freenode/#lisp - IRC Chatlog
Search
20:06:22
makomo
pjb: i think you linked it a few days ago. i took a quick look but can't remember much. i'll check it out :-)
21:02:46
drmeister
I wonder if anyone has insight into an issue I'm seeing with docker and quicklisp - I'm trying to do something unusual.
21:03:25
drmeister
I've got a docker image (docker hub: drmeister/cando) that I'm mounting a ~/.cache and ~/quicklisp/local-projects directory into.
21:04:18
drmeister
These directories are on the host and mounted into the corresponding directories in the docker image.≈
21:05:13
drmeister
Everytime I start the docker container - quicklisp recompiles the systems from the ~/quicklisp/local-projects directory
21:05:47
drmeister
I'm trying to also mount the .cache directory - so that it can use the cached fasls - but that is not working.
21:06:21
drmeister
quicklisp/asdf use time stamps of the source files and fasls to decide if it should recompile something - correct?
21:07:15
drmeister
Hmm - the docker image uses UTC time by default. Maybe it doesn't like that the compiled fasls look like they are from a couple of hours in the future.
21:07:47
drmeister
But - fasls compiled in the future are always better than fasls compiled in the past. Hmmph
21:10:02
akkad
drmeister: you can fix that with updating the /etc/localtime and other /etc/ timezone items to match the hosts
21:11:23
akkad
that way you can avoid installing time related packages, and playing the time difference gam
23:41:52
aeth
What's the best way to deal with "multiple return value" C functions where you pass in pointers that it sets? I'm thinking static-vectors:with-static-vector and then pass in pointers to each element of that array.
23:43:52
Bike
assuming the function just uses the pointers for returning values, i'd just stack allocate, yeah
2:06:07
smokeink
circle is a class, shape is the instance object of it. shape refers to the object you pass to the function as the 1st param
2:41:47
on_ion
try to search git for a page of the different backends, afaik there is also descibed how to use them
2:45:44
Xach
I don't think beagle is still a going concern. (But I am not much of an authority in this matter.)
7:23:54
xificurC
makomo: I just read what you wrote me, thanks. I enjoyed reading let over lambda despite the code-walking issues and elitism. The sorting network chapter was pretty cool. It's also interesting to see how closures can be the building block of OO and how one can create ad-hoc protocols from it
7:26:05
xificurC
someone else (I never remember who) mentioned he doesn't like that (rephrasing) the values in closures cannot be changed, which is very non-lispy
7:28:58
xificurC
"Several remarks that might seem offensive are actually just phrasing mistakes on my part"
7:29:52
xificurC
the terminology throughout the book really sounded that way to me. All the right vs wrong, lisp vs. blubs etc
7:30:52
schweers
people often complain that lispers are elists and look down on other languages. But really, this is to be expected, given the current state of our field :/
7:31:54
beach
I still don't see how that is "elitist" though. But never mind. I am in a rotten mood, so I should probably be quiet.
7:32:04
xificurC
schweers: I agree and I sound like that to some of my colleagues too. Still it's something to look out for. Noone likes to hear their choices are "wrong" or downright "stupid" and I don't feel competent to judge anyone
7:33:11
xificurC
beach: sorry if the term offended you in any way. It was a subjective feeling not to be confused with what the author actually meant.
7:33:53
schweers
I also wonder ... is being elitist wrong generally? I feed that it can be justified
7:35:24
xificurC
I would say - knowing you are better at something is ok. Being a jerk about is is not.
7:35:26
beach
xificurC: I have often been accused of being elitist, which is probably why I reacted that way. But the people who accuse me of that say that because I happen to think it is a good idea for programmers to actually have some training. Just like I would like for surgeons, lawyers, accountants, etc., to actually have some training.
7:36:16
beach
xificurC: Many people seem to think that anyone who has written a 10-line program should be allowed to practice programming professionally.
7:36:58
schweers
shrdlu68, xificurC: it may be associated with negative behavior, but I don’t think it need be that way. Anyway, being a jerk about it and having prejudices(?) is not good. Not my point though.
7:37:54
xificurC
beach: sounds like kids calling you a swot to be honest. I try not to take those people's aggressive behavior to heart
7:39:45
aeth
schweers: Yes, but even beyond language choice, I'm not sure that software engineering has a solid foundation. Or, at least, if it has one, it doesn't seem to be the majority position.
7:40:03
aeth
If software engineering were to be treated like engineering (and maybe it should be), then it should be engineering first.
7:40:36
schweers
aeth: no, I guess the field is too young for that kind of maturity. Also, software is very different from regular engineering
7:40:49
jach[m]
jdz: even homeopaths are trained to properly dillute the thing.. but also, you can write a 10 line program that does something useful, vs homeopaths that aren't actually useful
7:41:23
xificurC
I think too much work has come too quickly and now everyone is trying to do whatever they can to code. It's not exactly anyone's fault, the field is just young
7:41:25
beach
schweers: I agree that software is very different from the artifacts that engineering is concerned with.
7:42:21
schweers
xificurC: I agree. No one entity is at fault, but the result is still very bad and needs fixing. And I don’t see much fixing going on.
7:43:28
jdz
I think this is relevant, if you have ~40 minutes to spare: https://www.err.ee/836236/video-google-0-projekti-tarkvarainseneri-ettekanne-cyconil
7:43:43
shrdlu68
Does every other company need a web developer? This is a bizarre field, and Im struck by how ephemeral our work is.
7:43:50
jach[m]
A better analogy might be writing fiction. "Real authors" and readers of them tend to be pretty snobby against things like fanfiction, and even teenage authors producing highly derivative original-enough books.
7:43:55
schweers
xificurC: some, yes. I found a job where I can write software in CL, so that’s a start. I must say that I’m the only one here, and my collegue insists on using scheme :/
7:45:42
schweers
I know, but using CL really has made me see how it is a better language (and has better implementations) than others, even scheme.
7:48:09
aeth
That's the problem with Scheme. You're (usually) not a Scheme programmer, you're a programmer of some Scheme dialect.
7:48:23
aeth
And if you like an obscure one (like Gauche) you don't get access to much of an ecosystem.