<ivan>
annoyed at chromium for not publishing release tarballs when chrome is released
__monty__ has quit [Quit: leaving]
<Ericson2314>
@elvishjerricco pong
<Ericson2314>
But on phone waiting for train
<elvishjerricco>
Ericson2314: Well I was gonna ask how to make a cross target in nixpkgs with patched GCC and binutils so I could mess with the redox kernel, but I gave up when carnix couldn't handle relibc :P
<Ericson2314>
Bummer carnix couldn't
<Ericson2314>
You probably want that rust cross PR to land first too
<elvishjerricco>
Oh right. Thought it already had
<elvishjerricco>
Carnix was just struggling with a) git deps, and b) nested path deps
<andi->
yeah :/
<elvishjerricco>
Ericson2314: And my first goal was to get relibc working on Linux first, so I could have pkgsCross.linuxRelibc
<Ericson2314>
Oh interesting
<Ericson2314>
You could probably knock off that PR though
<Ericson2314>
I'm kinda busy af recently
<Ericson2314>
It just needs a little more
aleph- has quit [Ping timeout: 250 seconds]
hedning has quit [Quit: hedning]
drakonis has quit [Quit: WeeChat 2.3]
aleph- has joined #nixos-chat
lassulus_ has joined #nixos-chat
lassulus has quit [Ping timeout: 245 seconds]
lassulus_ is now known as lassulus
pie___ has joined #nixos-chat
pie__ has quit [Ping timeout: 240 seconds]
aleph- has quit [Ping timeout: 246 seconds]
endformationage has quit [Quit: WeeChat 2.3]
jasongrossman has quit [Quit: ERC (IRC client for Emacs 26.1)]
jasongrossman has joined #nixos-chat
iqubic has joined #nixos-chat
<jD91mZM2>
sphalerite, __monty__, elvishjerricco: Redox OS does support C++ but as far as I'm aware nobody has tried to port Nix. For a while it didn't have support for pthreads but now with the relibc standard library it finally does. Trying to build Nix on the latest master of Redox would be cool (can't release because of a blocking network issue that randomly occurs). The "everything is a URL" idea is actually
<jD91mZM2>
really cool. Daemons can provide their own special file handling, such as ipcd:hello (where hello isn't an actual file, sort of like tmpfs), event: (so the whole epoll system can be simplified to just opening a file), mem: (for checking memory status and mmap over memory). I recommend reading https://dev.to/legolord208/programming-for-redox-os-4124 for a full explanation of just how awesome schemes are.
<jD91mZM2>
The whole `socket` syscall family? Schemes. The whole `epoll` system over memory (NULL fd and a special flag)? Just map over a scheme. Sleeping for a duration asyncronously? Schemes.
<jD91mZM2>
Oh and /dev/null or /dev/zero special files? null: and zero:. IIRC there's no daemon to handle http/https, so calm down, it's not that high level
<sphalerite>
but there could be, right
<sphalerite>
?
<elvishjerricco>
seems like it
<elvishjerricco>
could even be an ssh scheme to act like sshfs or scp. Though I'm not sure how you'd sanely handle the auth? Can multiple daemons add resources to the same scheme?
<jD91mZM2>
There could indeed be a scheme for all those things, but programs couldn't depend on them being installed. My only solution for the sshfs idea is writing some auth info struct to `sshfs:` without specifying a path. Or perhaps you start the daemon manually with the auth info in command line arguments or a GUI
<jD91mZM2>
Schemes can communicate like any other process can communicate with schemes. A scheme dependant on a ssh authentication mechanism or password store could query that scheme for the details or something before starting up.
<elvishjerricco>
jD91mZM2: Yea, like sshfs, I think it'd be best to make it a process that you start manually and auth at start time. The issue is in allowing multiple of these daemons to run at a time, so that you can use sever remote servers
<elvishjerricco>
s/sever/several/
<jD91mZM2>
Indeed that's an issue. We have the same issue with filesystems and stuff like that. I think the current idea is the little hacky starting sshfs1 and stuff, and supporting inter-scheme symlinks so you can have a directory act like a mount. Not sure how this will be solved.
jasongrossman has quit [Ping timeout: 250 seconds]
iqubic has quit [Ping timeout: 240 seconds]
jasongrossman has joined #nixos-chat
jackdk has quit [Ping timeout: 246 seconds]
__Sander__ has joined #nixos-chat
globin has quit [Remote host closed the connection]
__monty__ has joined #nixos-chat
__monty__ has quit [Quit: Lost terminal]
__monty__ has joined #nixos-chat
tilpner has quit [Quit: WeeChat 2.3]
endformationage has joined #nixos-chat
drakonis has joined #nixos-chat
Synthetica has joined #nixos-chat
jasongrossman has quit [Ping timeout: 246 seconds]
andi- has quit [Ping timeout: 250 seconds]
andi- has joined #nixos-chat
__Sander__ has quit [Quit: Konversation terminated!]
drakonis has quit [Quit: WeeChat 2.3]
<gchristensen>
sphalerite: all my EU friends are sick :(
<gchristensen>
y'all have to stay healthy
<Synthetica>
__monty__: There are group invite links, but no public register
<Synthetica>
Also, groups are limited to 256
<__monty__>
This is still about the whatsapp stuff, Synthetica?
