gchristensen changed the topic of #nixos-chat to: NixOS but much less topical || https://logs.nix.samueldr.com/nixos-chat
<__monty__> nn, peoples
<infinisil> __monty__: Yeah so that argument seems really weak to me
<__monty__> Not *that* weak. "Oh, so you're a totalitarian gov that wants to snoop on dns? Just block port 865 (or whatever the DoT port is)."
<__monty__> Vs, "Oh ... snoop on dns? Just block port 443 and break the internet for everyone."
<infinisil> Yeah, but it's got nothing inherently to do with the protocol
<infinisil> Oh, I guess because it does go over HTTPS, which has this fixed port
<infinisil> Maybe it is inherent to the protocol
<infinisil> Yeah
<infinisil> Unless you can change these ports (why would you not be able to change them anyways)
<__monty__> Sure but you can't seperate "normal" HTTPS traffic from DoH traffic is the point.
<infinisil> Yeah I get the argument now
<infinisil> Not a bad one
<__monty__> Yep, but the layer violation tho.
<__monty__> It's like throwing away TCP for QUIC. TCP *is* the internet almost.
<infinisil> IIRC, QUIC is like an alternate TCP with more smarts
<infinisil> I don't think that's a bad thing, we don't need to be stuck with UDP and TCP forever
<__monty__> Yeah but more anything doesn't imply more better.
<infinisil> I still hate Google though lol
<infinisil> __monty__: Night :)
<__monty__> nn, thanks for the discussion!
<infinisil> Likewise
<infinisil> gchristensen: Ah yes, one more thing not to touch
<joepie91> lol
<gchristensen> well ... the thing is ... I can imagine USeful Purposes like signing
<gchristensen> or secrets!
<joepie91> gchristensen: does docbook have a "do not touch" tag
<joepie91> :P
<gchristensen> but definitely not something to use LIGHTLY.
<infinisil> gchristensen: Oh yeah, could be very interesting for secrets :O
lassulus has quit [Ping timeout: 240 seconds]
lassulus has joined #nixos-chat
__monty__ has quit [Quit: leaving]
<samueldr> these are not the kind of strings you want to see in a router management thingy https://stuff.samueldr.com/screenshots/2019/02/20190204192714.png
<samueldr> hopefully it asks for a password
<gchristensen> O.o
<samueldr> it's likely it does, looking at things using "dropbear" in the tp-link app
<samueldr> so it might just be as simple as ssh
jackdk has quit [Ping timeout: 245 seconds]
<samueldr> and it apparently uses `jsch` http://www.jcraft.com/jsch/
lassulus has quit [Ping timeout: 246 seconds]
lassulus has joined #nixos-chat
cransom has quit [Ping timeout: 250 seconds]
jackdk has joined #nixos-chat
<joepie91> lejonet: infinisil: elvishjerricco: progress update... https://strange-cougar-75.localtunnel.me - basically nothing works yet except for the conversation tabs, but pretty much the entire state layer is there already, it's not just a static mockup :P
<joepie91> (sorry, you're all on my chat project mailing list now!)
<infinisil> joepie91: Is this supposed to take like 5 seconds to load?
<joepie91> infinisil: yeah, blame localtunnel
<infinisil> Ah
<joepie91> it's proxied to my local PC but their proxy is excruciatingly slow and I'm not minifying any JS and it all comes with sourcemaps etc.
<samueldr> yeah, was going to say "at a glance it looks slim, must be hard to squeeze it through joepie91's connection"
<joepie91> lol
<joepie91> my part of the connection ain't the problem :D
<joepie91> though it *is* a little slow right now, slower than normally...
<samueldr> that's poorly tabbable (the tab key) :) but I guess you know of it and it's not a priority right now
<joepie91> samueldr: aye, I know
<joepie91> samueldr: https://www.speedtest.net/result/8014561049 -- if I can't serve a bundle.js through that, something is really very wrong
<joepie91> :P
<samueldr> are you sure you know JavaScript? ;)
<samueldr> better upload than download?
<infinisil> Argh why does everything need javascript in the web
<infinisil> I can't even watch the speedtest result without enabling js
<samueldr> is it because you're living under sea level?
