freenode/#lisp - IRC Chatlog

19:57:45 SpikeMaster hi

19:58:11 SpikeMaster what's more rock solid? SBCL or Clozur CL?

19:58:34 Shinmera Both are pretty solid.

19:58:52 Shinmera On Windows CCL is probably more solid than SBCL

19:59:38 SpikeMaster k, i'm on windows so CCl it is

20:01:39 phoe_krk SBCL gave me problems only when dealing with encodings.

20:01:47 rubengarcia complete noobie to programming can anyone recommend a good book?

20:01:49 phoe_krk And that's the only reason I've switched to CCL when on Windows.

20:02:03 phoe_krk rubengarcia: Gentle Introduction to Symbolic Computation by D. Touretzky

20:02:14 phoe_krk https://www.google.pl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwijyZzux_rKAhVl1XIKHXznDQoQFggfMAA&url=https%3A%2F%2Fwww.cs.cmu.edu%2F~dst%2FLispBook%2Fbook.pdf&usg=AFQjCNHzM32v612ibAe2--S-auWx7GbFDA&sig2=QRT3QA92NvLqX1tZ8graxg

20:02:16 phoe_krk wait

20:02:21 rubengarcia thanks.

20:02:21 phoe_krk https://www.cs.cmu.edu/~dst/LispBook/book.pdf

20:03:14 phoe_krk also, rubengarcia

20:03:35 phoe_krk #clnoobs <- that's a channel just for us Lisp newbs and learners!

20:03:51 rubengarcia thanks for yor help phoe_krk

20:04:03 phoe_krk rubengarcia: no problem! we're here for that.

20:08:50 phoe_krk Does Minion know anything about Lisp books such as the one I just linked above?

20:09:11 phoe_krk I mean, can I ask it "minion: gisc" to have the link to Gentle Introduction to Symbolic Computation pop up?

20:09:28 pavelpenev minion: tell phoe_krk about pcl

20:09:28 minion phoe_krk: direct your attention towards pcl: pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005).

20:09:40 pavelpenev phoe_krk: I guess yes :)

20:09:43 phoe_krk pavelpenev: thanks!

20:09:49 phoe_krk minion: tell phoe_krk about gisc

20:09:49 minion Sorry, I couldn't find anything in the database for ``gisc''.

20:09:55 phoe_krk That needs some polishing though.

20:10:06 jsgrant rubengarcia: Land of Lisp, is pretty fun.

20:10:11 pavelpenev minion: tell phoe_krk about gentle

20:10:12 minion phoe_krk: please see gentle: "Common Lisp: A Gentle Introduction to Symbolic Computation" is a smoother introduction to lisp programming. http://www.cs.cmu.edu/~dst/LispBook/

20:10:16 phoe_krk ...

20:10:18 phoe_krk gentle, I see.

20:10:26 phoe_krk Does minion have a database of what it knows, anywhere?

20:10:35 phoe_krk Publicly viewable?

20:10:40 phoe_krk I'd like to know what I can tell it to do.

20:10:59 jsgrant minion: Is that worth the read? I'm considering doing a learning log, starting from an introductory refresher text.

20:10:59 minion Is that worth the read? I'm considering doing a learning log, starting from an introductory refresher text: An error was encountered in lookup: Parse error:URI "http://www.cliki.net/Is%20that%20worth%20the%20read?%20I'm%20considering%20doing%20a%20learning%20log%2C%20starting%20from%20an%20introductory%20refresher%20text?source" contains illegal character #\? at position 161..

20:11:00 pavelpenev phoe_krk: it appears to be a kind af lisp arcana

20:11:06 Shinmera phoe_krk: https://github.com/stassats/lisp-bots/tree/master/minion is probably it

20:11:18 jsgrant Oh, didn't notice it was a bot. :^I

20:11:33 jsgrant phoe_krk: ^

20:11:35 PuercoPop try /msg minion help

20:11:38 phoe_krk I did!

20:12:05 jsgrant * probably needs a nap, but it's too late at this point. :^P

20:12:21 jsgrant minion: (+ 1 1) ?

