rwcom has quit [Read error: Connection reset by peer]
rwcom has joined #lisp
SGASAU has joined #lisp
karen-chan has joined #lisp
karen-chan has quit [Read error: Connection reset by peer]
Kundry_Wag has joined #lisp
renzhi has quit [Read error: Connection reset by peer]
shangul has joined #lisp
epony has quit [Remote host closed the connection]
shifty has quit [Ping timeout: 260 seconds]
shangul has quit [Ping timeout: 260 seconds]
Nilby has quit [Read error: Connection reset by peer]
zaquest has quit [Quit: Leaving]
zaquest has joined #lisp
rwcom has quit [Quit: Ping timeout (120 seconds)]
rwcom has joined #lisp
mono has quit [Remote host closed the connection]
Kundry_Wag has quit [Remote host closed the connection]
efm has quit [Read error: Connection reset by peer]
buffergn0me has quit [Ping timeout: 240 seconds]
efm has joined #lisp
Oladon has joined #lisp
<bitmapper>
great
<bitmapper>
i literally cannot use the free version of lispworks/allegro cl because of macos catalina
buffergn0me has joined #lisp
renzhi has joined #lisp
SGASAU has quit [Remote host closed the connection]
SGASAU has joined #lisp
choegusung has joined #lisp
renzhi has quit [Ping timeout: 240 seconds]
bitmapper has quit [Ping timeout: 240 seconds]
cyraxjoe has joined #lisp
MightyJoe has quit [Ping timeout: 260 seconds]
buffergn` has joined #lisp
buffergn0me has quit [Read error: Connection reset by peer]
buffergn` has quit [Client Quit]
buffergn0me has joined #lisp
CrazyPython has quit [Ping timeout: 240 seconds]
pilne_ has joined #lisp
pilne has quit [Ping timeout: 246 seconds]
EvW has quit [Ping timeout: 240 seconds]
Fare has quit [Ping timeout: 244 seconds]
shangul has joined #lisp
shangul has quit [Ping timeout: 246 seconds]
renzhi has joined #lisp
Fare has joined #lisp
cyberoctopi has joined #lisp
Jeanne-Kamikaze has joined #lisp
acolarh has quit [Ping timeout: 260 seconds]
<pfdietz>
I wonder if there should be a :standard-package-nicknames option, and a :use-standard-nicknames. The former attacjes a standard PLN to a package; the latter says use that package's standard PLNs as PLNs for it in this package.
<pfdietz>
If the standard PLNs ever collide in another package, the user can manually define different ones.
<pfdietz>
I really should get to sleep. ELS is soon.
ayuce has left #lisp ["ERC (IRC client for Emacs 28.0.50)"]
cyberoctopi has quit [Ping timeout: 244 seconds]
pilne_ has quit [Quit: IceChat - Keeping PC's cool since 2000]
acolarh has joined #lisp
buffergn0me has quit [Ping timeout: 240 seconds]
lxbarbosa has quit [Read error: Connection reset by peer]
shangul has joined #lisp
pfdietz has quit [Ping timeout: 240 seconds]
turona has quit [Ping timeout: 265 seconds]
shangul has quit [Ping timeout: 260 seconds]
cyberoctopi has joined #lisp
turona has joined #lisp
<slyrus_>
Oladon: hey
cyberoctopi has quit [Ping timeout: 260 seconds]
buffergn0me has joined #lisp
pjb has quit [Ping timeout: 265 seconds]
<beach>
Good morning everyone!
nowhere_man has quit [Remote host closed the connection]
nowhere_man has joined #lisp
mwgkgk has joined #lisp
shangul has joined #lisp
cyberoctopi has joined #lisp
Bike has quit [Quit: Lost terminal]
cyberoctopi has quit [Ping timeout: 240 seconds]
Bourne has quit [Read error: Connection reset by peer]
terpri has quit [Read error: Connection reset by peer]
torbo has quit [Remote host closed the connection]
Codaraxis has joined #lisp
cyberoctopi has joined #lisp
epony has joined #lisp
gxt has joined #lisp
efm has quit [Ping timeout: 240 seconds]
GuerrillaMonkey has joined #lisp
efm has joined #lisp
Jeanne-Kamikaze has quit [Ping timeout: 240 seconds]
cyberoctopi has quit [Ping timeout: 246 seconds]
MerlinTheWizard has joined #lisp
dddddd has quit [Ping timeout: 260 seconds]
<MerlinTheWizard>
Hey, so there should probably be a list of lisp resources in the channel banner.
<MerlinTheWizard>
And I'll put in my vote for the Common Lisp Quick Reference: http://clqr.boundp.org/
<MerlinTheWizard>
It's beautifully laid out and designed for printing or reading on your computer.
vlatkoB has joined #lisp
<verisimilitude>
You can specialize a method on those array specializations required, Oladon, such as BIT-VECTOR and STRING, but anything else is nonportable or bothersome.
<p_l>
MerlinTheWizard: due to how big the channel topic gets, I think I'd rather make a link to a minimal site listing such things, preferably one general to #lisp itself
<verisimilitude>
I've such a resource, but it's only intended for my Lisp group.
<MerlinTheWizard>
p_l, could you do that?
GuerrillaMonkey has quit [Remote host closed the connection]
<MerlinTheWizard>
It seems common lisp is probably not getting as many users as it could be because information about the language just seems to be scattered around the internet mostly. A lot of the bigger "hubs' and "authorities" are weirdly formatted, ancient looking sites.
<verisimilitude>
Why would the number of people using Common Lisp be an issue?
brutalist has joined #lisp
<MerlinTheWizard>
verisimilitude, with more users comes more resources for people wanting to use the language... I'm a new CL user and I need resources. CLHS is unfortunately proprietary and weirdly formatted. More users means CL will have more life on the internet and better support.
<verisimilitude>
I can read the CLHS in most any WWW browser ever made, so it's nice.
<no-defun-allowed>
I think having information being "scattered" is a good thing; in the sense that having fewer coherent outlets would require having fewer outlooks on the language and how to use it.
<verisimilitude>
I agree with no-defun-allowed; I usually ignore this channel, because I dislike Freenode and have my own entirely separate Lisp venues, as an example.
<MerlinTheWizard>
no-defun-allowed I would call that a non-sequitur. My use of "scattered" here means "disconnected", not "decentralized".
<no-defun-allowed>
In the vain of http://bettermotherfuckingwebsite.com/, it is also quite easy to make the CLHS look pretty nice with a few lines of CSS.
<MerlinTheWizard>
verisimilitude, you are preferring "private" to "public". I like "public" better.
<MerlinTheWizard>
no-defun-allowed, can you put it all on a single page with a little CSS?
<verisimilitude>
I like that Common Lisp is a standardized language without ``official'' resources.
<verisimilitude>
This channel is no more legitimate than my main IRC channel on a different network, and I like that.
<MerlinTheWizard>
No, it has official resources, who will sell you a badly scanned printout of the standard for $30.
<verisimilitude>
The standard document is clearly excluded.
<verisimilitude>
The opposite is garbage such as Rust where there's an official ``subreddit'' or whatever.
<MerlinTheWizard>
CLHS is based on the standard.
<no-defun-allowed>
A single page? Not easily, the CLHS is 2,300 hyperlinked documents, which was quite impressive when it was released.
<MerlinTheWizard>
no-defun-allowed, that's my main issue with it.
<verisimilitude>
I've a copy of ``ANSI Common Lisp'' I use as a reference at times.
<no-defun-allowed>
Yes, the standard costs money to access because it's a proper ANSI standard and all, and the CLHS is equivalent to the standard from what I have heard.
