brain worm
guy who buys a ups so he doesnt have to fix a nic race on boot
Replying to @StellaFoxxie
lower c++ and c# vs upper Java and other Products? lower music vs other tabs? also idk if that's expected but the cursor-reactive thing in the background doesn't go all the way to the bottom?
Replying to @StellaFoxxie
About says Firefox DevEd 87.0b9 (64-bit)
is tea a conflict mineral
Replying to @nabijaczleweli
normal day https://bugs.debian.org/1018919
Replying to @nabijaczleweli
"average coreutils program has 3 distinct bugs" factoid actualy just statistical error. average coreutils bug has half of a distinct bug. Spiders Stty, who lives in cave & encounters over 10,000 each day, is an outlier adn should not have been counted
i fucking hate the strict-PR-denominationalism github's monopoly created so much
Replying to @nabijaczleweli
unless its something hard-automated and/or controversial like rust-lang/rust splitting a patchset into five single-patch ones is just infinitely more annoying than useful. they'll all have an unchanged message and are mostly unrelated. you'll probably apply them in the same order
Replying to @AssembledGhost
https://twitter.com/nabijaczleweli/status/1565662818174701576
Replying to @nabijaczleweli
oh, sorry, it's three this time, and each of them changes a single, different, file. of course the "one PR" "mixes different changes" that's why they're in different patches
Replying to @nabijaczleweli
this is literally a worse workflow than posting to a list, by a wide margin
in 10-15 years we may record the first naturally-occurring "mom! mom! how do I look?" – "honey, you're dripping :)" interaxion
Replying to @nabijaczleweli
hohoho so deviant https://bugs.debian.org/1018958
Replying to @nabijaczleweli
this post is dedicated to the camel drip check sheesh allahu akbar video
whats the deal with "hiking". americans are always "hiking" stuff. is this because they need a fancy word for walking or
if you're a git disliker you are most likely encountering a perkele tree variant
Replying to @AssembledGhost
i don't, git is great. sounds in general like a skill issue. or you've encountered a perkele tree variant
hm. setting stty kill ^I then stty sane back breaks completion in bash (readline?)
Replying to @nabijaczleweli
pog pog i love debbugs https://bugs.debian.org/1019053
Replying to @nabijaczleweli
i love c++ because the error is always in the dead center of my screen
Replying to @dazabani
hm. good to know my old DevEd is ahead of firefox etat d'art, even if only sometimes
Replying to @greenTetra_
this is your least unhinged take so far to be honest
Replying to @mycoliza, @eddyb_r and @bitshiftmask
we can scour the literature for a Good and Cool solution: per analogiam to Early UNIX, i would simply move the epoch when the timer was close to overflowing
Replying to @mycoliza, @eddyb_r and @bitshiftmask
(less of a problem for the mains-frequency tickrate of the era at 828 days, the citation for both is https://www.tuhs.org/Archive/Distributions/Research/1972_stuff/Readme, https://www.tuhs.org/Archive/Distributions/Research/Dennis_v3/Readme.nsys; for the former also V2, V3 sleep(II))
Replying to @softminus
buddhist rocketry
200ml can-pack-branded not-for-sale can of bubbly is drip. too me.
Replying to @nabijaczleweli
thank.s. good to know one of us knows what the fuck the extension means
Why do they call it crème fraîche when you creme the fraîche crème fermes eat the crème fermant
Replying to @nabijaczleweli
the 3 years of french i did in midschool have not prepared me to contend with the wiktionary declension tables
the way POSIX describes ECHOK and the Linux uapi headers describe ECHOK ECHOKE is absolutely fucking psychotic
Replying to @eatijr
falconry?