20:12:21 minion please stop playing with me... i am not a toy

20:13:02 phoe_krk ,eval (+ 1 1)

20:13:07 phoe_krk sigh~

20:13:41 pavelpenev phoe_krk: evaling lisp code is rather hard to secure hence why it isn't available

20:13:52 phoe_krk pavelpenev: yes it is

20:14:03 phoe_krk over at #lisp-pl we have ood which has a custom shadow

20:14:19 jsgrant Is there a "go to" irc lib, for CL?

20:14:25 phoe_krk even through that shadow I was able to poke a hole through that and eval whatever I'd like, so. :P

20:19:24 kpoi What Common Lisp implementations do you guys use? I have a project that I want to make portable to the most common ones, right now it only works on SBCL, any tips?

20:19:42 PuercoPop I wonder why some people write (in-package #:cl-user) (defpackage ..) instead of (cl:defpackage) ..

20:19:49 jsgrant I use SBCL, I at somepoint want to move to SICL.

20:19:51 Shinmera kpoi: CCL and ECL are probably the next targets.

20:19:56 phoe_krk ^

20:20:19 Shinmera kpoi: followed by some order of ABCL, CLISP, AllegroCL, LispWorks

20:20:45 phoe_krk kpoi: what's not portable in your code, anyway?

20:20:54 phoe_krk maybe it can be aided through the use of some of the portability libraries

20:21:15 kpoi I'm not sure yet I haven't tested anywhere else but SBCL. It might work fine on all of them...

20:22:30 kpoi It doesn't do anything particularly menacing, it just interprets brainfuck.

20:22:36 phoe_krk kpoi: uhhh.

20:22:44 phoe_krk Are you using *anything* outside ANSI CL standard?

20:22:51 kpoi No.

20:22:54 phoe_krk Then it's portable.

20:23:01 phoe_krk * applauds

20:23:22 phoe_krk Congratulations!

20:23:23 kpoi I don't even need to test?

20:23:28 trinitr01 ** NICK trinitr0n

20:23:30 phoe_krk It's always worth to give it a test or two.

20:23:31 phoe_krk But.

20:23:36 phoe_krk If your code is ANSI CL-compliant.

20:23:43 phoe_krk And it doesn't work on an ANSI CL-conforming implementation.

20:23:54 phoe_krk Then it's not the problem of your code and the problem of the implementation.

20:24:22 Shinmera Note that some things in the standard are implementation-dependant, so it's always worthwhile to test anyway.

20:24:34 Shinmera (or even worse, unspecified)

20:24:35 PuercoPop phoe_krk: I am aware that portable is an overloaded term, but that is not enough for code to be portable http://www.lispworks.com/documentation/HyperSpec/Body/26_glo_p.htm#portable

20:24:36 kpoi That is what I was worried about^^^

20:25:16 phoe_krk PuercoPop: well. right.

20:25:27 phoe_krk and not relying on implementation-dependent things.

20:25:53 phoe_krk like (eq 2 2) evaling to T, which was something that gave me a moment of pure WTF a moment ago.

20:26:47 kpoi I don't use EQ.

20:26:59 phoe_krk Then, basically.

20:27:11 phoe_krk Scan the functions you use for whether they anyhow utilize implementation-dependent behaviour.

20:27:18 phoe_krk If not, that should be enough.

20:27:35 phoe_krk PuercoPop might correct me if I'm mistaken again. :P

20:28:26 kpoi Yeah. Well I'm off to install some virtual machines, Tchau!

20:28:39 kpoi How do I do the /away thing?

20:29:03 kpoi it says "You are no longer marked as being away" every time, but I WANT to be marked as away

20:29:14 pareidolia Use it with a message?

20:36:43 PuercoPop phoe_krk: I was trying to come up w/o a good example, that didn't seemed contrived but have fail to come up or remember one. However more than implementation-dependent it is a constraint upon relying only on documented behaviour. For example, setf literals is not implementation-dependent but undefined. Sorry I couldn't come up w/ a more illustrative example

20:37:19 phoe_krk Oh, undefined behaviour is never a good idea.

20:37:45 kpoi Pathnames

20:40:40 phoe_krk Pathnames are even more than implementation-dependent: they're OS-dependent.

20:40:56 Shinmera well

20:41:07 Shinmera even fixnums are os-dependant

20:41:51 Shinmera or rather, system-dependant.

