The major ones, like Java, or C# are not offering it. You usually see it in languages built on the aforementioned, such as Groovy, Boo, Nemerle, ...
18:33 TimToady Java is a bit like an assembly language that way, in that you have to say everything
18:33 shortcircuit We _really_ need to get the SMW stuff working right.
18:33 TimToady "Anything allowed is mandatory." is the usual quote here... :)
18:33 TimToady SMW?
18:34 Hypftier TimToady: Semantic MediaWiki
18:34 shortcircuit If there were a task like the exercise I want to do, it'd be a real shame to see more lines of {{omit from|language}} than actual code. Doing something based off of an SMW query for [[runs on::JVM]] would be much cleaner.
18:34 Hypftier http://rosettacode.org/wiki/S​pecial:Browse/:Category:Java
18:34 fedaykin http://rosettacode.org/wiki/runs_on::JVM (Doesn't exist.)
18:34 fedaykin "Browse wiki - Rosetta Code" http://rldn.net/1308M
18:35 * shortcircuit is semi-afk for a bit
18:36 Hypftier shortcircuit: I might get to you with a larger idea in the next few days, by the way. I just need to think about it a bit
18:36 TimToady though that can be a problem too; early Unix programs were full of #ifdef SysV and such
18:36 TimToady when most of the time you want #ifdef ThisFeatureIsPresent
18:36 Hypftier ... which came with autotools :)
18:37 TimToady which came with metaconfig first :)
18:37 TimToady autotools is a NIH thing :)
18:38 shortcircuit Pretty soon, these two will be arguing about who had the slowest modem.
18:38 TimToady but the features-not-systems idea came even before metaconfig
18:38 Hypftier TimToady: we're there again today with Javascript ;)
18:38 TimToady unless you used Baudot, you can't beat me
18:39 TimToady every generation has to learn that lesson anew, it would seem
18:39 shortcircuit My step-dad had an acoustic coupler, my slowest was a 1200bps, 1200baud. At one point, I had an actual, brand-name Hayes.
18:39 TimToady I started on 110 baud ASR Teletypes :)
18:40 TimToady when the lights would flicker, I'd call up the modem line and whistle at the modem, and if it answered back, I knew the computer was still up...
18:41 TimToady otherwise it was a hike up the hill to make sure it was rebooting...
18:43 TimToady anyway, conditional compilation based on platform is a design smell of non-future-proofness
18:44 TimToady or perhaps of corporate lock-in :)
18:45 shortcircuit TimToady: Ideally, I'd love to see library-by-library, platform-by-platform comparisons, it's just been very difficult to get that side of things rolling.
18:46 shortcircuit The language side takes off as nicely as it does because it's easy to get folks to advocate their languages constructively. It should be so easy to do it for libraries, but the existing static structure makes that difficult to enable.
18:51 TimToady you almost need a way to factor out the language difference, because most folks will concentrate on the surface syntax and not see the point of comparing underlying semantics
18:52 sorear TimToady: mberends is doing JVM for 6model now, not niecz
18:52 TimToady ah, thanks
18:52 TimToady but hopefully it's all convergent :)
18:53 TimToady except for the diversions...
18:54 shortcircuit TimToady: One of the things I love about Rosetta Code is that it helps me learn algorithms and patterns. By seeing what's *common* in the implementations of the same algorithm in very different languages, I get a more direct understanding of the algorithm myself.
18:54 shortcircuit It'd probably be easier to understand algorithms directly if I knew squat past Calc 2, but I don't, really. >.>
18:55 TimToady yes, well, same here
18:56 TimToady at some point the breadth-first people have to yield the road to the depth-first people.
18:57 TimToady the penalty for being a Ronald Reagan sort of person is that sometimes you don't know you don't know, and spout off anyway :)
18:57 shortcircuit Don't I know it. :)
18:57 TimToady :)
18:59 TimToady one can get away with it only by manifesting abject humility from time to time :)
19:01 TimToady I'm fortunate to come from a culture that values vulnerability as a sign of strength, not a sign of weakness.
19:01 shortcircuit Possibly one reason I love hanging out with people I believe to be smarter and/or more knowledgable than me.
19:01 TimToady well, pretty much everyone is smarter than me occasionally :)
19:02 TimToady and occasionally I'm smarter than everyone else :)
19:02 TimToady but usually it's just a smart mouth...
19:02 shortcircuit It scares me how familiar that sounds. :)
19:03 shortcircuit Speaking of familiar...I noticed in the video that you look exactly like the photo on your home page, which I don't think has changed in at least five years.
19:03 TimToady they must have put a blur filter on me :)
19:04 shortcircuit Well, bad autofocus, but the hair and shirt looked the same to me. :)
19:04 TimToady a couple of years ago Damian Conway did a photoshoppy talk in which he claimed I was thousands of years old and never aged
19:04 TimToady I immediately started getting gray hairs :)
19:05 TimToady I just wish my insides were aging as slowly as my outsides.
19:13 shortcircuit TimToady I wouldn't worry too much; you can apparently still jump. :)
19:37 shortcircuit TimToady: I noticed you were speaking Japanese a few times. I take it the presentation was in Japan?
19:38 TimToady Hai!
19:38 shortcircuit What area? Were they badly hit by yesterday's earthquake?
19:38 parsleyfirefly TimToady: Good thing we've been watching Shogun here.
19:38 shortcircuit parsleyfirefly: Well, my otaku background gave me enough to recognize 'Hai'. :P
19:39 parsleyfirefly fiiiine
19:39 sorear what video are you discussing?
19:39 TimToady my talk at YAPC::Asia last year
19:40 TimToady it was mostly to a Perl crowd, so I was much more interested in converting Perl 5 programmers than in converting Python programmers :)
19:41 TimToady it was west of downtown Tokyo, so I assume they aren't in too bad of shape
19:41 TimToady though I very well remember seeing the building that was burning in Odaiba (was never inside it though)
19:50 mwn3d_phone left #rosettacode
19:54 parsleyfirefly left #rosettacode
20:31 mwn3d_phone joined #rosettacode
21:16 TimToady I believe I have have found the only portable solution to the Morse Code problem
21:17 TimToady ;)
21:38 mwn3d_phone Lol cute
21:51 Mathnerd314 joined #rosettacode
22:06 BenBE left #rosettacode
22:25 BenBE joined #rosettacode
22:42 r4m left #rosettacode
22:47 r4m joined #rosettacode
23:04 mwn3d_phone left #rosettacode
23:37 MigoMipo left #rosettacode

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