if g-easy is a rap enjoyer then im b-walkin
Replying to @nabijaczleweli
(a) what the hell is this post, (b) why am i so walkingpilled lately
Replying to @eatijr
through the facsimile of snooker the modern briton reaches for the purpose and thrill of spear-fishing
hm i think im on that big scoob soul musik drip
racemic mixture of limbs (me)
Replying to @nabijaczleweli
spent entirely too long on this
Replying to @nabijaczleweli
lmao, and of course it blames to start-of-git
they call me contractual, the way I'm much obliged
coming up with a hard-posting/soft-posting dichotomy to quantify the anguish
Replying to @nabijaczleweli
https://lore.kernel.org/linux-doc/20220906003828.aqrcmjzuri5do4uq@tarta.nabijaczleweli.xyz/T/#u
Replying to @nabijaczleweli
yo pog
mandoc moment https://inbox.vuxu.org/mandoc-discuss/20220906113438.4ad6gv5w2frukp5h@tarta.nabijaczleweli.xyz/T/#u
Replying to @nabijaczleweli
noooo dont find two mandoc bugs from the same manual your so sexy haha https://inbox.vuxu.org/mandoc-discuss/20220906122412.5noxbtocludm3tn6@tarta.nabijaczleweli.xyz/T/#u
fucking gnu debugger moment
Replying to @nabijaczleweli
staggering indictment of C as a platform that gdb manages to pass argv + 2 to exec for the first r but explodes for the second
Replying to @nabijaczleweli
this is apparently expected and supposed to happen. it may even make sense! except not if --args. https://sourceware.org/gdb/onlinedocs/gdb/Forks.html
Replying to @nabijaczleweli
pog someone looked at it https://sourceware.org/pipermail/libc-alpha/2022-September/141906.html
i think ive independently developed a simple present/be+present participle split
britcoin
Replying to @__phantomderp
isn't that just __builtin_add_overflow()?
Replying to @nabijaczleweli
there's something to this v/v the sterling post-brexit and clepto post-2021 i think
Replying to @nabijaczleweli
i love processing text https://bugs.debian.org/1019344
Replying to @nabijaczleweli
this is the 6th for stty, which beats out du's 5.5
notepad for next time I need to set the window size in a VM: (stty -echo; printf '\033[s\033[999;999;H\033[6n\033[u'; IFS='[;' read -dR a b c; stty echo rows $b cols $c)
Replying to @nabijaczleweli
bog (bublik pog)
pog pog the courier railed me :0
Replying to @Morwenn_Ed
grey to sit on https://twitter.com/nabijaczleweli/status/1184835796681023488
Replying to @Morwenn_Ed
grey's so small here what the hell
Replying to @nabijaczleweli
lmfao just closed https://bugs.debian.org/210475, which was reported Sep 2003 with the last post on Jan 2008
Replying to @nabijaczleweli
and a POSIX bug to round it out http://www.mail-archive.com/austin-group-l@opengroup.org/msg10123.html
this is a certified bruno powrożnik DING DONG THE WICKED BITCH IS DEAD moment
Replying to @nabijaczleweli
it is in these moments i am reminded of the tom scott talk from a decade(?) ago, whose core tenet was: "the question will not be »did you see what happened?« but »were you there when it happened?«" and, on jod, i was here, i was watching, and it's pure dopamine into my pog cortex
Replying to @nabijaczleweli
(the last part was an eon away linguistically at the time, but, it's strongly implied)
Replying to @nabijaczleweli
8 full pages after STANDARDS. no coherent EXAMPLES, no HISTORY
Replying to @nabijaczleweli
probably the most psycho thing I've ever written, but I couldn't come up with a way to debashify this /and/ have it not take .2s :(
Replying to @dazabani
wash, left is sadder (but also nonstandard, pick right), similar (but without them side by each you'd read them both as right), "sleepy face" apparently but see next, yes if you're japanese
Replying to @dazabani
emojis are great because they tell you what to feel so you don't have to guess except when you do, the XPG calls the latter two "carries a cultural dependency", and, since that dependency is unchecked, this implies that emojis are development tools
Replying to @dazabani
(citation on that is dude just trust me for now, the XPG PDFs are agony enough on a desktop), but this also means that unicode should at some point succumb to the tide of internationalisation, much like who am I -> who -m, imo
skeleton-based emoji when vs emotion-composition-based emoji when
pwease 🥺 any unix enjoyers got an idea?
(the problem statement corollary is: the L5 printf makes '\33[%d;%dR' $r $c appear on the input) https://twitter.com/nabijaczleweli/status/1568052142077796353
Replying to @dazabani
> It is expected that the output from development utilities will be read by an expert user who can interpret it, and no warning is given where their output format is incompletely specified or has a language or cultural dependency.
immeasurably gutted this is way too tall to fit where i need it https://www.olx.pl/d/oferta/szafa-rack-sun-microsystems-szafa-serwerowa-CID99-IDQMMVH.html
Replying to @nabijaczleweli
seven grams of (poorly-extracted probably) seeds
RT @nabijaczleweli: pwease 🥺 any unix enjoyers got an idea?