20:42:21 Shinmera Both of which boil back down to implementation-dependant, since the implementation could try to bridge things over as much or little as it wants.

20:42:28 phoe_krk Well. Right/

20:42:29 phoe_krk .

20:47:07 Guest59381 ** NICK HDurer

20:49:48 kpoi I used ASDF. Maybe that would be an issue aswell

20:50:05 Shinmera Nah

20:50:09 phoe_krk ASDF seems fairly implementation-independent, actually.

20:50:13 phoe_krk Sturdy as hell.

20:50:25 Shinmera ASDF is probably the most portable code out there, as long as you only rely on features from ASDF2.

20:50:36 Shinmera Since some implementations still haven't upgraded to 3 :(

20:51:02 Shinmera Either way, it's the standard build system and as such well tested on all implementations.

20:51:59 kpoi Any feedback would be appreciated, it is my first "Open source" project: https://github.com/equwal/LispFuck

20:54:11 phoe_krk Hum. You might want to put an optional reader macro there so you can go like #F>>>><

20:54:13 phoe_krk until the end of line.

20:54:30 phoe_krk also, hahahaha! you actually did what I wanted to suggest you :D

20:54:39 kpoi ?

20:54:40 phoe_krk (brain:fuck string)

20:54:51 kpoi Oh yeah I thought that was amazing.

20:55:11 phoe_krk it is, at least to me.

20:55:48 kpoi I have never heard of a reader macro, I just finished reading Peter Siebel's book and he said to come here for help. Why would I want to go like "#F>>>><"?

20:56:06 kpoi Are the reader macros the #+ and the #- things?

20:56:23 phoe_krk Yes yes.

20:56:25 phoe_krk I mean.

20:56:28 Shinmera Reader macros allow you to modify how the text is read, so they allow syntax modification.

20:56:35 phoe_krk Instead of going (fuck "asdfasfsfsdfasfdafsdaf") in, let's say, REPL.

20:56:43 phoe_krk You can go #Fasdfasfsfsdfasfdafsdaf

20:57:01 phoe_krk and it'll work the same way, although giving you a less Lispy and more Brainfucky feel I think.

20:57:19 kpoi Oh so another reader macro that is built in would be #b100101

20:57:35 kpoi That is neat I'll do that.

20:57:54 Shinmera Note that ( is a reader macro too

20:58:06 Shinmera as is "

20:58:18 kpoi Also for the record "asdfasfsfsdfasfdafsdaf" is just a brainfuck comment.

20:58:22 phoe_krk yes, I know :D

20:58:28 kpoi I thought you did yeah.

20:58:32 phoe_krk so you can as well go #FUCK

20:58:45 phoe_krk and have it evaluate to whatever an empty line evaluates to in your reader

20:59:27 phoe_krk after using https://github.com/nvbn/thefuck I would consider this to be a very fine addition to an average REPL.

21:00:18 kpoi That is hilarious

21:00:48 kpoi "Magnificent app which..."

21:01:24 phoe_krk kpoi: it IS magnificent.

21:02:27 kpoi They should have a shorter version like "f" so you only have to type one letter

21:02:50 kpoi Of course that would be easy to do yourself

21:04:06 phoe_krk uh

21:04:06 kpoi Like this guy https://github.com/nvbn/thefuck/issues/453

21:04:12 phoe_krk kpoi: you can customize it of course

21:04:20 phoe_krk alias f fuck

21:04:21 phoe_krk in bash

21:05:10 kpoi Yeah just did.

21:17:03 rpg Does anyone know about QL internals? How does it do introspection? Does it have to load everything up, or does it do some kind of heuristic parsing of libraries' asd files?

21:18:52 Bike i think it tries to load and puts a handler on the system not found condition.

21:19:46 Bike ah, no, it's a bit more complicated check ql:autoload-system-and-dependencies, it uses asdf to compute the load strategy and then does the handler bit.

21:26:36 rpg Bike: Thanks. We were arguing over whether QL did some sort of groveling where it needed to be able to load or read a system definition /without/ actually loading anything.

21:27:08 rpg ASDF is not (cannot be) pure, so if you load an ASD file, you may change the image by loading systems needed to process the defsystem form.

21:28:02 rpg We have been trying to figure out how QL and ASDF interact, with an eye towards fixing some ASDF stuff.

