libera/commonlisp - IRC Chatlog
Search
11:04:39
jackdaniel
pve: I've developed a personal distaste for the phrase "I am not a lawyer" - it is too often used as a hand-wavy argument to not do something. (I'm not saying that you've used it that way)
11:05:07
jackdaniel
generally there is a very vocal anti-copyleft group of people on the internet forums
11:06:08
aeth
The MIT license seemed more common in CL than the BSD license when I last checked, which was ages ago. Maybe things have changed in recent years.
11:07:09
aeth
The confusing part with licenses is the middle ground ones like the LGPL/LLGPL. Everyone knows that the GPL means, or is supposed to, anyway.
11:08:34
jackdaniel
lgpl is like putting a cookiejar in public and saying: don't take it, claim ownership and resell at "fAiR MArkeT PrICe" :)
11:09:32
aeth
LGPL is saying "package it in a specific way so a user can upgrade the library in-place without relying on you, except it's worded in a C-specific way, except we say it applies to all languages"
11:12:44
beach
I wouldn't say "C specific", but as I recall, there are phrases that suggest traditional languages with traditional separate compilation and transitional linking technology.
11:14:40
jackdaniel
no, let over lambda is a fun set of experiments with closures, on lisp is a book that discusses interesting and very useful common lisp techniques
11:14:42
beach
That, and On Lisp is certainly good enough to learn some very sophisticated macro use.
11:16:28
jackdaniel
regarding wording in lgpl, /I think/ that there was a clarification from fsf after someone (not lisp related, but some other non-c-like language) asked about the phrasing around linking etc
11:16:46
jackdaniel
but I don't feel like digging that up in the internet, so I'll leave it as a hearesay
11:17:24
jackdaniel
(generally - technical details aside, it applies the same to libraries where "linking" does not a C sense)
11:17:57
aeth
you just have to know that 'java' is the magic word when this conversation comes up again in 2 years
11:55:35
pve
jackdaniel: Ok, I just meant that to a layman such as myself the argument I quoted seems sound.
12:14:08
lispm
Probably gonna be off-topic here. I am not a channel regular. But just wanted to alert here that there is a certain lisp123 here who pretends to be multiple people on different forums (HN, Reddit, perhaps here too) and self-promote their links related to ashok-khanna while pretending to be someone else. Case in point:
12:14:09
lispm
https://hn.algolia.com/?dateRange=all&page=0&prefix=false&query=ashok%20khanna&sort=byDate&type=all
12:18:40
jackdaniel
(since they did not break any rules on this channel it just went as something people acknowledged - lisp123 sometimes attends this channel)
12:19:17
phoe
lispm: the channel logs say https://irclog.tymoon.eu/libera/%23commonlisp?around=1658750032#1658750032
12:22:29
lispm
Ah! I see you guys have misaken me for Rainer. I am not. :-) Unfortunate choice of irc nick. Will drop this nick so that it is available for the real Rainer.
12:27:32
jackdaniel
in your code you are yet to disclose or with promoting your own work under different names?
12:28:54
Shinmera
with hugely unique names like "lispm" and "lisp123" it's impossible for there to be different people with the same name
14:07:11
prxq
how is it with slime nowadays. Still the best to use the bleeding edge (ie master in the repo)?
14:26:22
nij-
While loading an uncompiled system, the files gets read as piles of strings. I wonder if those strings are retained somewhere in the system? I want to compute the hash of it to identify its 'version'.
14:30:48
beach
nij-: What I do for SICL, though, is I first read the entire file into a vector of lines, where each line is a string. Then I have the compiler read from a Gray stream that traverses that vector. That way, source position in compiled code is indicated as a triple: vector, line number, column. And the position of an expression as a pair of those (beginning and end).
14:44:12
jackdaniel
here it gives me variable |CHAR-#| is unbound because it gets confused about a missing colon
14:53:08
mfiano
This reminds me of ruricolist/cmd, a wrapper over uiop:*-program, with keyword symbols for Unix pipeline operators like :>, :>>, and my favorite: :|\||
15:34:04
pjb
jackdaniel: the problem isnot the REPL, but emacs that is not able to parse correctly Common Lisp. Even worse when it has to try to parse backward.
15:57:35
mfiano
I think Common Lisp as a community really shined brightly this year. It seems like there were lots more new exciting developments and collaborative efforts than I ever remember previously anyway. In any case, congrats, everyone!
15:59:21
mfiano
No, because he also allows using the SYMBOL-NAME when it is annoying to do otherwise, so "|" works.
15:59:34
NotThatRPG
Certainly the idea of having something special for those is neat. Are they just translated and shoved through to the shell, or are they unpacked to operators on the streams produced by run-program?
15:59:55
mfiano
I am not the author. I didn't particularly care for using that library when I tried though. I ended up re-implementing it :/
16:00:59
NotThatRPG
Invoking sub-shells is frowned upon, but I find trying to do anything with "naked" run-program is miserable
16:29:42
beach
mfiano: Well, as it turns out, I am working on Second Climacs a bit now. Today, I implemented the EXCHANGE-EXPRESSIONS command, sort of like C-M-t in Emacs. And it should work even if there are invocations of strange (even user-defined) reader macros in the buffer.
16:32:02
beach
As it turns out, scymtym is making great progress on higher-level analyses of the buffer contents and I felt I was holding everything back by not working on Second Climacs.
16:35:25
mfiano
Yes, you had tried to hire me to help on Climacs last year, when I was busy. Ironically, now I'm looking for work.
16:45:38
mfiano
In any case, I still think my previous statement about this year holds true. Congrats everyone.
16:51:12
Shinmera
if the sales of Kandria are going to be as I expect them to be it'll be a bad start :)