(the problem statement corollary is: the L5 printf makes '\33[%d;%dR' $r $c ap…
Replying to @nabijaczleweli
8.5pg, second-biggest by size too, and still without HISTORY :v
Replying to @nabijaczleweli
bookending with classic C integer parsing moment :) http://bugs.debian.org/1019463
Replying to @nabijaczleweli
omg pogpogpgogp −xcase removal will be properly noted in Issue 8 Draft 3 https://www.mail-archive.com/austin-group-l@opengroup.org/msg10129.html
Replying to @nabijaczleweli
and a symmetrical (it's not symmetrical, but close enough) mismatch for cooked: https://bugs.debian.org/1018844#10
Replying to @nabijaczleweli
uhoh new one, kill '' is kill ^@ (apparently, you wouldn't know), and kill ^żupan is valid http://bugs.debian.org/1019466
Replying to @nabijaczleweli
aaaaa what the fuck. they strace identically. the 1000000 speed sticks when it's specified what the hell is wrong with coreutils
Replying to @nabijaczleweli
waa http://bugs.debian.org/1019468
Replying to @nabijaczleweli
this is the 10th for stty :) they grow up so fast
Replying to @nabijaczleweli
okay, this is actually a POSIX formatting bug since Issue 6 https://www.mail-archive.com/austin-group-l@opengroup.org/msg10139.html https://twitter.com/nabijaczleweli/status/1565502555521785856
Replying to @__phantomderp
this may appeal to your standards editor sensibilities :)
(this archive interface seems to eat <stuff>s but only in the subject, thankfully, so the empty ""s should be "<undef>"s)
why do people like glass doors on racks so much
Replying to @nabijaczleweli
is mesh too horny or what
Replying to @pixelbeat_
you need to compare cfget[io]speed() instead of the c_[io]speed fields directly – https://git.sr.ht/~nabijaczleweli/voreutils/commit/dcdc8f916b8f503fcddc769ca002498f9b809f08 [0BSD]
up late thinking abt her (sunmicro rack)
Replying to @nabijaczleweli
holy shit if you substitute the default special characters to something reasonable, you could include the three pages of /dev/tty0 ... tty5 (IV) verbatim in a modern manual, and you'd do a better job of explaining how teletypes work than 99% of modern-written bull shit
Replying to @nabijaczleweli
there are – count'em (heh) – icanon with its at-most-a-line reads and line processing, -icanon (min 1 time 0), echo, ^D processing (which, until hard-reading the POSIX description I wouldn't've guessed worked as described), cttys, ^C/^\ (sans pgroups)
need this for ext4
Replying to @nabijaczleweli
hm. the UNIX Programmer's Manual notes that the(ir?) PDP-11 has DC-11 (DC11?) asynchronous modem interfaces, to which stty applies (in contrast to the console to which it doesn't), and the first two arguments set the input/control and printer flags, deferring to the DC-11 manual
Replying to @nabijaczleweli
bitsavers has multiple microfiches of /test programs/ for it, but not so much as a product brief, much less a manual (there is a DEC disk controller DC-11 product brief, but.); wikipedia doesn't seem to acknowledge the PDP-11 has a console or capacity for modem peripherals
Replying to @nabijaczleweli
am i fucking stupid?
Replying to @nabijaczleweli
i am, i just fucked up the query the first time :) all but the first are for the same line of device, and the second-to-last is the right one :)
don't tell me you don't want one. the immense amount of drip in DEC/PDP-11 materials is frankly insulting
Replying to @nabijaczleweli
am i slowly becoming a 1950s retvrn account but for 1972
Replying to @nabijaczleweli
do not get it twisted: this would fuck equally hard if we did a similar type of brochure design language and product shots but with 2022 computers
Replying to @nabijaczleweli
i became a real POSIX bug! dreams do come true! https://www.austingroupbugs.net/view.php?id=1604
Replying to @nabijaczleweli
what if bublik but it wasn't? today's youth asks scary questions – here's how to answer them.