21:38:47 N3vYn /reconnect

21:40:40 PuercoPop rpg: I was going to ask, is the purpose of asdf-user to put all the system definitions in that package?

21:41:25 rpg PuercoPop: Yes, that was fare's intention. I am old school: I have an emacs template for my ASD files and just make a new package for each one.

21:42:05 rpg PuercoPop: Fare didn't like the old proliferation of anonymous packages. I'm not entirely sure why, since they provide namespace hygiene.

21:42:13 rpg you should ask him.

21:42:20 rpg possibly resource issue

21:43:36 PuercoPop rpg: thanks

21:45:29 fiddlerwoaroof rpg: anonymous package?

21:45:49 fiddlerwoaroof Is it possible to make a package without a name?

21:45:49 rpg ASDF used to create a new package for each system, and load the system definition file into that package.

21:46:03 rpg Not really anonymous -- had gensyms for names.

21:46:40 fiddlerwoaroof Ah,

21:48:29 kpoi That makes sense.

21:51:12 kpoi So they did it like this: (make-package (gensym))

21:51:46 rpg I think it was more like (gensym "ASDF") but you get the picture.

21:52:11 kpoi I never thought about doing something like that.

21:52:14 kpoi neat

21:52:52 kpoi Say, how do you get out of a package that doesn't inherit COMMON-LISP? I am trapped in my brand new G1334 package

21:53:17 mordocai (cl:in-package) ?

21:53:40 kpoi Right.

21:59:42 phoe_krk kpoi: I did that once

22:01:38 rpg What's the best source for anaphoric conditional macro?

22:01:49 phoe_krk Source?

22:04:30 Fare rpg: they failed to provide hygiene. They just made symbols unreachable when you needed them

22:05:01 rpg Fare: I thought it was something like that, but I forgot the rationale.

22:05:20 rpg phoe_krk: library

22:05:39 PuercoPop rpg: you can replace aif with alexandria:let-if, the result will likely be mroe readable

22:06:11 axion IF-LET

22:06:43 Fare Shinmera, actually since last year it looks all implementations have ASDF 3.

22:07:06 rpg axion, PuercoPop: Thanks! didn't realize that there was anything because of the declaration against anaphoric macros.

22:07:18 axion i like the conditional LETs much

22:07:19 Fare except that clisp is stuck with an antique prerelease and isn't maintained anymore

22:10:08 Fare in practice, with the temporary ASDF~d packages, you still had to create your own package for anything serious, so all that was left of the pseudo-hygiene was a non-standard thing that made some symbols inaccessible. All the disadvantages, none of the advantages.

22:10:10 rpg bye for now.

22:10:14 Fare bye!

22:32:57 brunov0id ** NICK d3lf0

22:35:25 ASau` ** NICK ASau

22:46:04 kpoi Fare: Clisp has a way of doing it's own thing.

22:47:21 Fare gcl is worse

22:53:18 rpg Fare: if you are still around, is there an easy way to run all the lisps on a single test (instead of all the tests on a single lisp)?

22:56:09 Fare maybe make test-all-scripts t=test-name.script

23:00:20 rpg OK. I'll have a look. BTW, I'm trying to do this in lisp, with WITH-ALL-LISPS, and it seems like the GET-LISPS doesn't use the envar by default....

23:01:20 rpg I see INITIALIZE-ENVIRONMENT

23:46:24 rpg Quick SLIME question: should I expect SLIME to properly indent :METHOD inside DEFGENERIC?

23:46:30 rpg not working for me...

23:50:19 Bike i think you might need a special indentation contrib.

23:51:44 phoe_krk rpg: silly question, but are you including slime-indentation inside your slime-setup?

23:51:52 phoe_krk this fixed quite a lot of indentation issues for me.

23:52:03 rpg phoe_krk: do I need to do that in addition to slime-fancy?

23:52:25 phoe_krk rpg: I believe you do.

23:52:45 phoe_krk (slime-setup '(slime-fancy slime-asdf slime-mrepl slime-indentation inferior-slime))

23:52:49 phoe_krk This is my piece of init.el

23:53:00 phoe_krk And it did fix a few things for me.

23:53:04 Bike inferior superior lisp interaction mode?