<no-defun-allowed>
"M-x hyperspec-lookup RET cons" is probably the easiest way to search that document.
<verisimilitude>
Sure, MerlinTheWizard, but it does lessen the legitimacy of any others.
<MerlinTheWizard>
Nope
nowhere_man has quit [Remote host closed the connection]
<MerlinTheWizard>
That's illogical
<verisimilitude>
I could be ostracized here, and I wouldn't even care; in other languages, being ostracized is effectively a ``death-sentence''.
nowhere_man has joined #lisp
<MerlinTheWizard>
Ostracism has little to do with the connectivity properties of a network of information sources.
<no-defun-allowed>
If one service is more popular than the other, it will attract more users just by being more popular. Not that you use e-mail these days, but how much mail do you get from people that isn't from their ISP's email server or from gmail?
<verisimilitude>
We may not be understanding each other, perhaps.
<p_l>
I see silly political discussion centered about not doing things, where people talk past each other on completely unrelated topics that don't actually conflict
<verisimilitude>
Most of my mail doesn't come from gmail, once I delete the spam, almost all of which comes from gmail.
<MerlinTheWizard>
"authority" just means a good source of information as presented in the linked webpage. One good source of information does not cancel out another.
<MerlinTheWizard>
So, yeah. We're talking past each other.
<beach>
MerlinTheWizard: There are many theories about why Common Lisp is not as popular as some might want it to be. But I can assure you that having information that is less scattered will not have any significant impact on the popularity of the language.
<no-defun-allowed>
(And, well, our ISP's email server is horrible as it has tiny quotas, has a mediocre spam filter, and accepts mail from phony email addresses. Someone told the users of that "Here's your email!!" and they didn't bother to look for anything better.)
<verisimilitude>
Whatever; back to the beginning, Common Lisp being popular or not isn't going to have much of an impact, especially if most new people just want to write libraries that are actually C libraries or other such things.
<no-defun-allowed>
(In that sense, if we assert one platform to be "the" platform, then people are not going to look at alternatives, regardless of quality.)
<MerlinTheWizard>
beach, I don't know about that. If people go to the irc channel on freenode because they were looking for good info on CL, and they find one or more solid links right off the bat, it makes them more likely to use the language.
<MerlinTheWizard>
no-defun-allowed, different people are different.
rgherdt has joined #lisp
<no-defun-allowed>
How does one find #lisp without looking for information on Lisp? I suppose it's possible you could guess the name, but it's not likely.
<verisimilitude>
Perhaps the user wants a channel for speach impediments.
<MerlinTheWizard>
My point exactly. If you came here, you were probably looking for something.
<beach>
MerlinTheWizard: I seriously doubt it would have any significant impact. There are some very strong psychological forces at play that prevent people from using Common Lisp.
<verisimilitude>
Why did you start learning CL, MerlinTheWizard?
<no-defun-allowed>
I don't think "if you came here, then you were looking for something" and "if you were looking for something, then you came here" are the same.
cyberoctopi has joined #lisp
<MerlinTheWizard>
I'm interested in functional programming and I want a language that let's me do just about anything.]
<beach>
MerlinTheWizard: Well, Common Lisp is not a functional programming language in that sense.
* no-defun-allowed
gets back to making a lightning talk presentation.
<beach>
MerlinTheWizard: It is a multi-paradigm language. In particular, it has excellent support for object-oriented programming.
<MerlinTheWizard>
no-defun-allowed, most IRC channels that I've frequented were not "leaf nodes" in that sense. They had useful resources available.
<saturn2>
is cliki.net "weirdly formatted and ancient looking"?
<p_l>
cliki is mostly dead
<p_l>
that's bigger problem
<MerlinTheWizard>
I'm not that interested in the OO aspect.
<MerlinTheWizard>
^
<beach>
MerlinTheWizard: If you are interested in functional programming, then something like Haskell or Clojure might be better for you.
<MerlinTheWizard>
I'll check out haskell later.
<verisimilitude>
Well, cliki doesn't ask for my email or other information to make changes, display advertisements, and I presume it's not trying to track me, so it's outdated, yes.
<no-defun-allowed>
verisimilitude: Heh.
<saturn2>
it doesn't use bootstrap.js or material design so it must be bad
<no-defun-allowed>
I'm not good at writing long texts, but apparently I struggle to write short texts too. Fun stuff.
<verisimilitude>
Try doing something else until inspiration strikes you, no-defun-allowed; would you like some more in-depth writing advice from me?
<p_l>
saturn2: the issue is the content, not everything else
<no-defun-allowed>
Thanks, but there isn't really time. I'm almost done though.
<verisimilitude>
Alright.
<MerlinTheWizard>
Yeah, I like the cliki layout just fine. But the web of CL content should be larger, more well connected, more up to date, and less proprietary.
<saturn2>
p_l: what do you mean by "dead"?
<verisimilitude>
I prefer the distributed ``random guys website'' model, MerlinTheWizard.
<verisimilitude>
It's a shame so much CL is on github.
<p_l>
saturn2: It pretty much stopped being updated much, especially after ASDF-Install stopped being a thing
<MerlinTheWizard>
verisimilitude, how do you find the "random guys website"? Can you tell me how many of them there are?
<verisimilitude>
No, and that's part of the point.
<MerlinTheWizard>
So it's better if information is harder to find?
<verisimilitude>
One finds such websites by looking.
<no-defun-allowed>
You use a search engine. There's probably at least two or three at the very least.
Grauwolf has quit [Read error: Connection reset by peer]
<MerlinTheWizard>
That knowledge I had before I even came to this IRC channel.
<verisimilitude>
Being scattered and distributed is better than everyone being on Facebook or similar filth, yes.
Grauwolf has joined #lisp
<MerlinTheWizard>
I didn't say "put all the information on facebook"
<MerlinTheWizard>
Again, talking past each other.
<MerlinTheWizard>
Wrong comparison. Strawman.
<verisimilitude>
That's an example of centralization, is all.
<verisimilitude>
Anyway, good luck with finding a Common Lisp programmer who cares much about this issue, MerlinTheWizard.
<MerlinTheWizard>
And I told you, I wasn't arguing for more centralization. Just more and better "authorities" and "hubs", as defined in the linked page.
<MerlinTheWizard>
Obviously, you speak for all of them, king verisimilitude.
<MerlinTheWizard>
The people who built, cliki, I'm pretty sure they cared about this issue.
<MerlinTheWizard>
So you've been making nothing but ridiculous points this whole time.
<verisimilitude>
This is just the general attitude I've observed over time.
<MerlinTheWizard>
I'd guess your perspective is rather biased, not to mention short sighted.
<verisimilitude>
Also, if you want to deal in functional programming, also consider Scheme.
<verisimilitude>
I disagree with several of those here, yes.
<MerlinTheWizard>
I do. But I want to do stuff.
<MerlinTheWizard>
So it's CL for now.
<p_l>
MerlinTheWizard: there are some attempts to make "friendly" first contact website (for example, http://lisp-lang.org) but I think careful shepherding of Wikipedia might be of more benefit to people in first contact
dale has quit [Quit: My computer has gone to sleep]
<MerlinTheWizard>
p_l, precisely. People need "hubs" so they can get started.
<verisimilitude>
I didn't need a ``hub''.
<p_l>
verisimilitude: great for you, Oh Great Gatekeeper
narimiran has joined #lisp
<verisimilitude>
Or, at the least, I wouldn't look back on it like that.
<MerlinTheWizard>
LOL
<saturn2>
i think hubs are good
<MerlinTheWizard>
Admission of biased and skewed perspective.