Replying to @nabijaczleweli
maybe im just vehemently anti-gamer
Replying to @nabijaczleweli
holy shit you can trace like half (its not half) of the standard flags to the DC11 manual lmao
Replying to @nabijaczleweli
lmao for (a) teletypes got signals before signals got signals; we /been/ knew job control is king from day 1 and (b) "It should be easier to resume after an interrupt, but I don't know how to make it work." men will literally go to therapy vs. inventing absurd architectural abuse
Replying to @nabijaczleweli
think im all wrote out on V1 after a mere 1.5pg
Replying to @nabijaczleweli
im "deep in it" but ive found the original gamer, lads
Replying to @nabijaczleweli
sorry! sorry! electronic sportsman. i shouldn't flame kids, even if they are 60
Replying to @__phantomderp
ADa (Language)
Replying to @nabijaczleweli
how the fuck do you cite a usenet post
Replying to @nabijaczleweli
like this i guess. decided that "alt.folklore.computing" is the Journal and "IBM Selectric parallel/serial interface" the Article. idk if the ggroups top-level M7H thing is actually stable and refers to the msgid of the top post but that'd make a stellar (awful) Issue number
Replying to @nabijaczleweli
V2 moment
Replying to @atax1a
this would be obvious except GGroups refuses to show me the actual raw post, so idk if in https://groups.google.com/g/alt.folklore.computers/c/M7H9LZrK894/m/-ncoZ9gVzPAJ M7H9LZrK894 is the root's msgid and -ncoZ9gVzPAJ the reply's, or if they're entirely unrelated, or if there's a usenet archive that wouldnt make me want to kernel mo
they call me scientific the way i con and am sus
Replying to @nabijaczleweli
sahrry, sahrry
Replying to @nabijaczleweli
"con to send sus posts" would be better
left: V2; right: V3
what makes mongodb more mongolian than any other database, given the extent and lasting legacy of the mongol empire?
Replying to @nabijaczleweli
mongolian dairy bulb
Replying to @nabijaczleweli
zamn now i want horse
Replying to @nabijaczleweli
oh, of course! mongodb must just have columns instead of rows
super fucked up we gave up on punched paper tapes imo
Replying to @nabijaczleweli
(its probably for the best because jesus christ imagine right but still. they do b funy ideologically)
:)
Replying to @nabijaczleweli
okay okay i know this looks like a self-parody off a big screen and to some extent it is because come on but i think its still a reasonable point. too me
Replying to @nabijaczleweli
finished off a bag of flour
still very funny to me that because erase was # and typewriters were king the Normal way to do emphasis was Stty^H^H^H^H____ and the only way to render this normally is to actually run it off because less chokes on it, col canonicalises to _s, and displaying it is the same as col
Replying to @nabijaczleweli
havent considered that this may just be a less bug before but it may just be a less bug https://bugs.debian.org/1019956
Replying to @nabijaczleweli
this reads like something @oculisprimed would post about
🧈 ← burro; burrito → 🌯
gracias jair!
Replying to @nabijaczleweli
glass can't possibly be cheaper, right?
Replying to @arne_mertz
appx. all the ones I looked down to the description for were described as "front door: glass, vented" and all do appear to have vents so idk
Replying to @nabijaczleweli
cant believe dmr fucking gnomed me from beyond the grave. what the hell
Replying to @__phantomderp
phd politics streamer but for c arc?
Replying to @nabijaczleweli
HOLY FUCK SHE'S GAMEING>?!?!?
Replying to @nabijaczleweli
come on!
Replying to @nabijaczleweli
Teletype Corporation kinda based doe??
Replying to @nabijaczleweli
(literally)
Replying to @nabijaczleweli
this is a certified R. A. the Rugged Man Who Do We Trust? moment!
Replying to @nabijaczleweli
i was haha joking i may have 20 pages but i just got done with V5 and im a fifth of the way through page 17 so unless the scope doesn't reduce drastically (which thankfully it has been, as expected) this may actually be 30
Replying to @nabijaczleweli
i think i have like two full pages' worth of bull shit about the IBM 2741 alone
Replying to @nabijaczleweli
also if someone can decipher what the scribble here says please tell me :)
Replying to @MrDOS
!!!!!!
Replying to @MrDOS
this is supported by a modern mode described identically by POSIX, so i think you're right!
Replying to @MrDOS
i would never have guessed that in a million years (I've also forwarded this to my parents who also came up blank, so I'm assuming this is a different line drawing discipline between poland and america?), thanks loads!
Replying to @greenTetra_
"can nava get a jo" or "you could never get a nava jo" and ship it tbh
Replying to @nabijaczleweli
"noo noo i only use bimbows for gameing !!" cope + seethe unix did it first and unix did it better
if deadmen tell no tales then why tf the queen lying in state rn?? sth up fr
becoming a fujo but just for ken/dmr.. @ATFujoshi is this a healthy way to engage with media?
Replying to @nabijaczleweli
"During the last decade we have discovered a number of new powerful pattern-matching algorithms that are useful for locating patterns in text."
uh, yeah. i bet you fucking have. what the hell
Replying to @nabijaczleweli
taming text formatting like a wild yak
Replying to @__phantomderp
verbal abuse :) https://twitter.com/nabijaczleweli/status/1386044286668525570
Replying to @__phantomderp
the same fucker tried that shit again earlier this year but i cant find the post now, i chewed him out again, its fun :)
i think with age i just get older
Replying to @dazabani
[compartmentalises yuor cat]
oooo nft this, nft that .. my brother in christ just make the tokens funny !