23:53:28 phoe_krk https://github.com/slime/slime/blob/master/contrib/inferior-slime.el

23:53:33 rpg d'oh! slime-indentation is indeed, *not* included in slime-fancy.

23:53:44 phoe_krk Inferior SLIME mode: The Inferior Superior Lisp Mode for Emacs. indeed.

23:53:52 rpg oh. wait. I'm actually including slime-indentation.

23:54:02 phoe_krk oh.

23:54:06 phoe_krk then that's weird for me.

23:54:14 rpg slime-fancy -indentation, -tramp, and -asdf

23:54:26 PuercoPop luis: you should consider adding slime-indentation to slime-fancy?

23:54:31 phoe_krk Then you *might* need a custom indent of sorts.

23:54:44 rpg so what I see is that :METHOD forms have the first form of the body lined up under the argument list, instead of "outdented"

23:55:10 phoe_krk I don't know how to make custom indentation rules, I'm afraid.

23:56:22 rpg I may be inadvertently crushing it....

23:56:27 rpg I'll restart and check, thanks!

23:56:40 rpg I have old emacs config code from before slime-indentation.

23:56:58 rpg no, take it back -- that's not loaded.

23:57:52 rpg only loaded when using Franz common lisp mode. Hm.

0:01:17 rpg phoe_krk: looks like it's defined -- would you mind testing it? It's definitely not working for me, and slime-indentation is in features....

0:05:19 phoe_krk rpg: I'm dying to sleep

0:05:21 phoe_krk but gimme

0:05:26 rpg agh. Gottta go anyway.

0:05:29 phoe_krk rpg: tomorrow

0:05:30 phoe_krk cya

0:05:34 rpg Catch you tomorrow! Thanks!

0:28:23 kpoi (#FUCK

0:28:35 kpoi Oh buffer

0:28:38 kpoi Wrong buffer

0:32:22 phoe_krk XD

0:32:35 phoe_krk kpoi: you're doing what I think you're doing, aren't you

0:39:36 kpoi Yes

0:40:11 kpoi I didn't know you could make your own reader macros so I am reading about them in Let Over Lambda

0:40:23 kpoi PCL doesn't mention your ability to do that until the LAST CHAPTER

0:40:33 phoe_krk :D

0:40:53 pillton kpoi: Why is this outrageous?

0:40:54 kpoi He (paraphrasing) writes "I haven't shown you how to parse XML"

0:41:05 kpoi Because they seem like a useful thing

0:41:34 phoe_krk well

0:41:37 phoe_krk they are useful

0:42:05 kpoi phoe_krk: most certainly

0:43:13 phoe_krk torlakur: wb

0:43:14 pillton I don't think I have ever written one.

0:43:49 pillton I'm not saying they aren't useful, but I can understand why they are mentioned in the last chapter.

0:43:57 phoe_krk pillton: I have, and I plan on submitting a quicklisp packet based upon them.

0:44:19 phoe_krk basically they interfere with Lisp's usual syntax which is, to put it lightly, a generalized lack of syntax. :P

0:44:32 kpoi pillton: They aren't really mentioned, he just alludes to their existence with the XML quote I sent a minute ago

0:44:53 phoe_krk there's an example online on how to parse JSON with a reader macro on #\[

0:45:04 kpoi phoe_krk: I saw that one.

0:45:20 kpoi First thing on google

0:45:32 phoe_krk yes

0:45:52 kpoi That is a really great example, he turns the REPL into a full blown JSON parser.

0:46:05 pillton To do what?

0:46:23 phoe_krk kpoi: that's how you can essentially write a compiler with reader macros.

0:48:20 kpoi You type something like

0:48:21 kpoi > [1, 2, 3] ;This is some JSON markup

0:48:21 kpoi in the repl and it gets converted right into lisp:

0:48:21 kpoi ;I don't think this is exactly how it worked, but you could make it work like this if not.

0:48:24 kpoi He also had it for the Object syntax curly braces

0:48:27 kpoi > { null: true}

0:48:30 kpoi

0:48:33 kpoi

0:48:39 kpoi damn I wasn't done with that

0:49:40 Fare kpoi: you can also use reader-interception...

