<andi->
latency to the endpoint is around 10ms.. They recently started offering one that is just ~1.3ms away. Not sure if that does help in any noticable amount for the narinfo lookups..
<pie__>
andi-, arigato gozaimashita
pie__ has quit [Ping timeout: 252 seconds]
jasongrossman has joined #nixos-chat
drakonis has joined #nixos-chat
endforma1 has quit [Quit: WeeChat 2.4]
endformationage has joined #nixos-chat
jasongrossman has quit [Quit: ERC (IRC client for Emacs 26.2)]
_e has quit [Quit: WeeChat 2.4]
jasongrossman has joined #nixos-chat
_e has joined #nixos-chat
_e has quit [Client Quit]
_e has joined #nixos-chat
drakonis has quit [Quit: WeeChat 2.4]
jD91mZM2 has quit [Ping timeout: 258 seconds]
aminechikhaoui has quit [Read error: Connection reset by peer]
jD91mZM2 has joined #nixos-chat
endformationage has quit [Ping timeout: 252 seconds]
kgz has quit [Ping timeout: 258 seconds]
kgz has joined #nixos-chat
<MichaelRaskin>
__monty__: putting people who do write JS into the defensive mode makes it harder to have nuanced discussions about tradeoffs of mixing languages for web development
jasongrossman has quit [Ping timeout: 248 seconds]
<MichaelRaskin>
__monty__: and JS disappearing (or NPM disappearing) won't fix the economics — including the economics of education — preexisting education imposed on users, I mean — around the software development.
jasongrossman has joined #nixos-chat
jasongrossman has quit [Remote host closed the connection]
jasongrossman has joined #nixos-chat
<simpson>
Maybe. JS is so popular (my prior is that fully half of all devs are JS-first or JS-only!) that it is *very* hard to tell whether JS is bad, JS authors tend to be bad at writing correct JS, or JS is fine and humans have no idea how to write code.
<simpson>
Add onto that that JS's design flaws are relatively unique. For example, JS doesn't have ints; that is a *bold* and rare design move that I cannot easily compare to most other platforms.
<MichaelRaskin>
simpson: neither of the above; programmers are bad at fixing large-scale economics effects, and code is not the only mass-market product with low manufacturing quality
<simpson>
MichaelRaskin: Sure, but e.g. in mass manufacturing of consumer goods, tradeoffs like ABS vs. PLA are made based on material science, based on the chemical properties of the substances in question. What's the equivalent of chemistry for writing code? How can we measure tradeoffs between JS and other languages?
<MichaelRaskin>
simpson: it's not between JS and other languages
<MichaelRaskin>
It's also between having _any_ spec for the product or no
nh2[m] has quit [Read error: Connection reset by peer]
<simpson>
MichaelRaskin: Between many different languages, of which JS is one. I'm not trying to pick on JS in particular; my view is that *many* languages are not great.
alienpirate5 has quit [Write error: Connection reset by peer]
dtz has quit [Write error: Connection reset by peer]
zimbatm has quit [Remote host closed the connection]
colemickens has quit [Remote host closed the connection]
Ralith has quit [Write error: Connection reset by peer]
<MichaelRaskin>
It's also between having any programmers _ever_ speak to end-users and considering this idea a horrible threat
<MichaelRaskin>
Language is irrelevant, just as materials are sometimes not the entire story — a lot of unreliable parts have been manufactured out of good steel
<simpson>
Sure. This is a perennial question of management: Do managers do anything? In this case, are managers necessary for facilitating customer-dev communication?
<ar>
writing a spec for anything using js would take enough time for half of the dependencies to get deprecated ;)
* ar
runs
<simpson>
But, unfortunately, language *does* matter, for no other reason than that choice of language determines Kolmogorov complexity, which determines which idioms and programs will be short.
<simpson>
And usually further for the reason that choice of language determines which formal methods can be brought to bear on the produced artifacts.
<simpson>
ar: Between JS and Python at my current dayjob, it's not clear which is easier to maintain when the application gets large. It doesn't appear to be any worse to design a feature in JS compared to Python, when considering the cost of years of maintenance.
<MichaelRaskin>
Formal methods are useless when noone will look at the problem reports anyway
<simpson>
MichaelRaskin: Like I said, it's very hard to tell whether the language is bad or whether humans are bad at thinking in the language or whether humans are bad at *any* programming task.
<MichaelRaskin>
Humans are bad at tasks they are almost explicitly expected not to do well
<MichaelRaskin>
Hm. Maybe the value of Ada was proof of work — one had to pay _some_ attention to make the code even compile
<MichaelRaskin>
Ada, the blockchain of programming languages.