<MerlinTheWizard>
A CL user thinks hubs are good? AMAZING!
frgo has quit [Remote host closed the connection]
<p_l>
having a FAQ link in topic is, IMO, perfectly normal and could fulfill both easier ways to find #lisp (due to findability by web crawlers) as well as serve as good point to get people to jump further from
<MerlinTheWizard>
Yes, I agree.
<saturn2>
there aren't many rewards for creating beginner resources, i guess, without a source of capital that needs to grow the user base
frgo has joined #lisp
emys has joined #lisp
MerlinTheWizard has quit [Ping timeout: 244 seconds]
frgo has quit [Ping timeout: 244 seconds]
jayde has quit [Quit: too sleepy today]
Cymew has joined #lisp
<beach>
So that was yet another "Here is what YOU should do to make Common Lisp more attractive".
<verisimilitude>
Oh, that happens often here?
<beach>
Regularly, but perhaps not often.
<no-defun-allowed>
Infrequently (or when I'm asleep and haven't checked the logs), yes.
<verisimilitude>
How amusing.
emys has quit [Ping timeout: 260 seconds]
<beach>
They all claim some deep understanding as to why Common Lisp is not more popular, and it is always a single issue that is easy to fix if the person in charge would just put in a little more work.
<beach>
And, of course, they all assume that Common Lisp popularity is the main goal for everyone here, and that we are all quite puzzled as to why it is not more popular.
<verisimilitude>
While entertaining here, I'm glad to be free of such ignorance in my usual venue.
<verisimilitude>
Not a single language I know has me caring about anyone else using it.
<verisimilitude>
This is probably inconceivable for plenty of inexperienced newcomers.
gravicappa has joined #lisp
<verisimilitude>
Just today I saw some idiot claiming everyone must copy and paste code from Stack Overflow.
<no-defun-allowed>
"How did they make Stack Overflow without using Stack Overflow?" but in an actually funny context.
<verisimilitude>
Oh, that was an actual question on the site?
<verisimilitude>
Imagine deriving any ego from programming, and then just copying and pasting or gluing together other people's code.
<no-defun-allowed>
No, it's just a joke that is repeated quite frequently in "mainstream" programming humour places; up there with having most-positive-fixnum compiler errors and trying to balance parens.
<verisimilitude>
Oh, alright. Stack Overflow can be a decent source of entertainment with such questions.
<beach>
p_l: You mean MerlinTheWizard? Sure, but instead of humility, we often see this behavior, insisting that WE fix things so that the obvious problems of similar newbies will disappear.
frgo has joined #lisp
<p_l>
beach: I didn't read it as "insisting that we do it" but asking if it wouldn't be useful. The discussion immediately derailed with push that not only it's not useful, it should be abhorred because <start a long political nonsense>
<p_l>
and this spiraled in vicious cycle further
buffergn0me has quit [Quit: ERC (IRC client for Emacs 26.2)]
<no-defun-allowed>
How long should my lightning talk be?
shangul has quit [Ping timeout: 244 seconds]
<verisimilitude>
What's a lightning talk; I take it by the name it's intended to be quick.
<verisimilitude>
If this is the only consideration, then I'd make it as short as I could, no-defun-allowed.
<no-defun-allowed>
A format of presentation that ELS has, which is pretty short.
<no-defun-allowed>
Someone suggested that I could write one for this year, given that it will be presented online.
<beach>
no-defun-allowed: 5 minutes max I think.
emys has quit [Ping timeout: 244 seconds]
<phoe>
yes, a hard limit of 5 minutes
<phoe>
also good morning
<no-defun-allowed>
Got it.
<p_l>
no-defun-allowed: target 2 minutes
<verisimilitude>
Oh, that's right, conventions are being shuttered now; yes, considering this, make it as short as possible, no-defun-allowed.
<p_l>
the remaining 3 will be eaten by technical issues and/or heckling
<verisimilitude>
If you make the shortest one with useful information, it's an achievement.
<no-defun-allowed>
I see.
<p_l>
no-defun-allowed: especially given the broadcast nature of this year's ELS
<phoe>
verisimilitude: it's weird to hear "conventions are being shuttered" about something that has been going on unchanged for at least several ELSes
<verisimilitude>
Was this always an online event?
<beach>
No, first time.
shangul has joined #lisp
<p_l>
my normal methods of estimation quickly go into negative time for lightning talks
<verisimilitude>
Why would it be strange to read, then?
<beach>
What do you mean?
<verisimilitude>
I ask phoe, to clarify.
emys has joined #lisp
<phoe>
oh, you mean *that* sort of conventions - sorry, don't mind me, I'm waking up
<beach>
verisimilitude: phoe is just saying that 5-minute lightning talks have been a feature of ELS for many year. Perhaps since the beginning.
<phoe>
while also trying to read the sad stuff in the chat log up there
<verisimilitude>
Oh, I see the misunderstanding now.
<verisimilitude>
I referred to events and he practices, yes.
<no-defun-allowed>
Great, this looks like a ten-minute presentation so far.
<verisimilitude>
What's the topic?
<beach>
:(
<no-defun-allowed>
verisimilitude: I'm trying to give a high level introduction to my Netfarm distributed object system. It took me two minutes to explain the aim and the centralised/federated/distributed terms, which is 1/5 of my presentation.
<verisimilitude>
I suppose I can see why shortening that would be difficult.
<beach>
no-defun-allowed: The trick is to realize that you don't have to provide all the information.
cyberoctopi is now known as iAmDecim
_paul0 has quit [Remote host closed the connection]
paul0 has joined #lisp
terpri has quit [Quit: Leaving]
emys has quit [Ping timeout: 240 seconds]
MerlinTheWizard has joined #lisp
terpri has joined #lisp
<no-defun-allowed>
8:20 for the whole presentation.
<no-defun-allowed>
beach: Yeah, I want to mention the "distributed hash table", "object system", "script machine" and some of the "other stuff" with some examples and definitions, which is tricky.
<no-defun-allowed>
It wouldn't be nice to throw new terms without explaining them at the audience though.
<phoe>
this certainly seems more like a full-size talk then
<no-defun-allowed>
Dammit.
shka_ has joined #lisp
<phoe>
if anything, I'd do a lightning talk that only shows the examples and the most important practical parts *without* introducing the new terms at all - "I have several networked computers, if I do this on one computer, then this thing pops up on all these computers" three or four times
<no-defun-allowed>
Sure.
<phoe>
and/or do a full size 25-minute talk that actually explains the theoretical basics and then performs the examples
<phoe>
and/or just record this stuff and throw it at the ELS mailing list and Reddit; in a way, we have ELS all year long this year due to the current pandemic circumstances
<no-defun-allowed>
Right then.
* easye
chants "ELS ... All ... Year"
<verisimilitude>
An obvious suggestion is heavily compressing all of your sentences, no-defun-allowed.
<verisimilitude>
That could be bothersome to actually listen to, however.
<phoe>
verisimilitude: I tried that with my first ELS lightning talk ever
<phoe>
the effect was poor
<phoe>
I can't therefore really recommend it; human listening and understanding bandwidth isn't good enough for compression to be really viable
<phoe>
easye: every day is lisp day™
<verisimilitude>
Using a language such as Lojban isn't an option, yes.
<no-defun-allowed>
Okay, I dropped the less interesting parts and it should be shorter now.
<beach>
no-defun-allowed: It's the hard part of any talk, knowing what not to say.