0:49:40 kpoi Anyway type something like

0:49:40 kpoi > [1, 2, 3]

0:49:40 kpoi (1 2 3)

0:49:40 kpoi > {null: true}

0:49:43 kpoi (NIL (T))

0:49:46 kpoi And that is similar to what he did.

0:50:10 kpoi What is that

0:50:12 kpoi ?

0:50:20 kpoi Fare:

0:52:05 Fare http://cliki.net/reader-interception

0:52:16 learning does lisp have support for audio in the standard library or do i need to use a third party lib?

0:52:46 learning im searching the docs but im not finding anything

0:53:23 Bike not in the standard.

0:53:35 Bike i think you can use mixalot.

0:55:17 learning lame, i'll have to write my own then it looks like

0:55:24 learning i want to do really low level stuff

0:56:11 learning bookmarked mixalot

0:57:58 learning if anyone else has worked with audio and has an open source library to reccomend please do

1:00:18 PuercoPop learning: not my lib, but there is this WIP https://github.com/rick-monster/cl-alsaseq

1:01:06 learning bookmarked. thank you

1:33:15 Fare learning: there's mixalot if you don't care too much about timing, or incudine if you do

1:33:48 Fare I found that with mixalot I get a lot of jitter. But I'm probably not using it well.

1:34:20 Fare for really low-level stuff, try incudine

1:34:48 Fare or then again, outside cl, there's supercollider or extempore. Or in clojure, overtone or klangmeister

1:34:59 Fare learning: are you still there?

1:35:36 Fare racket has stuff, too

1:37:57 kpoi You could use FFI

1:38:28 kpoi or UFFI

1:38:58 edgar-rft There's also cl-portaudio: <http://filonenko-mikhail.github.io/cl-portaudio/>

1:39:24 kpoi Probably better not to use FFI/UFFI, there is plenty of native code.

1:56:40 learning ye

1:59:53 learning incudine is what im looking for i think

3:33:58 drmeister Is there anyone who uses LaTeX and includes Common Lisp source code snippets?

3:35:01 drmeister What do you use to include source code? verbatim mode looks ok but I haven't figured out how to get captions or labels/refs.

3:35:17 drmeister I tried the {listings} package - but that makes Common Lisp look like dog food.

3:54:50 Bike that is probably only beach.

3:55:19 Bike i used listings, but that was for python, and i still had to fix some shit.

3:56:51 Bike it looks like listings only professes to support autolisp. is that why it's dog food?

4:28:10 beach Good morning everyone!

4:28:36 blub hi beach

4:30:05 beach drmeister: Look in the SICL Specification directory. Look for \inputcode in TeX files and look at the shell script named codify.

4:31:08 Opodeldoc beach: I shot you an email couple or few hours back. Wasn't sure which to use so I sent it to the one in the paper

4:31:44 beach Might not work. Try robert.strandh@gmail.com.

4:35:35 Opodeldoc beach: resent \o/

4:36:05 beach Got it! Thanks!

4:36:30 beach I'll get to it as soon as I am a bit more awake.

4:36:58 Opodeldoc beach: What seemed actually important e.g. typos are marked with =>

4:37:12 Opodeldoc The rest is largely overcaffeinated babbling :)

4:37:24 beach OK. Good to know. :)

4:39:28 beach Opodeldoc: Some of your remarks I'll have to run by my wife, who has a PhD in English and who is a member of the IBM style and word-usage council.

4:39:49 Opodeldoc beach: Sounds good

4:40:12 Opodeldoc beach: Are you going to wind up using this for climacs?

4:40:23 beach Yes, for Second Climacs.

4:41:11 Opodeldoc Hell yes.

4:41:33 beach For (First) Climacs, it is more complicated, because it used Drei which is part of McCLIM, so I would have to make sure I don't break input editing in McCLIM if I were to use Cluffer there.

4:53:32 beach I take it you approve of putting Cluffer into Second Climacs, yes?

4:53:46 Opodeldoc beach: I never got my head around clim-anything tbh. I'd just love to see some syntax hilighting that isn't just blind matching of characters :)

4:54:59 beach OK. (First) Climacs already does a better job than Emacs in that respect, but the technique used is wrong and unmaintainable. For Second Climacs I have a much better idea of how to do it.