<ar>
that's not a fair comparision. Ada (and SPARK) have their uses
jasongrossman[m] has joined #nixos-chat
jasongrossman has quit [Remote host closed the connection]
jasongrossman has joined #nixos-chat
jasongrossman has quit [Remote host closed the connection]
jasongrossman has joined #nixos-chat
zimbatm has joined #nixos-chat
Ralith has joined #nixos-chat
colemickens has joined #nixos-chat
<MichaelRaskin>
ar: I do not say that proof-of-work in blockchain sense is never useful, either
<MichaelRaskin>
And in Ada proof-of-work is mutual, the language designers and implementors _also_ provide a proof of work that gives people hope that the language is not entirely built out of landmines
jasongrossman has quit [Remote host closed the connection]
jasongrossman has joined #nixos-chat
pie_ has quit [Ping timeout: 258 seconds]
pie__ has joined #nixos-chat
jasongrossman has quit [Read error: Connection reset by peer]
jasongrossman has joined #nixos-chat
jasongrossman has quit [Ping timeout: 255 seconds]
jasongrossman has joined #nixos-chat
jasongrossman has quit [Ping timeout: 252 seconds]
<simpson>
Also! Looking back at the original topic, it would seem from https://github.com/features/package-registry that the builtin npm-scope feature (https://docs.npmjs.com/misc/scope) is used to give each Github user their own package scope to which they can publish. So perhaps the compatibility problems are limited to packages becoming rescoped, and folks having to update their dependency listings. I wonder how
<simpson>
much changes for nixpkgs?
jasongrossman has joined #nixos-chat
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
<joepie91>
simpson: nope
<joepie91>
it being scope-only doesn't make much difference
pie_ has joined #nixos-chat
<joepie91>
you *can* configure npm to look up different scopes in different registries
<joepie91>
which appears to be what github is doing
<joepie91>
but that still doesn't give you cross-registry references
<joepie91>
and the use of a scope on github doesn't prevent registration of the same-named scope on npm
<joepie91>
so you still have the same compatibility problem
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
pie_ has quit [Remote host closed the connection]
<infinisil>
It's always a bit magical when two different interests lead you to the same thing
<infinisil>
I just watched a random video about Pluto, when it showed the names of its moons, Nix, Hydra and Charon :)
<pie__>
:p
<MichaelRaskin>
That's called «how else are we supposed to call stuff»
<MichaelRaskin>
I think the proper spelling for Pluto's moon is Nyx, though
<MichaelRaskin>
(How I called VMs: list of moons of Saturn)
<ar>
speaking of. does Nyxos have a NixOS community?
<ar>
erm, Naxos
<MichaelRaskin>
infinisil: I did my search before choosing!
<MichaelRaskin>
infinisil: note that they also have official _numbers_ too!
aminechikhaoui has joined #nixos-chat
<pie__>
anyone used anki on mobile apple stuff?
<__monty__>
Only the site.
<pie__>
figured i might shell out for the app :I
<__monty__>
If you use it heavily it's probably worth it.
<pie__>
i should take advantage of it while i still have an hour commute to work
<pie__>
i mean, if we're really fair, if i learn anything substantial 25 bucks is worth it
<MichaelRaskin>
Well, learning anything substantial also depends on packing it into a fitting format
<pie__>
MichaelRaskin, i dont follow but i probably agree
<MichaelRaskin>
I mean, you need to cut the thing you want to learn into the pieces that can be put on a card, but that will also form a web of associations once you do remember them
<pie__>
yeah
<pie__>
i really dont want to have to put the effort in making my own card sets...but i will probably have to, to get anything decent
<__monty__>
It's supposed to work much better if you make your own cards.
<pie__>
damn man why i always gotta do everythin myself? :P
<pie__>
jk jk
<pie__>
hm. does the mobile app support addons?
qyliss has quit [Quit: bye]
qyliss^work has quit [Quit: bye]
qyliss has joined #nixos-chat
qyliss^work has joined #nixos-chat
jasongrossman has quit [Ping timeout: 258 seconds]
jasongrossman has joined #nixos-chat
endformationage has joined #nixos-chat
Synthetica has joined #nixos-chat
qyliss^work has quit [Quit: bye]
qyliss has quit [Quit: bye]
qyliss has joined #nixos-chat
qyliss^work has joined #nixos-chat
drakonis has joined #nixos-chat
<manveru>
wut, it costs money on iphones?