<phoe>
verisimilitude: neither is gzipping the stream of data you present and leaving uncompressing it as an exercise for the reader, or *even* just speaking faster
<Shinmera>
no-defun-allowed: Lightning talks traditionally are '5 minutes, and if you go one second more, Didier kills you live on stage'
<no-defun-allowed>
Haha, I'm immortal now!
<verisimilitude>
Hello, Shinmera.
<verisimilitude>
My notes on that dns-client were noticed, right?
<Shinmera>
Yeah
random-nick has joined #lisp
<Shinmera>
Not doing anything about it at the moment though.
<verisimilitude>
I've further notes on that DECODE-HOST; I'd use DPB and LDB instead of your subtraction by a binary constant and I'd simplify it by making the increment conditional; the only reason I understand what that constant did was because I've read some of the DNS documents.
<Shinmera>
Sure. I just copied another implementation because I was in a hurry.
<verisimilitude>
Alright.
emys has joined #lisp
<verisimilitude>
I don't want to unduly criticize, but the special variables specifically for Google and Cloudflare DNS servers is something I'd never write.
Bourne has joined #lisp
<no-defun-allowed>
7:00 now. If I extrapolate this, it should take me about 2 more takes to get it under 5 minutes.
<Shinmera>
you can just... not use them if you don't like them.
<verisimilitude>
Well, I'm going to write my own DNS library anyway; it's simply my final note on it.
<Shinmera>
no-defun-allowed: don't speak too fast, though...
* Shinmera
shrugs
SGASAU has quit [Remote host closed the connection]
* no-defun-allowed
cuts out more stuff
SGASAU has joined #lisp
orivej has joined #lisp
barodaret has quit [Read error: Connection reset by peer]
jprajzne has joined #lisp
<no-defun-allowed>
5:30 now. Maybe I can speak a tad faster.
<verisimilitude>
Remove any greetings.
<beach>
I don't think it will be rejected at that length.
orivej has quit [Ping timeout: 260 seconds]
frgo has quit [Remote host closed the connection]
emys has quit [Ping timeout: 260 seconds]
heisig has joined #lisp
<beach>
... especially since there seem to be relatively few lightning talks today.
<no-defun-allowed>
Hm, I don't want to find out what the online equivalent of being killed live on stage is, nor take too long given the circumstances.
frgo has joined #lisp
<beach>
I understand.
<phoe>
Didier used to get killed himself during one of his talks that went for, like, six and half a minute
<verisimilitude>
Just look at what happened to RMS, no-defun-allowed.
<phoe>
and look he's still around to organize ELS this year
<Shinmera>
Being dead ain't stopped nobody before
<beach>
Yeah, being alive is overrated anyway.
<verisimilitude>
I don't even like being alive.
frgo has quit [Remote host closed the connection]
iAmDecim has quit [Ping timeout: 240 seconds]
<no-defun-allowed>
Right, now it's pretty short. I can probably cut out :30 removing my stuttering.
<beach>
no-defun-allowed: Are you aiming for today or tomorrow (UTC+2) for your lightning talk?
<no-defun-allowed>
I'm not sure, I just need to do some quick editing now.
MerlinTheWizard has quit [Ping timeout: 240 seconds]
<beach>
It doesn't matter much I guess. Shinmera will fit it in when he can.
emys has joined #lisp
iAmDecim has joined #lisp
<verisimilitude>
What will Shinmera give a talk on?
<Shinmera>
nothing.
<Shinmera>
I'm organising.
<phoe>
How To Organize ELS Over the Internet: A Practical Approach
<beach>
Heh!
<Shinmera>
step 1: despair
<beach>
A lightning talk should be long enough for that, right? :)
emys has quit [Ping timeout: 260 seconds]
<Shinmera>
Either that, or it's a life long experience :V
<Shinmera>
should be able to download the VODs, too.
<Gnuxie[m]>
Shinmera: thank you :)
v_m_v has quit []
milanj has joined #lisp
terpri has quit [Ping timeout: 240 seconds]
xantoz has quit [Read error: Connection reset by peer]
<Krystof>
right, streaming at 1080p means I have to be very close to our router (and my wife must not be on a simultaneous call)
<Krystof>
I'm getting the gist of the talks though :-)
<Shinmera>
It's less the resolution, and more the bitrate. I thought twitch would offer rescaling options like it usually does, but they're not on for whatever reason.
<easye>
Krystof: if you have a tablet, I think there is a "Twitch App" which might do optimizations.
frgo has quit [Remote host closed the connection]
<mgr_>
but the image quality is nice and shiny when it works :)
<phoe>
it's also nice and shiny when it doesn't work, except then also it doesn't move
frgo has joined #lisp
<Krystof>
Shinmera: I think you have to be a "partner" for your viewers to get options
<easye>
Other than Marco's gorgeous presentation, 1080p has been mostly unneeded.
<Shinmera>
Possible that they changed that, but I thought my usual stream also got them.
<Shinmera>
and I'm not partnered
<MichaelRaskin>
It seems to be true that for a given target bitrate, higher starting resolution gives better results…
<MichaelRaskin>
But Shinmera did not even have enough wall-clock time between getting the videos and putting them up to recode anything
narimiran has quit [Ping timeout: 244 seconds]
<easye>
Well, we could maybe get Krystof to give us a MAP/REDUCE instance to recode.
<easye>
(not a serious suggestion)
<Krystof>
*phew* :)
<Shinmera>
MichaelRaskin: I downloaded the last video for today this morning...
<Krystof>
that is entirely consistent with my usual timetable for writing slides
<MichaelRaskin>
Shinmera: even if it were yesterday, still not enough time in practice to recode…
Bourne has joined #lisp
scymtym__ has joined #lisp
<cpape>
Is a audio-only mode possible? My bandwidth is too limited for the 1080.
scymtym_ has quit [Ping timeout: 260 seconds]
scymtym__ has quit [Client Quit]
scymtym has joined #lisp
sroh has quit [Read error: Connection reset by peer]
milanj has quit [Quit: This computer has gone to sleep]
emys has joined #lisp
xantoz has joined #lisp
<Krystof>
speaking as someone who has been a local organizer for a conference - I think the fact that we have anything at all in the way of ELS, let alone something that is as good as this is pretty cool
<Shinmera>
MichaelRaskin: yes, plus, as I said, I was doing this on the assumption that there would be live reencodes for people with worse connections.
<Krystof>
and even when my video cuts out, getting the twitch chat is great :)
<Shinmera>
Krystof: I think so too
<luis>
Shinmera: yes, thanks for this!
<cpape>
It is great, yes, thank you Shinmera! I follow the chat; will watch the videos async.
scymtym has quit [Quit: ERC (IRC client for Emacs 28.0.50)]
Lord_of_Life_ has joined #lisp
Lord_of_Life has quit [Ping timeout: 240 seconds]
Lord_of_Life_ has quit [Excess Flood]
shangul has joined #lisp
Lord_of_Life has joined #lisp
scymtym has joined #lisp
Bike has joined #lisp
<phoe>
(incf Shinmera)
monokrom has joined #lisp
ljavorsk has quit [Ping timeout: 260 seconds]
<jmercouris>
(decf Shinmera)
<Bike>
wow, rude.
<White_Flame>
(setf Shinmera #c(1 1))
<jmercouris>
I'm just bringing Shinmera back to its original value
<phoe>
(complex-incf Shinmera)
EvW has joined #lisp
<jmercouris>
(setf Shinmera (- 1 (+ 1 Shinmera)))
hiroaki has joined #lisp
<phoe>
so you basically tumble him around for a moment and then set him back in his place
rgherdt has quit [Remote host closed the connection]
amerlyq has quit [Quit: amerlyq]
<splittist>
Very well done Shinmera and presenters
akoana has joined #lisp
thodg has joined #lisp
<thodg>
hello
luckless has joined #lisp
<cgay>
What's the relationship between ultralisp.org and quicklisp? "Ultralisp is a quicklisp distribution, which updates every 5 minutes" Does that mean it has the same stuff as quicklisp but it's updated more often?