4:59:29 beach For Second Climacs, I also plan to have a better isolation between the model and the user interface, so that I can put a CLIM II GUI on it now, and a CLIM3/CLIMatis interface later when it is available.

5:03:39 Opodeldoc I may have to take a try at getting climacs running :)

5:05:33 beach (First) Climacs seems to work out of the box. Second Climacs is not ready yet.

5:05:59 beach It should be available through Quicklisp.

5:16:01 fiddlerwoaroof I just quickloaded (first) climacs and it seemed to work fine.

5:16:37 Quadrescence fiddlerwoaroof type (climacs:climacs) !!

5:16:39 beach Great! Thanks.

5:16:55 fiddlerwoaroof Quadrescence: I did

5:17:46 fiddlerwoaroof Although, when I hit C-x f in a blank buffer, it enters the debugger

5:18:14 beach :(

5:19:12 beach I confirm. :(

5:19:46 fiddlerwoaroof http://paste.lisp.org/+6L67

5:20:10 beach Yes, I can reproduce it here.

5:20:10 fiddlerwoaroof I was also just checking out gsharp via quickload

5:20:44 fiddlerwoaroof That didn't want to work at all.

5:20:58 beach I am so sorry to hear that.

5:21:15 beach That's another one I wanted to make a second version of.

5:21:57 beach But my remaining life expectancy is only around 20 years, and I am not sure I will have time to do all this.

5:22:51 beach ... which is why I need an apprentice or 10.

5:23:49 Opodeldoc beach: I can't for the life of me find a reference, but I'm pretty certain both nvi and vim also just use a gap buffer

5:24:02 Zhivago On the positive side, it will probably all be irrelevant in 20 years in any case. :

5:24:24 beach Opodeldoc: Entirely possible, yes.

5:25:30 beach Zhivago: Thanks for the encouragement.

5:25:47 Zhivago No problem.

5:26:03 beach * takes a break.

5:26:15 Opodeldoc It seems the vim author is also behind http://www.zimbu.org/Home

5:26:29 Opodeldoc I guess if you don't just use lisp you have to reinvent everything...

5:42:45 opal ___

5:42:46 opal < λ >

5:42:47 opal ---

5:42:48 opal \ __

5:42:49 opal \ (oo)

5:42:50 opal \ ( )

5:42:51 opal \ /--\

5:42:52 opal __ / \ \

5:42:53 opal UooU\.'@@@@@@`.\ )

5:42:54 opal \__/(@@@@@@@@@@) /

5:42:55 opal (@@@@@@@@)((

5:42:56 opal `YY~~~~YY' \\

5:42:57 opal || || >>

5:43:02 opal that is all

5:55:45 moore33 beach: Around?

5:58:48 Opodeldoc Was ~half hour ago

6:01:43 beach moore33: Yes.

6:02:58 moore33 beach: Good morning! I'm reading http://www.infoq.com/articles/no-more-mvc-frameworks, which you might find interesting with your CLIM hat on.

6:04:08 moore33 A lot of JavaScript / Web jargon, but that's not too hard to navigate.

6:04:11 beach Looks interesting, yes.

6:05:01 beach I have been against explicit view updates with (say) observers for a very long time.

6:08:57 beach moore33: It's a bit long. I need to read it a little at a time.

6:09:22 emaczen What runtime tools exist for viewing/querying the last called methods?

6:09:23 minion emaczen, memo from pjb: yes, there's at least one OSX app in the App Store written in CCL: 'Clozure CL.app'.

6:09:23 minion emaczen, memo from pjb: you need to write a wrapper Objective-C class with a lisp slot refering to your lisp object.

6:10:42 Opodeldoc I'm given to understand the smalltalk folks are starting to frown on MVC too

6:15:36 moore33 beach: yeah, it's long. Good bus / bathroom reading.

6:15:54 Opodeldoc "you don’t necessarily get this warm feeling that, even at Google, people know what they are doing:" haha

6:16:42 moore33 Opodeldoc: yes, it's quite funny.

6:17:07 Opodeldoc "The core package of Angular 2 alone has 180 semantics," my god is that for real

6:17:16 Opodeldoc Good day to not be a web dev sheesh

6:18:31 blub not a semantic too many

6:19:56 moore33 Off to walk Zep.