<infinisil> Lol
<samueldr> here it's generally the other way around
<gchristensen> nix-shell -p speedtest-cli --run speedtest
<infinisil> gchristensen: Like the link joepie91 posted
<gchristensen> speedtest-cli does the same thin
<infinisil> joepie91: :D
<joepie91> samueldr: I'm /supposed/ to have 200/200, and normally I get 200/260... it's been that way since they upgraded me to 160/160 a while back, so I think they typoed the upload cap and it got stuck
<infinisil> Damn, I'm jealous of that speed
<joepie91> samueldr: before that, I got a neatly capped 100/100
<joepie91> but it's a bit slow today, normally I'm between 180-200mbps, not 160
<samueldr> just noting here: if you're ever wanting to play with a decompiled android apk, using jadx I was able to output passable java (without names) and then exporting it as a java project intellij IDEA (community) can use; and then I can use `refactor` and `find usage` from IDEA
cransom has joined #nixos-chat
<joepie91> infinisil: this is actually not even the fastest plan :P
<joepie91> but there's no point in upgrading to 500/500 given that my HDDs can barely keep up with the speed I have now
<joepie91> infinisil: but yeah.... FttH is quite nice, it really annoys me that it isn't getting rolled out in more places
<joepie91> it gets quite close to "internet as it should be"
<infinisil> joepie91: Oh no, the more people have such a connection, the more bloated javascript websites we'll see!
<joepie91> whoa. what? OVH has an australian location now?
<joepie91> ah, not unmetered :P
<joepie91> infinisil: unlikely :P
<joepie91> infinisil: resource demands of web-based stuff scale specifically with the connections that the developers have
<joepie91> (I wish I was kidding)
<infinisil> Yeah, but the more people have such a connection, the more devs have such a connection
<joepie91> infinisil: devs already have that connection :P
<infinisil> devs are people too after all! I think..
<iqubic> Download: 37.28 Mbit/s
<joepie91> companies generally have comparable-speed internet
<iqubic> Upload: 6.34 Mbit/s
<joepie91> tech companies that is
<infinisil> Ah yeah
<infinisil> So it's even worse, damnit
<joepie91> infinisil: my favourite rant on this topic: https://eev.ee/blog/2016/03/06/maybe-we-could-tone-down-the-javascript/
<infinisil> Haven't fully read it, but that's a nice post
<samueldr> it's oddly amusing to do RE on java code, like crosswords
<samueldr> or that picross game
jasongrossman has joined #nixos-chat
<iqubic> I find Regexes in general to be a pain.
iqubic has quit [Remote host closed the connection]
jasongrossman has quit [Quit: ERC (IRC client for Emacs 26.1)]
nodyn has joined #nixos-chat
drakonis has quit [Quit: WeeChat 2.3]
jasongrossman has joined #nixos-chat
<samueldr> sphalerite: without a proper dump, I cannot verify something weird in the code; the name of the packet type is "TDP" in the code, unsure how it should act in a bi-directional way... the headers you have are of this format AFAICT from the code https://gist.github.com/samueldr/7aecccfdd20c1480b13f8bf92cb33a16
<samueldr> except for length and checksum, it is unclear what the others are
<samueldr> sn is a name that came from debug log information in the apk
<samueldr> I'm thinking it might be serial number, but not as in device number, but as in Packet N; the device ID is discovered inside the data of such packets
nodyn has quit [Ping timeout: 246 seconds]
<samueldr> (the data is also AFAICT in network byte order; FWIW
lassulus_ has joined #nixos-chat
lassulus has quit [Ping timeout: 268 seconds]
lassulus_ is now known as lassulus
pie__ has joined #nixos-chat
pie___ has quit [Ping timeout: 240 seconds]
jackdk has quit [Ping timeout: 240 seconds]
endformationage has quit [Quit: WeeChat 2.3]
jasongrossman has quit [Quit: ERC (IRC client for Emacs 26.1)]
jasongrossman has joined #nixos-chat
jasongrossman has quit [Client Quit]
jasongrossman has joined #nixos-chat
lassulus has quit [Quit: WeeChat 2.2]
jasongrossman has quit [Ping timeout: 272 seconds]
jasongrossman has joined #nixos-chat
__Sander__ has joined #nixos-chat
__monty__ has joined #nixos-chat
lassulus has joined #nixos-chat
hedning has quit [Quit: hedning]
Jackneill has quit [Ping timeout: 240 seconds]
LnL has quit [Ping timeout: 245 seconds]
Guest5079 has joined #nixos-chat
hedning has joined #nixos-chat
Jackneill has joined #nixos-chat
Jackneill has quit [Remote host closed the connection]
Jackneill has joined #nixos-chat
nodyn has joined #nixos-chat
nodyn has quit [Read error: Connection reset by peer]
hedning has quit [Quit: hedning]
drakonis has joined #nixos-chat
endformationage has joined #nixos-chat
hedning has joined #nixos-chat
aminechikhaoui has quit [Quit: The Lounge - https://thelounge.github.io]
aminechikhaoui has joined #nixos-chat
Guest5079 is now known as LnL
drakonis has quit [Quit: WeeChat 2.3]
hedning has quit [Quit: hedning]