<thodg>
you can push your own projects from github
<phoe>
cgay: quicklisp is actually two things
<phoe>
the first is quicklisp client, which is a piece of lisp stuff you run to download things
<phoe>
the second is quicklisp dist, which is a set of Lisp software updated ~once a month
<phoe>
the main quicklisp dist updates once a month
<phoe>
the main ultralisp dist updates on every change pushed to any of the ultralisp git projects
<phoe>
so it's "stable" versus "bleeding edge"
<cgay>
i see, thanks!
q-u-a-n23 has joined #lisp
EvW has quit [Ping timeout: 240 seconds]
<ralt>
the important thing about quicklisp dist is that all the dependencies are compiled together, i.e. the set of libraries in a given distribution are guaranteed to at least compile together with all the given versions
emys has quit [Ping timeout: 265 seconds]
Inline has quit [Remote host closed the connection]
<phoe>
and ultralisp makes no such guarantees
edgar-xyz has joined #lisp
SGASAU` has joined #lisp
roelj has quit [Remote host closed the connection]
<cgay>
you mean they compile together without warnings, or what? because IME for Lisp code to "compile together" isn't a very strong guarantee. :)
<housel>
If it breaks you get to keep the pieces, as they say
<cgay>
:)
Adamclisi has quit [Ping timeout: 240 seconds]
lonjil has quit [Quit: No Ping reply in 180 seconds.]
<phoe>
cgay: they are loadable without ASDF going crazy
stepnem has quit [Ping timeout: 260 seconds]
<Shinmera>
cgay: It means every system is loaded at least once before the dist is created.
<Shinmera>
each being loaded in its own fresh image.
gxt has quit [Ping timeout: 240 seconds]
edgar-rft has quit [Ping timeout: 244 seconds]
<Shinmera>
if errors happen during this, a bug is filed and the dist release delayed.
HDurer_ has joined #lisp
stepnem has joined #lisp
dilated_dinosaur has quit [Ping timeout: 260 seconds]
trittweiler has quit [Ping timeout: 260 seconds]
froggey has quit [Ping timeout: 256 seconds]
jprajzne has quit [Ping timeout: 265 seconds]
duncan_ has quit [Ping timeout: 260 seconds]
cmatei has quit [Ping timeout: 244 seconds]
<cgay>
Tests?
lonjil has joined #lisp
<Shinmera>
There's no standardised interface to run tests and get results from them, so they're not run.
gxt has joined #lisp
emys has joined #lisp
<phoe>
asdf:test-system attempted to be one, but even it admit that there's no standardization there
_heisig has joined #lisp
jprajzne has joined #lisp
roelj has joined #lisp
<Shinmera>
it could have been one if it had taken the time to standardise the behaviour
<Shinmera>
oh well.
heisig has quit [Ping timeout: 260 seconds]
thodg has quit [*.net *.split]
Kundry_Wag has quit [*.net *.split]
makomo has quit [*.net *.split]
gabiruh has quit [*.net *.split]
SGASAU has quit [*.net *.split]
scymtym has quit [*.net *.split]
whiteline has quit [*.net *.split]
kamid has quit [*.net *.split]
__jrjsmrtn__ has quit [*.net *.split]
rixard has quit [*.net *.split]
dtman34 has quit [*.net *.split]
sjl has quit [*.net *.split]
jackdaniel has quit [*.net *.split]
ck_ has quit [*.net *.split]
HDurer has quit [*.net *.split]
Grue` has quit [*.net *.split]
Blukunfando has quit [*.net *.split]
axion has quit [*.net *.split]
nchambers has quit [*.net *.split]
q-u-a-n2 has quit [*.net *.split]
scal_ has quit [*.net *.split]
loli has quit [*.net *.split]
jbgg has quit [*.net *.split]
fengshaun has quit [*.net *.split]
cracauer has quit [*.net *.split]
copec has quit [*.net *.split]
pok has quit [*.net *.split]
loke has quit [*.net *.split]
brass has quit [*.net *.split]
fiddlerwoaroof has quit [*.net *.split]
cg505 has quit [*.net *.split]
heisig has joined #lisp
Inline has joined #lisp
ck_ has joined #lisp
verisimilitude has left #lisp ["ERC (IRC client for Emacs 24.5.1)"]
trittweiler has joined #lisp
heisig has quit [Client Quit]
duncan_ has joined #lisp
cmatei has joined #lisp
nchambers has joined #lisp
fiddlerwoaroof has joined #lisp
_heisig has quit [Ping timeout: 260 seconds]
froggey has joined #lisp
brass has joined #lisp
<Bike>
what would a standard interface return? T for pass? NIL for pass and something else as a representation of failures?
<phoe>
Bike: primary value: true if everything is as expected, false if someone needs to take a look at the results
<phoe>
secondary value: test-dependent object that shows the test result(s)
Lycurgus has quit [Remote host closed the connection]
lonjil has joined #lisp
rgherdt has joined #lisp
gravicappa has joined #lisp
shifty has joined #lisp
jayde has joined #lisp
Mawile has quit [Quit: Leaving]
gxt has quit [Quit: WeeChat 2.8]
Archenoth has joined #lisp
nullman has joined #lisp
varjag has quit [Quit: ERC (IRC client for Emacs 25.2.2)]
EvW1 has joined #lisp
schjetne has quit [Ping timeout: 260 seconds]
scymtym_ has joined #lisp
MerlinTheWizard has quit [Ping timeout: 244 seconds]
scymtym has quit [Ping timeout: 246 seconds]
rixard has quit [Remote host closed the connection]
rixard has joined #lisp
<Shinmera>
Bike: you could go as far as standardising objects for test unit results.
rixard has quit [Ping timeout: 260 seconds]
karlosz has joined #lisp
karlosz has quit [Client Quit]
karlosz has joined #lisp
pfdietz has joined #lisp
<pfdietz>
On the subject of test APIs, I have opinions.
<phoe>
please let me know it
<phoe>
s/it/them
* cgay
waits
<pfdietz>
It should be possible to decompose the test suite into individual tests that can be run separately.
shifty has quit [Ping timeout: 256 seconds]
shifty has joined #lisp
<pfdietz>
Results should include expected/unexpected passes and fails. That is, it should be possible to include tests that are expected to fail, without rendering the test suite failing.
<pfdietz>
There should be a standard way of reporting success/failure to ASDF.
emys has quit [Ping timeout: 260 seconds]
<phoe>
I can wholeheartedly agree with this very last point
Adamclisi has quit [Ping timeout: 240 seconds]
froggey has quit [Ping timeout: 240 seconds]
dilated_dinosaur has quit [Ping timeout: 256 seconds]
lonjil has quit [Quit: No Ping reply in 180 seconds.]
<pfdietz>
There should be some sort of traceability between individual tests and parts of the system under test. This may not be part of the testing API per se, but I want to be able to answer questions like "to test X, run the tests in the set Y".
duncan_ has quit [Ping timeout: 260 seconds]
cmatei has quit [Ping timeout: 256 seconds]
lonjil has joined #lisp
<pfdietz>
(asdf:test-system <systemname>) should load the tests if they are not already loaded.