6:22:46 beach Recently, from reading articles and now this web page, I have detected a change in the way "API" is used.

6:23:28 beach It now appears to mean "The means by which a web server communicates with the client web browser".

6:27:32 smokeink when it comes to web dev, api usually means the backend that responds with 'json data'

6:28:02 beach I see.

6:29:46 beach No wonder I had a hard time understanding those articles I read, given the definition in my mind of "API".

6:32:36 Opodeldoc It's safe to assume anything web means something different, as they're reinventing bloody everything.

6:32:48 smokeink haha

6:33:19 beach I am beginning to understand that.

6:34:38 Opodeldoc I still read "sandbox" as "half-assed version of process isolation"

6:57:00 fiddlerwoaroof beach: by "explicit view updates" do you mean that the view should not be able to request an update?

7:02:30 fiddlerwoaroof I was just reading an article by a smalltalk guy about how everyone gets MVC wrong.

7:02:32 fiddlerwoaroof http://blog.metaobject.com/2015/04/model-widget-controller-mwc-aka-apple.html

7:03:04 fiddlerwoaroof Anyways, I'm off.

7:03:44 Opodeldoc Ask them how they feel about c++ "objects" :]

7:08:39 roscoe_t` ** NICK roscoe_tw

7:09:22 Opodeldoc "Right now we write UIs by poking at them, manually mutating their properties when something changes, adding and removing views, etc. This is fragile and error-prone. Some tools exist to lessen the pain, but they can only go so far. UIs are big, messy, mutable, stateful bags of sadness."

7:16:06 emaczen What are you guys using for UIs?

7:18:54 z0d curses or web mainly

7:24:47 emaczen z0d: What is curses?

7:25:02 Opodeldoc roguelike

7:25:32 Opodeldoc https://en.wikipedia.org/wiki/Ncurses

7:25:56 Opodeldoc curses is part of posix standard, the actual implementation tends to be ncurses

7:27:30 emaczen interesting -- now I know what the GUI is made from (or somethign similar) when my OS crashes lol

7:28:16 z0d curses is terminal based.

7:28:30 Opodeldoc lynx, elinks, w3m (browsers), or games like nethack use curses

7:28:36 z0d yeah

7:28:44 z0d or Midnight Commander

7:29:04 z0d heck, even 'emacs -nw'

7:29:30 Opodeldoc duh! of course. vi and emacs and nano and pico and mutt and.....

7:30:25 Opodeldoc I think freebsd's installer is curses

7:31:12 Opodeldoc I knew there was an obvious example. I can't believe I didn't think of editors hah

7:32:29 Opodeldoc screen and tmux probably use the hell out of curses

7:46:02 p_l I've recently seen a rare nowadays beast - a graphical dashboard in curses

7:46:13 p_l which graphed system stats

7:47:25 myrkraverk Oh, that's interesting. People did a lot of cool things with softfonts in real terminals too.

7:47:58 myrkraverk Anyway, I am wondering, the SBCL garbage collector is essentially single threaded, right?

7:48:29 myrkraverk So, if one thread holds a reference to something, another thread will not garbage collect it? Is what I mean to ask.

7:48:48 p_l myrkraverk: no, SBCL stops all threads for GC

7:48:53 p_l i.e. "Stop the World"

7:49:46 p_l then, a single-threaded GC happens, in this case single-threaded means it marks and collects from a signle thread

7:50:35 myrkraverk I am going to use a CFFI binding in multithreaded envirnmont, and am wondering what (if anything) I need to make thread safe on the C level. (I have already reviewed the C++ code I'm binding)

7:52:10 Opodeldoc p_l like htop or something else?

7:52:28 p_l Opodeldoc: no, htop does simple horizontal bars - I mean things like line charts :)

7:52:34 Opodeldoc ooo

7:52:47 Opodeldoc mrtg type stuff

7:53:08 p_l and that's on a limited ANSI terminal, unlike on advanced stuff in XTerm (which does vector graphics if you ask it) or one of the crazier DEC VT-series with programmable fonts

7:55:05 myrkraverk * pats his crazy VT with programmable fonts.

7:56:31 p_l https://raw.githubusercontent.com/algernon/potential-happiness/master/docs/screenshots/screenshot-2.png