__Sander__ has quit [Quit: Konversation terminated!]
<infinisil> Great, doing the StackOverflow survey
<infinisil> "Which of the following programming, scripting, and markup languages have you done extensive development work in over the past year, and which do you want to work in over the next year?"
<infinisil> Neither Haskell nor Nix are in the list, and there's only a single "Other: ..." field..
<samueldr> if you're not doing JavaScript, you're not programming, scripting or marking up languages
* samueldr ducks
<infinisil> :P
<infinisil> I guess I could put "Haskell, Nix" in the field..
<infinisil> Goddamnit! "Blockchain / cryptocurrency technology is primarily:"
<infinisil> blablabla, "A passing fad", "An irresponsible use of resources"
<infinisil> It's a single choice question, but I want to mark both
<Ralith> haha
<Ralith> that's a tricky one indeed
iqubic has joined #nixos-chat
<lejonet> samueldr: pffft, amature, everyone knows that vbscript is the only REAL programming language ;)
iqubic has quit [Ping timeout: 240 seconds]
nokomprendo has joined #nixos-chat
<samueldr> a friend of me asked what r13y stood for, I answered reproducibility; another friend jested arthirteeny :)
<gchristensen> haha
<gchristensen> could make a asciilogo, R<small>eproducibilit</small>y
<cransom> it really seems like there should be alternate lyrics for https://www.youtube.com/watch?v=5Qk9o_ZeR7s in there for reproducability
drakonis has joined #nixos-chat
<drakonis> catchy.
<gchristensen> nice
nokomprendo has quit [Quit: Leaving]
srhb has quit [Remote host closed the connection]
srhb has joined #nixos-chat
<__monty__> gchristensen: Wow, that's terrible. It breaks I18N. Just terrible.
* __monty__ dejectedly shakes his head
<gchristensen> lol
<infinisil> gchristensen: You can combine three 3 I's under the fraction to I³
<infinisil> :)
<gchristensen> bahaha
<__monty__> But I^3 is III surely, not IIII?
<infinisil> Oh but r13y.com is nice, haven't seen it before
<gchristensen> thanks, infinisil! :)
<gchristensen> it was a 2-evening hack
<srhb> Is there some kind of mnemonic to r13y that I'm not guessing? I'm never going to be able to remember that url :P
<infinisil> Wait, https://isdebianreproducibleyet.com/ says it's 92.5% reproducible, does this mean we surpassed debian without even trying?
<__monty__> Who'd've thunk perl and python'd be such offenders?
<srhb> infinisil: Far smaller package set, no?
<infinisil> Ah
<gchristensen> for a MUCH smaller package set
<srhb> I think there's a chance we're "pretty well off" by sheer accident of our infrastructure.
<srhb> But probably nowhere near Debian.
<srhb> __monty__: _what_
<srhb> oh, eproducibilit
<gchristensen> maybe I'll just put at the top "(it means reproducibility)"
<srhb> Thanks xD
<gchristensen> the trouble with cute domains
<infinisil> srhb: Same as i18n for internationalization
<infinisil> s/Same/Similar
<srhb> infinisil: I didn't even know that.
<gchristensen> m4 means macro
* srhb cries
<gchristensen> k8s -> k[ubernete]s
<__monty__> You should refer to it as "The NixPkgs repro" for ameliorated confusorium.
<srhb> gchristensen: wtf...
<srhb> Stop messing with my mind
<tilpner> And a11y (accessibility)
<gchristensen> a11y is super cool since it is also ally
<srhb> Yeah, I'll forgive that one.
<infinisil> I'm gonna change my nick to i7l lol
<srhb> The rest can go die in a bucket.
<gchristensen> let's rename nix to n1x
<__monty__> Why is macro such an odd duck?
<infinisil> Lol
<joepie91> ,locate winetricks
<srhb> >:|
<{^_^}> Found in packages: winetricks
<joepie91> hmmk.