<phoe>
pfdietz: you are already describing a unit test framework
<pfdietz>
There should be a way of describing how to generate tests.
emys has joined #lisp
<pfdietz>
There should be a way to evaluate the adequacy of the tests.
scymtym__ has joined #lisp
schjetne has joined #lisp
<pfdietz>
I've been working on something for the latter that motivates some of those earlier points.
<pfdietz>
It's a mutation testing framework, something that mutates a function (or a set of methods) and runs tests against the mutated code.
karlosz has quit [*.net *.split]
gravicappa has quit [*.net *.split]
Grue` has quit [*.net *.split]
axion has quit [*.net *.split]
scal_ has quit [*.net *.split]
loli has quit [*.net *.split]
jbgg has quit [*.net *.split]
fengshaun has quit [*.net *.split]
cracauer has quit [*.net *.split]
copec has quit [*.net *.split]
loke has quit [*.net *.split]
pok has quit [*.net *.split]
cg505 has quit [*.net *.split]
scymtym_ has quit [*.net *.split]
nullman has quit [*.net *.split]
thodg has quit [*.net *.split]
Kundry_Wag has quit [*.net *.split]
makomo has quit [*.net *.split]
gabiruh has quit [*.net *.split]
whiteline has quit [*.net *.split]
dtman34 has quit [*.net *.split]
sjl has quit [*.net *.split]
jackdaniel has quit [*.net *.split]
kamid has quit [*.net *.split]
__jrjsmrtn__ has quit [*.net *.split]
<pfdietz>
So, it needs a standardized way of determining if the tests passed.
<pfdietz>
It would benefit from only running the tests that exercise the mutated function/methods.
<pfdietz>
If tied with a test input generator, it could automatically look for inputs that kill mutants the test suite missed. This is beyond using that test input generator just for property based testing for properties derived from requirements.
duncan_ has joined #lisp
edgar-xyz has quit [Quit: Leaving]
dilated_dinosaur has joined #lisp
karlosz has joined #lisp
axion has joined #lisp
jbgg has joined #lisp
cracauer has joined #lisp
loke has joined #lisp
copec has joined #lisp
cg505 has joined #lisp
scal_ has joined #lisp
Grue` has joined #lisp
gravicappa has joined #lisp
loli has joined #lisp
fengshaun has joined #lisp
pok has joined #lisp
cmatei has joined #lisp
froggey has joined #lisp
nullman has joined #lisp
thodg has joined #lisp
Kundry_Wag has joined #lisp
makomo has joined #lisp
gabiruh has joined #lisp
kamid has joined #lisp
whiteline has joined #lisp
sjl has joined #lisp
dtman34 has joined #lisp
jackdaniel has joined #lisp
__jrjsmrtn__ has joined #lisp
__jrjsmrtn__ has quit [Max SendQ exceeded]
dtman34 has quit [Max SendQ exceeded]
edgar-rft has joined #lisp
<pfdietz>
I would like the suite to be multithreaded, or execute in multiple processes, if possible.
Adamclisi has joined #lisp
dtman34 has joined #lisp
<pfdietz>
I would like it to tell me how long individual tests take to run.
<pfdietz>
(That's important in the mutation tester, so a timeout can be set to detect looping in the mutants.)
<phoe>
this has escalated quickly from the point where we said that it's hard for ASDF to know if the test suite for a given project ran successfully or not
<pfdietz>
Yes, I have needs. :)
msk has joined #lisp
<pfdietz>
I'
<pfdietz>
I'm driven here by a lack of adequate testing in many ql systems.
<pfdietz>
I want something that can check how well or poorly a system is tested, and help mass produce new tests if more are needed.
<flip214>
ugh, 13.4 GB for one of the ELS videos...
<cgay>
pfdietz: i am in favor of all the things on your list. what do you mean exactly by "there should be a way of describing how to generate tests"?
<flip214>
13.7 even
<Shinmera>
cgay: I presume a grammar of sorts that can be randomly instantiated to create a valid test form.
<Shinmera>
Something I have been thinking about for Parachute, though not much yet.
<Shinmera>
flip214: I'm streaming at about 6Mbps, so yeah, it adds up quick.
<pfdietz>
What he said.
<Shinmera>
I'll probably lower it to 4Mbps for tomorrow in hopes it'll help for people with worse internet.
<pfdietz>
This would be used for two things. First, you can express a property of some function, then bombard it with random inputs and see if the property is ever violated.
<flip214>
Shinmera: I'd have hoped to find videos of individual talks... and perhaps in different sizes and/or compression formats
rixard has quit [Ping timeout: 260 seconds]
<Shinmera>
someone has to put in the work for that.
<pfdietz>
Second, it can be used with some metric of coverage to generate test inputs that improve that metric. This is useful even if you don't have an oracle that tells you if the output is correct.
<flip214>
and while I'm posting wishes, why not include a slides+voice video channel as well?
<Shinmera>
eh?
<Bike>
the live ended, what, like a couple hours ago? these sound like things that take time even whe someone is actually doing them
<flip214>
Shinmera: I'm well aware of that. And we all (me, at the very least!) are very grateful to you!
__jrjsmrtn__ has joined #lisp
Inline has quit [Ping timeout: 265 seconds]
<flip214>
well, I haven't looked at the videos yet... but at that size I guess there are not that many slide pictures and more persons talking
edgar-rft has quit [Quit: Leaving]
<flip214>
slides (as still pictures) plus a bit of sound can't be _that_ big, right?
<Bike>
would that not depend on the video compression
efm has joined #lisp
<MichaelRaskin>
There seems to be no good video format for vector graphics + voice
<flip214>
well, the vector graphics shouldn't really matter - just put still pictures in every few seconds, won't take that much bandwidth
edgar-rft has joined #lisp
<flip214>
MichaelRaskin: animated gif .... or animated SVG.... or just PDFs of the slides that can be clicked-through along to an audio-only stream
<MichaelRaskin>
PDFs are … getting there
sjl_ has joined #lisp
<MichaelRaskin>
Note that it is not BigBlueButton where you would have a recording of slide change times + audio, here Shinmera receives videos as input
<bitmapper>
i contacted lispworks support, they said no to getting a version that will work on IRIX
<bitmapper>
hmm
<MichaelRaskin>
(and synchronised seek through SVG animation + audio is not a well-supported task in modern software, as far as I know)
frgo_ has joined #lisp
CrazyPython has joined #lisp
frgo has quit [Ping timeout: 244 seconds]
KDr24 has joined #lisp
KDr23 has quit [Ping timeout: 260 seconds]
thodg has quit [Remote host closed the connection]
karlosz has quit [Quit: karlosz]
heisig has joined #lisp
Kundry_Wag has quit [Remote host closed the connection]
Kundry_Wag has joined #lisp
mangul has quit [Ping timeout: 260 seconds]
man213 has joined #lisp
man213_ has joined #lisp
karlosz has joined #lisp
emys has quit [Ping timeout: 260 seconds]
frgo has joined #lisp
EvW1 has quit [Ping timeout: 272 seconds]
rand_t has joined #lisp
frgo_ has quit [Ping timeout: 246 seconds]
edgar-rft has quit [Ping timeout: 260 seconds]
Adamclisi has quit [Ping timeout: 240 seconds]
froggey has quit [Ping timeout: 240 seconds]
duncan_ has quit [Ping timeout: 260 seconds]
cmatei has quit [Ping timeout: 240 seconds]
jprajzne has quit [Ping timeout: 265 seconds]
dilated_dinosaur has quit [Ping timeout: 260 seconds]
jprajzne has joined #lisp
CrazyPython has quit [Ping timeout: 240 seconds]
frgo_ has joined #lisp
man213 has quit [Ping timeout: 260 seconds]
frgo has quit [Ping timeout: 240 seconds]
edgar-rft has joined #lisp
duncan_ has joined #lisp
schjetne has quit [Remote host closed the connection]
efm has quit [Quit: Konversation terminated!]
schjetne has joined #lisp
dilated_dinosaur has joined #lisp
froggey has joined #lisp
cmatei has joined #lisp
man213 has joined #lisp
Adamclisi has joined #lisp
EvW has joined #lisp
rixard has joined #lisp
Fare has joined #lisp
rixard has quit [Ping timeout: 260 seconds]
rixard has joined #lisp
milanj has joined #lisp
|Pirx| has joined #lisp
rixard has quit [Ping timeout: 244 seconds]
man213 has quit [Ping timeout: 260 seconds]
ym has quit [Quit: Leaving]
hiroaki has quit [Ping timeout: 260 seconds]
<pve>
pfdietz: this mutation testing sounds interesting.. do you know if it is widely used?