<MichaelRaskin>
That's what you get when you cannot sideload and also publishing in the store costs money
Synthetica has quit [Quit: Connection closed for inactivity]
<infinisil>
Because I just thought of this and it's offtopic for #nixos:
<infinisil>
I should register a domain like dot.dot.lowercaseA.com
<infinisil>
xD
* joepie91
doesn't get it
<samueldr>
impossible to share verbally, a good edge case
<MichaelRaskin>
Isn't it infringing on Slashdot trademark?
<samueldr>
"e-mail me at infinisil.at@dot.dot.lowercaseA.com"
<MichaelRaskin>
Too bad the case doesn't matter in domain names
<infinisil>
Ah right
<__monty__>
But the plebs doesn't *know* this >: )
<infinisil>
"e-mail me at infinisilatgmaildotcom@dot.dot.com"
<gchristensen>
watch out for domain squatters at dotdotdot.com
<samueldr>
or "infinisil@gmail.com"@upperasedashone.com
<samueldr>
uppercasedashone*
<__monty__>
Surely you'd use "e-mailme@infinisilatgmail.com"
<infinisil>
We should all become villains we're so evil
<gchristensen>
upper case 7 for example is &, uppercase 1 is !, etc.
<__monty__>
lol
<samueldr>
gchristensen: that's not right !"/$%?&*() here!
<samueldr>
:3 and then there's azerty which uses shift to access numbers already
<gchristensen>
:D
<samueldr>
keyboards are actually way harder to handle right than most people would think... and then I always learn more and it never gets easier
<gchristensen>
definitely true
<__monty__>
And terminals.
<pie__>
e-dashmail
<infinisil>
samueldr: Same with programmer dvorak which I'm using
<infinisil>
(numbers on shift)
<infinisil>
Can be pretty annoying tbh..
<infinisil>
But symbols without shift is wonderflu
<infinisil>
wonderful
<__monty__>
Is it worth the tradeoff though?
<MichaelRaskin>
Well, if you want to avoid magic constants in your code…
<gchristensen>
+KARMA_INCR
<__monty__>
Thing is I don't program 100% of the time.
<MichaelRaskin>
Big mistake
<gchristensen>
you mean FULL_PCT% of the time
<MichaelRaskin>
gchristensen: I think even with Caps Lock underscore requires Shift key anyway
<gchristensen>
aw
<MichaelRaskin>
Although maybe not, not sure what happens when digits require Shift
<infinisil>
__monty__: Yeah I think it's worth
<infinisil>
I need these symbols way more often than numbers
<infinisil>
(these are on my number row: &[{}(=*)+]!#)
<gchristensen>
for whatever reason my keyboard has a native § key
<gchristensen>
s/native/physical/
<__monty__>
No $, no ^, no @, no %?!
<infinisil>
gchristensen: Yeah swiss keyboards have that too for some reason..
<infinisil>
__monty__: There aren't enough keys in that row for all of them!
<gchristensen>
and it shifts to ±
<__monty__>
Yeah §, shift-§ = ±, which is more useful.
<__monty__>
I like the idea of qmk's layers.
<__monty__>
Have an entire layer for symbols.
<MichaelRaskin>
I think that's called Level3 Shift?
<MichaelRaskin>
I more or less use it like that
<MichaelRaskin>
Well, and also for deadkeys
<__monty__>
I think the § keyboards may be ISO layouts or something?
<samueldr>
105 keys more likely the right name I guess
<infinisil>
I have the problem that whatever keyboard layout I decide on, I need to adjust it to two different physical layouts.. My ergodox EZ and my macBook air one
<samueldr>
gchristensen: is it right of the left shift?
<gchristensen>
it is!
<__monty__>
This is an apple keyboard.
<samueldr>
iso layout is more related to the size of the enter key
<__monty__>
Oh, it's left of 1 here.
<gchristensen>
it is a Kinesis Advantage, and it only has this key in Dvorak mode
<__monty__>
samueldr: Yeah, my enter's iso shaped.
<samueldr>
__monty__: just making sure, you mean its height is two rows?
<__monty__>
Yes with the typical wider at the top shape.
<samueldr>
tired: "iso layout" wired: 105 keys with tall enter key, with 105th key located $location
<__monty__>
And mayb eit's teh return key, rather than the enter key?
<samueldr>
or don't you mean the ⏎ key?
<__monty__>
I do mean that one.
<samueldr>
:3
<samueldr>
tired: control; wired: ✲, ⎈ sometimes ^ or ⌃
<infinisil>
Is that emoji??
<__monty__>
Yeah, I don't speak machine either.
<infinisil>
You know what, I have this shitty emoji keyboard lol