<__monty__> ∠( ᐛ 」∠)_
<tilpner> Hey joepie91, I have some time for docs
<samueldr> srhb: stands for arthirteeny
<srhb> samueldr: Thanks, I will accept this as the correct answer.
<joepie91> tilpner: I'm already in menta shutdown mode for tonight unfortunately :( going to sleep a bit early
<joepie91> mental*
<samueldr> they're called numeronyms
<tilpner> joepie91: Could you point me in the right direction? I'll write them
<__monty__> I've considered getting a t8m domain name.
<tilpner> But I'll figure them out myself too
<gchristensen> oh speaking of which, report #2 for r13y is almost done
<srhb> samueldr: I thought numeronyms had to be homonyms when considering the number's pronunciation
<srhb> samueldr: Like l8r
<srhb> Oh well...
<srhb> Learn something new every day.
<samueldr> apparently they're too; sounds like the best definition is "word with a number" :/
<gchristensen> yeah, canine -> k9 -> I guess, if k[anine\0\0\0\0]
<joepie91> tilpner: the best I can do right now without it turning into a lengthy conversation would be: the docs should answer 1) why should you care, 2) which cases is and isn't this appropriate for (especially if they look alike), 3) at least a simple and a more complex example, 4) list any edge cases that you need to keep in mind, and 5) a formal listing (ie. reference docs) for the various exposed methods and what they do, when you need them, and
<joepie91> exactly what arguments they take and what they return etc.
<srhb> gchristensen: Right!
<__monty__> I actually thought k8s was à la k9 rather than i18n.
<gchristensen> watch out for buffer overflows, you might overwrite the next word.
<srhb> samueldr: I guess that is technically what numeronym means.. numbername.. :P
<joepie91> tilpner: https://mail.python.org/pipermail/python-list/2013-February/641772.html addresses a lot of 'what makes for good docs' points and https://gist.github.com/joepie91/5232c8f1e75a8f54367e5dfcfd573726 is a now somewhat-outdated unfinished rant about issues with NixOS documentation and usability that might be of use
<samueldr> yeah
<__monty__> I.e. kubern80s.
<srhb> __monty__: Yes, I hear it pronounced as kates all the time.
<__monty__> I pronounce it cube-err-netties though.
<gchristensen> not cube-err-neats?
<srhb> cube-err-neet-ees here..
<infinisil> coob-err-neetes
<infinisil> For me, I think
<tilpner> joepie91: Thanks! :)
<srhb> mdash: :|
<mdash> that bad eh
<__monty__> I think the closest correct pronunciation is que-berr-n... I'm giving up. Too hard to spell phonetically in english.
<joepie91> cue-burr-nee-tuss
<lejonet> cue burn :D
<__monty__> joepie91: Yeah but that's not very close. Upsilon is basically impossible to render in english afaict. And so is eta.
jackdk has joined #nixos-chat
<colemickens> sandboxfs uses FUSE for FS sandboxing, aimed at speeding up Bazel
<infinisil> colemickens: Damn, interesting!
<infinisil> Could be used to get something like buildFHSUserEnv
<infinisil> But better, probably
<colemickens> Yeah, I think there's quite a few things that could benefit from this.
iqubic has joined #nixos-chat
<colemickens> There's lot of places where runInVM is used just for FS manipulation that could be done outside.
<colemickens> (Though I think LUKS still requires root. There are some LUKS/FUSE projects but they all seem dead.)
<__monty__> Is it compatible with macos fuse?
<samueldr> someone should call tim cook and ask pretty please for more accessible macOS build hardware :/ https://stuff.samueldr.com/screenshots/2019/02/20190205184339.png
<__monty__> I assume that's the reason for runInVM rather than runInContainer?
<__monty__> samueldr: Yeah, it's a cluster duck. GHC has had lots of trouble with OSX and now windows build hosts.
<samueldr> wasn't even thinking about software in particular, just making it less of an issue to get a macOS build host, so either real good hardware, or leniency on the EULA
<gchristensen> hrm,bigmac fell out
<samueldr> in other news, looks like hydra was bumped with the timestamps patch :)
<samueldr> gchristensen: then it's "good" news, as I had assumed it still was building, and just overloaded
__monty__ has quit [Quit: Lost terminal]
<samueldr> the moment it disappeared I think https://stuff.samueldr.com/screenshots/2019/02/20190205184902.png
<samueldr> or not
<samueldr> the moment it should have been available though :)
<gchristensen> :D