<pfdietz>
It was invented decades ago (in Lisp!) but was mostly in fringe things like flight control software until recently. Now Google is using it as input to code reviews.
<pve>
there's no lisp testing framework that does that now, right?
brutalist has joined #lisp
<pfdietz>
Well, none that's been made public yet. I have a prototype working.
hiroaki has joined #lisp
<Aurora_v_kosmose>
How does mutation testing differ from formal modeling?
<Aurora_v_kosmose>
(Haven't read the linked paper, busy)
<pve>
and your prototype will become public at some point?
<markasoftware>
a lisp code optimizer could be written as a macro, right?
<pfdietz>
Very different. It's basically a way of estimating the adequacy of a test suite.
<phoe>
markasoftware: yes, although it *should* be written as a compiler macro instead.
<phoe>
Or, if possible, as an inlineable function.
<Aurora_v_kosmose>
pfdietz: Oh I see. That sounds like a rather different goal.
<pfdietz>
It's testing of testing.
<markasoftware>
is a compiler macro just a macro that's part of the compiler?
<phoe>
you can define a compiler macro for a function, that's a difference.
<phoe>
you can't do the same with standard macros.
shifty has quit [Ping timeout: 260 seconds]
<Aurora_v_kosmose>
pfdietz: Ah, got it. Thanks.
<Bike>
it is not a macro that's part of the compiler, though the compiler can use them.
shifty has joined #lisp
<saturn2>
a compiler macro is only for optimizing a function call though, not any lisp code in general
MerlinTheWizard has quit [Ping timeout: 244 seconds]
gigetoo has quit [Ping timeout: 240 seconds]
pfdietz has quit [Remote host closed the connection]
<markasoftware>
do you mainly write compiler macros for your own functions?
pfdietz has joined #lisp
makomo has quit [Quit: WeeChat 2.7.1]
leo_song has quit [Quit: ZNC 1.7.2+deb3 - https://znc.in]
<saturn2>
yes
leo_song has joined #lisp
<markasoftware>
What power does it give you that you wouldn't get from just writing your original function in a more performant way?
karlosz has quit [Quit: karlosz]
<SAL9000>
special-casing based on arguments
<SAL9000>
(or global state)
<phoe>
compile-time type inference
MerlinTheWizard has joined #lisp
<edgar-rft>
Common Lisp - build with alien technology and tested by mutants
<markasoftware>
ah i see
<markasoftware>
thank you!
man213 has joined #lisp
man213 has quit [Client Quit]
gigetoo has joined #lisp
clintm has joined #lisp
clintm has quit [Changing host]
clintm has joined #lisp
<cgay>
I've found examples of define-compiler-macro for collection-related facilities (remove, find, etc.). They basically transform those calls at compile time into more efficient internal calls that don't use keyword arguments.
<pve>
pfdietz: in that paper, lisp has a suspiciously low survival rate (1%).. what could explain it? (they don't seem to give an explanation)
<phoe>
I posted it once on twitch chat during a lunch break but might have easily been missed
<Shinmera>
I did miss that yes
milanj has joined #lisp
z147 has quit [Quit: z147]
schjetne has quit [Ping timeout: 260 seconds]
<phoe>
there should be a special ELS prize for providing aesthetic input for the conference
<phoe>
right now I am torn between heisig raising the bar for the quality of the visual medium and Shinmera defining the quality of the European Lisp Symposium OST
<Shinmera>
lol
<karlosz>
what is the type? is it a german or gothic one?
<selwyn>
chill beats to lisp to
Lycurgus has joined #lisp
femi has quit [Ping timeout: 265 seconds]
MerlinTheWizard has joined #lisp
tburdick has joined #lisp
* heisig
blushes
tburdick has quit [Remote host closed the connection]
tburdick has joined #lisp
<phoe>
karlosz: neither, it's my custom invention that shamelessly mixes gothic texture, fraktur, humanist minuscule and a few other stuff into one rather freeform writing style
<phoe>
that's already #lispcafe matter though
tburdick has quit [Remote host closed the connection]
<|3b|>
(though the ccl and ccl32 there are actually sbcl, since roswell got confused by ccl package names and i didn't add a workaround in hopes it would be fixed soon)
<Shinmera>
|3b|: what about allegro?
<axion>
by we he means me also :)
<Lycurgus>
how complete is clasp as a cl at this point if current sbcl was 100%?
<axion>
Allegro is in the actgions link above
<|3b|>
Shinmera: thats in the CI
<Shinmera>
oh. I missed it, my bad
* Shinmera
is tired
<Shinmera>
Lycurgus: it's quite complete features wise, but still has rough edges
<|3b|>
and they are looking at the bug pngload found (which apparently only affects linux express edition, not windows or full version)
<pfdietz>
I liked the presentations, except when lips got out of sync with the sound. Uncanny valley.
schjetne has joined #lisp
<Lycurgus>
Shinmera, so equivalent use wise to sbcl?
* |3b|
will probably add clasp to the CI once it works and i build some binaries the CI can use
<|3b|>
and windows ccl once ros figures out how to load it again, since i think 1.12 has the fix for windows hangs
<Shinmera>
Lycurgus: I don't know what you mean by that
<Lycurgus>
eg, I would say sbcl was 70% of acl to talk in a very loose manner
<Shinmera>
I still don't know what that means
<Lycurgus>
k
pve has quit [Quit: leaving]
* Lycurgus
will wait for fewer "rough edges"
SGASAU` has quit [Remote host closed the connection]
SGASAU` has joined #lisp
Oskidosw has joined #lisp
Oskidosw has quit [K-Lined]
Dreaskeity has joined #lisp
Dreaskeity has left #lisp [#lisp]
milanj has quit [Quit: Leaving]
<p_l>
Preventive censorship commencing
<p_l>
(yes, fresh spam masquerading as freenode server ops)
<p_l>
Xach: same as last time, just without telling you to login to efnet
<p_l>
I hope I did the right mode this time (/mode #lisp +m $-a
tburdick has joined #lisp
Cairn[m] has joined #lisp
<p_l>
/mode #lisp +q $~a should work... dunno how I mixed up tilde and minus
<Cairn[m]>
Anyone have a set of timestamps for the ELS recording?
<Cairn[m]>
Or will I just have to be patient for the videos to be uploaded?
<p_l>
Cairn[m]: the vids should already be available as separate clips on twitch
SGASAU` has quit [Remote host closed the connection]
<p_l>
ehhh, I seem to fail at twitch ultimately, they aren't
<p_l>
it just looked like it
SGASAU` has joined #lisp
<Cairn[m]>
Yeah, I only see the full VODs
<Shinmera>
I think the clips are mostly people being confused by the twitch UI
<Shinmera>
the timestamps should be easy enough to figure out from the programme.