<Synthetica>
Yes
<__monty__>
That was days ago! But thanks though.
<Synthetica>
Oh, I see now, hadn't been on in the mean time
<__monty__>
I wonder how they'll go about syncing OTR conversations though.
<__monty__>
Also, doesn't facebook messenger send everything to facebook even before you hit return? That'd defeat the e2e completely, right?
<gchristensen>
that was my point when we were discussing it last
<gchristensen>
doesn't matter if the chat is e2e, if the ends are sharing the info
<gchristensen>
and, their business isn't exactly privacy
drakonis has joined #nixos-chat
aleph- has joined #nixos-chat
aleph- has quit [Ping timeout: 245 seconds]
jasongrossman has joined #nixos-chat
ivan has quit [Quit: lp0 on fire]
harrow has quit [Ping timeout: 245 seconds]
harrow has joined #nixos-chat
ivan has joined #nixos-chat
<lejonet>
I wouldn't trust whatsapp, even before facebook bought em, further than I can throw em
<joepie91>
the more I'm reading the documentation of libsignal-protocol-javascript, the less surprised I am about the security issues with Signal Desktop...
<joepie91>
what a mess
<lejonet>
whaat, javascript causing a mess? That has never happened before... :P
<joepie91>
lejonet: no, it's not JS causing a mess, it's developers who clearly have no fucking clue what they're doing...
<joepie91>
there are *so many problems* with the example code in that repo alone
<gchristensen>
scary
<lejonet>
joepie91: that is not uncommon when JS is involved in my experience
<lejonet>
Yes, I know, I'm overly pessimistic about JS and devs using it, but I'm a bit bitter after maintaining a nodejs site for 3-4 years
<joepie91>
lejonet: in my experience that largely limits itself to the subset of "JS developers" who aren't actually JS developers but who normally work in a different language, don't consider JS a 'real' language, and therefore don't bother learning how to use it correctly
<lejonet>
joepie91: indeed true
<joepie91>
this particular kind of crap at least
<joepie91>
there's also the subset of "lolol move fast and break things" idiots but that's a - while vocal - pretty small subset
<lejonet>
Thankfully
<joepie91>
unfortunately they are influential enough that well-meaning developers pick up bad habits from them...
* joepie91
angry glare at MongoDB
<lejonet>
Imagine then this: A PHP developer that "picked up" JS... that is the code I've been maintaining
<joepie91>
ah yes
<joepie91>
well, it could be worse
<lejonet>
indeed, it could be actual PHP, or COBOL
<gchristensen>
ouch
<joepie91>
lejonet: topping the charts for "writes crappy JS" are Java, C++ and .NET developers
<joepie91>
in my experience anyway
<drakonis>
i wrote php once, every second of it was truly agony
<drakonis>
it wasn't fun
<lejonet>
joepie91: I can imagine that, I think the "lack of structure" flabergasts them
<joepie91>
though on C# there's a weird divide where like one half of the C# devs uses JS very well, and the other half makes a total mess of it
* gchristensen
was a full time PHP dev for many years
<joepie91>
correlates to how functional the code is that they write in C#, it seems
<gchristensen>
the people who develop in PHP don't deserve the disdain
<joepie91>
lejonet: that's one part of it; the other part is specifically JS not being class-based
<joepie91>
I have had people try to die on the hill that a language that doesn't use classes is not a real language
<drakonis>
gchristensen: they don't, but the language is kind of terrible to deal with
<lejonet>
gchristensen: I know, I've meet, and know, plenty of sane PHP developers, and I would argue that the disdain PHP usually get (at least from me) is from PHP's past by now
<jasongrossman>
I was a full time COBOL dev! A while ago.
<joepie91>
and they are so stuck on that viewpoint that they genuinely do not understand the concept of free-standing objects in JS...
<lejonet>
jasongrossman: do you still have your fingers intact?
<joepie91>
lejonet: anyway, all things considered, I've found that most PHP devs who move to JS are willing to learn, they just have years of bad habits to get rid of
<joepie91>
but PHP devs are surprisingly undogmatic
<jasongrossman>
lejonet: I don't understand.
<lejonet>
huuuuge frikin disclaimer from my part: I'm not a programmer by trade, but by hobby, so take my comments with a proper amount of amateurness
<joepie91>
unfortunately, when there's nobody to point out the bad habits... you get crap JS still
<lejonet>
jasongrossman: the joke I've heard about COBOL is "COBOL fingers" which is synonym with the verboseness of the language causing pain in hands and fingers
<lejonet>
joepie91: indeed, and that is the result of the code I've got, luckily I hopefully get to put that code to the archives on friday
<jasongrossman>
lejonet: Thanks. I hadn't heard that.