<Cairn[m]>
Yeah, I've just settled to watching it in order, but I just wanted to ask in case anyone had done timestamps yet.
<Cairn[m]>
Thanks!
<LdBeth>
Should be vaporwave
schjetne has quit [Ping timeout: 244 seconds]
MerlinTheWizard has quit [Remote host closed the connection]
Lycurgus has quit [Quit: Exeunt]
cosimone has quit [Quit: Quit.]
SGASAU` has quit [Quit: ERC (IRC client for Emacs 26.3)]
shifty has quit [Ping timeout: 260 seconds]
MerlinTheWizard has joined #lisp
shifty has joined #lisp
SGASAU has joined #lisp
krid has joined #lisp
tburdick has quit [Ping timeout: 265 seconds]
dddddd has quit [Ping timeout: 260 seconds]
man213__ has joined #lisp
man213_ has quit [Ping timeout: 264 seconds]
gravicappa has quit [Ping timeout: 264 seconds]
keep_learning has joined #lisp
man213__ has quit [Quit: Going offline, see ya! (www.adiirc.com)]
pjb has quit [Ping timeout: 265 seconds]
sjl_ has quit [Ping timeout: 246 seconds]
Adamclisi has quit [Ping timeout: 244 seconds]
froggey has quit [Ping timeout: 260 seconds]
dilated_dinosaur has quit [Ping timeout: 260 seconds]
duncan_ has quit [Ping timeout: 260 seconds]
holycow has joined #lisp
random-nick has quit [Ping timeout: 260 seconds]
dilated_dinosaur has joined #lisp
froggey has joined #lisp
X-Scale` has joined #lisp
Adamclisi has joined #lisp
|Pirx| has quit [Remote host closed the connection]
|Pirx| has joined #lisp
<markasoftware>
in SLIME, waht's the shortcut ot access the result of the last expression?
<markasoftware>
i seem to remember there was something simple
X-Scale has quit [Ping timeout: 256 seconds]
X-Scale` is now known as X-Scale
<jasom>
markasoftware: *
<markasoftware>
huehuehue thank you
<jasom>
markasoftware: and that should work in any REPL, not just slime
<|3b|>
in slime repl, you can hit enter on previous printed results to get them as input
<jasom>
* ** ***, / // ///, + ++ +++ are respectively the first/second/third most recent primary-value, values as a list, forms
* |3b|
isn't sure exactly when that preserves identity or not though, so might need to use * when that matters
<jasom>
and - is the form currently being evaluated in the repl, though I haven't used that one before.
<anlsh>
So what's the purpose of the alexandria-2 package?
papachan has joined #lisp
MerlinTheWizard has quit [Ping timeout: 244 seconds]
SGASAU has quit [Remote host closed the connection]
SGASAU has joined #lisp
pjb has joined #lisp
toorevitimirp has quit [Remote host closed the connection]
SGASAU has quit [Remote host closed the connection]
|Pirx| has quit [Remote host closed the connection]
SGASAU has joined #lisp
<pfdietz>
To not break systems that :use Alexandria by adding new exported symbols to that package.
<|3b|>
Xach: does quicklisp find .asd files in subdirectories of projects?
<Xach>
|3b|: yes
<|3b|>
ok, so would need to file an issue to avoid that if needed?
lxbarbosa has joined #lisp
* |3b|
is considering some test systems that would be intentionally broken for example
<Xach>
to ignore a system file? like for a template asd file or something?
<Xach>
yeah, i can work with that.
tburdick has joined #lisp
<|3b|>
ok, will do that if i end up adding them to a project in ql
brutalist has quit [Ping timeout: 240 seconds]
Oladon has joined #lisp
froggey has quit [Ping timeout: 244 seconds]
Adamclisi has quit [Ping timeout: 246 seconds]
dilated_dinosaur has quit [Ping timeout: 260 seconds]
froggey has joined #lisp
Adamclisi has joined #lisp
MerlinTheWizard has joined #lisp
ArthurStrong has quit [Quit: leaving]
Kundry_Wag has quit [Ping timeout: 240 seconds]
pfdietz has quit [Remote host closed the connection]
wxie has joined #lisp
rgherdt has quit [Ping timeout: 272 seconds]
rixard has joined #lisp
Lord_of_Life_ has joined #lisp
buffergn0me has quit [Ping timeout: 240 seconds]
Lord_of_Life has quit [Ping timeout: 260 seconds]
Lord_of_Life_ is now known as Lord_of_Life
rixard has quit [Ping timeout: 246 seconds]
dilated_dinosaur has joined #lisp
shifty has quit [Ping timeout: 264 seconds]
MerlinTheWizard has quit [Ping timeout: 244 seconds]
shifty has joined #lisp
<Xach>
hmm
schjetne has joined #lisp
<anlsh>
Making a new package only kicks the can down the road though doesn't it?
<Xach>
anlsh: how so?
<anlsh>
What would stop them from having to make an alexandria-3 years down the road
<Xach>
anlsh: Nothing - but the way you phrase the question suggests you think that is a problem?
<Xach>
people who want new things get new things, people who use the old do not break - seems ok to me?
<aeth>
it sounds like it would be better to build versioning into DEFPACKAGE so people who don't USE it can always go with the latest version, but people who USE it can lock to a version.
tburdick has quit [Ping timeout: 265 seconds]
<anlsh>
It just doesn't seem like a good solution, having to add a new package every time you'd like to export a new symbol or set of symbols
torbo has joined #lisp
<Xach>
anlsh: why?
<aeth>
the real solution is to not USE packages that aren't your own
<aeth>
you can IMPORT-FROM a symbol or two that you actually use, or you can give it a local nickname
choegusung has quit [Quit: leaving]
<Xach>
Packages may change what they provide even if you import selectively, or reference with prefixes.
<Xach>
Committing to a set of supported symbols is one option and it seems reasonable to me.
<|3b|>
having things break randomly depending on what you load (because someone defined a function that clobbered a function they don't know about from a package they :use) doesn't seem like a good solution either :/
remexre has quit [Quit: WeeChat 2.7.1]
wxie has quit [Ping timeout: 244 seconds]
* |3b|
agrees with "don't :use packages" as real solution, but suspects it is also an unlikely one, particularly for old, mostly unmaintained projects
<|3b|>
(but which are otherwise perfectly fine, as long as their :USEd packages don't change)
Kundry_Wag has joined #lisp
<aeth>
The real fun thing is that if you use FLET, you can break macros by (perhaps locally) redefining functions you're not even aware of if you USE.
Kundry_Wag has quit [Remote host closed the connection]
Kundry_Wag has joined #lisp
zooey has quit [Ping timeout: 240 seconds]
zooey has joined #lisp
<anlsh>
I agree that not :use'ing things is the best solution: what about including some sort of disclaimer that new symbols may be exported at any time?
<anlsh>
You can't stop people from :use'ing it, but it would at least make the breakages their fault
<Xach>
Where do you put the disclaimer?
<|3b|>
"their fault" doesn't help 3rd parties who get hit by it due to some dependency 3 levels down :p
<anlsh>
Well you couldn't do it with existing libraries obviously :|
<MichaelRaskin>
Ship with the dictionary and export random additional symbols on each import?
<anlsh>
but for instance, doing it with a new package would be feasible
<MichaelRaskin>
Well, blame-shifting doesn't help
<|3b|>
yeah, hoping for "don't :use this" to work is a bit more reasonable now that more lisps support p-l-n
<MichaelRaskin>
Exporting T and NIL not eq to CL:T and CL:NIL might, actually