<gchristensen>
joepie91: s/JS //
<jasongrossman>
lejonet: As I'm sure you agree, verboseness isn't a TYPING problem. It's a READING problem. And yeah, COBOL is pretty painful to read.
<joepie91>
gchristensen: I can only authoratively speak for JS :)
<jasongrossman>
s/verboseness/verbosity
<lejonet>
jasongrossman: indeed
<joepie91>
gchristensen: insufficient data on other languages
<lejonet>
copy-paste exist for example :)
<gchristensen>
joepie91: leave someone unattended and learning in any environment and they'll produce crap.
<joepie91>
not necessarily
<gchristensen>
oh
<joepie91>
it really depends on where they end up in terms of learning materials
<lejonet>
gchristensen: s/crap/less than optimal code at best, crap at worst/
<joepie91>
and more importantly, how forgiving the tools are of crap
<joepie91>
(and I don't *just* mean the language)
<joepie91>
I strongly suspect that a big factor in the crapness of a lot of PHP is that the path of least resistance is to shoehorn PHP into your HTML, which makes it effectively impossible to add structure later on
<jasongrossman>
lejonet: I'm not sure the mainframe I was working on did have copy and paste, actually!
<joepie91>
because that's how people learn to start writing PHP
<joepie91>
and so they get used to the mess
<lejonet>
I wish I had the insight to prod the programmer in question in more of the right direction at the time, but then, I wasn't much of a better JS programmer than him either honestly (he was a sane programmer, just inexperienced with the language)
<lejonet>
jasongrossman: oh right, mainframes might not have such luxuries
* lejonet
wishes he actually got to experience maintaining a mainframe or such
<joepie91>
lejonet: btw, for future reference, if you run into people producing poor Node.js code, *especially* those coming from other languages, please direct them to #Node.js and tell them to ask for a code review :)
<lejonet>
mainly for the insight, I've heard that it was... special
<jasongrossman>
lejonet: It was just about acceptable if you were getting paid for it. It was very very boring.
<joepie91>
lejonet: there's a pretty decent community of friendly-but-competent regulars there
<lejonet>
jasongrossman: hehe, I can imagine that, not much "new features" being developed I take it?
<joepie91>
(who have gotten used to un-teaching bad habits from other languages)
<lejonet>
joepie91: good to know for the future :)
<jasongrossman>
lejonet: Right, especially since they were (and are, I guess) used in systems that have really paranoid testing regimes, which means that changing anything fundamental is very expensive.
<lejonet>
We'll see if I take any such projects in the foreseeable future, I'm not relying on my own company for salary nowdays
<joepie91>
I'm actually pretty happy with how #Node.js turned out
<jasongrossman>
lejonet: I was programming payroll systems.
<joepie91>
over the past few years
<lejonet>
jasongrossman: ah, yeah, then you want 300% test coverage, because payroll processing is NOT allowed to go wrong
<jasongrossman>
lejonet: Right. More time spent testing than everything else put together.
<joepie91>
it's not without its problems, but overall it's struck a good happy middle between "straightforward and unsugarcoated feedback" and "friendly attitude"
<lejonet>
joepie91: any community of any considerable size have its problems :)
<lejonet>
jasongrossman: I'm not surprised
<joepie91>
aye
<joepie91>
lejonet: we're actually cheating a bit I think, ##javascript seems to get the brunt of the abuse
<lejonet>
joepie91: hehe
<joepie91>
I'm not entirely sure why or how, but it appears to act as a first-line filter
<lejonet>
joepie91: maybe because people think "oh, this is js, lets ask the js people first" instead of jumping at nodejs people directly
<joepie91>
lejonet: dunno, a lot of people still end up directly in the Node channel
<lejonet>
mindboggling mystery!
<joepie91>
I'm wondering if there's maybe some hidden causation somewhere in where the channels are documented, and what kind of people are likely to end up in those points of documentation...
<lejonet>
Maybe
<jasongrossman>
I know I'm disinclined to try out new channels on IRC when I have a problem, because IMO it's important to find out the local culture before asking questions, and that can take ages (unless the channel has easily available logs). So it matters what order I discover channels in.
<__monty__>
jasongrossman: I find that how a channel handles noobish questions is a good litmus test for the quality of the community.
<jasongrossman>
__monty__: That's a good thought.
<lejonet>
__monty__: that's what I've found too
<lejonet>
__monty__: and how they handle if someone bashes someone that asks a noobish question
tertl3 has joined #nixos-chat
jackdk has joined #nixos-chat
endformationage has quit [Quit: WeeChat 2.3]
endformationage has joined #nixos-chat
aleph- has joined #nixos-chat
<etu>
lejonet: And that's just one of the reasons why NixOS is amazing <3
jasongrossman has quit [Ping timeout: 250 seconds]