Replying to @nabijaczleweli
done with V7 at 21 pages flat :)
ozymandias a bitch. shouldve come up with farming to have a steady world supply imo
Replying to @nabijaczleweli
lets go babeyyy
Replying to @nabijaczleweli
bitches love getting couriered their weight in steel
mfers b like "fuck you and your hand-outs!!" then b lactose intolerant......... ok we get it you're ashamed your land is so rich ..!
Replying to @nabijaczleweli
no vat invoice. i hate it here. the best online retail experience is provided by Some Guy, and the more physical locations the worse it gets
Replying to @nabijaczleweli
pog pog i can get a modem implant any time after noon tomorrow
Replying to @nabijaczleweli
catch me skrrting in 3:30 at like
Replying to @nabijaczleweli
im pogging?
Replying to @nabijaczleweli
get in
Replying to @__the_sylph__
now branded spikevax apparently?
Replying to @nabijaczleweli
the WHAT port?
Replying to @__the_sylph__
rj11 sticking out my abdomen like
hehehe, not my goofy ass deforming elastically!
[this user underwent plastic deformation]
Replying to @nabijaczleweli
https://twitter.com/mycoliza/status/1572662128431230976
Replying to @nabijaczleweli
yeah just spew on your floppy, mate, she'll be 'right
Replying to @nabijaczleweli
before/after getting modemmed. i just wish i'd #knowntherisk
Replying to @nabijaczleweli
(this is a good (bad) pretext to post absolutely the worst selfies imaginable; also lol im wearing the same thing)
Replying to @frameslip
40U
Replying to @nabijaczleweli
total 8 man-hours. the back door doesnt fit because think different the rails have protruding bits that interfere with the support posts. they still haven't sent me the invoice and it do b a piece of shit kinda so like 50/50 im gonna have to take it apart and return which is fun
dataphonotactics
Replying to @eatijr
alas, the cats are out so i couldnt grab a shot of shaking one of their hand; sorry
[programs your data processor]
Replying to @nabijaczleweli
there are big bubliki among us
Replying to @nabijaczleweli
😳 [extends your virtual addresses] 🙈
MTG cards are an unregulated security
Replying to @nabijaczleweli
officially bigger than a ~V2 file can be :v
Replying to @nabijaczleweli
(CORRECTION: max seek offset, goes to 64k because it was a single 16-bit word and yes this is the reason why we got block-addressed disk I/O[1]; max file size on ~V2 is described like it goes up to 1Mi but there's two bytes (+1b(?))[2], and all of userspace gets 2B[3,4]. unclear)
a what driver sorry?
Replying to @nabijaczleweli
nooooo bitsavers doesn't have an ML11 manual
Replying to @nabijaczleweli
..do they?
baby you're the lempel to my ziv
holy zamn is this like UNIX-domain sockets.. in 3BSD?
Replying to @greenTetra_
+2+2+2+2
Replying to @nabijaczleweli
it's actually remarkable just how unremarkable 4BSD job control is; the PREFACE says [1], but AFAICT this straight-up doesn't hold for job control which beside living in -ljobs and requiring stty new is so nauseatingly boring and fully-formed you might as well read credentials(7)
if cats have nucleotides then why do i keep putting in meats instead of them being nuclear. nonproliferation can suck my shit hamburger style
Replying to @sovietvisuals
p. 1: "A monster!" — thinks the mouse.
p. 2: Fish scatter when they see me.
Replying to @nabijaczleweli
free mandoc bug, regression against at-most-3BSD troff, likely earlier https://inbox.vuxu.org/mandoc-discuss/20220925161826.mkpuxk43bckq2zxa@tarta.nabijaczleweli.xyz/t/#u
Replying to @nabijaczleweli
so fucking true
Replying to @nabijaczleweli
1 BSD = 1 page (converges at 4BSD)
Replying to @nabijaczleweli
maybe DEC just had line-based drip. have we considered this
llvm bad dragon
Replying to @nkizz11
idk something like this :)
RT @nabijaczleweli: @nkizz11 idk something like this :)
yeah we gay, keep scrolling
Replying to @nabijaczleweli
this says a lot about our society
non-builtin cd that just chdir()s then execs $SHELL. would love to know your thoughts on this
Replying to @nabijaczleweli
citing another usenet post. my ant sisters are not smiling upon me today
oh that's your programming? then why is it conversing with me?
Replying to @nabijaczleweli
sometimes i feel like i shouldnt have the power to go off on these poor people but then i remember this is for the switching equipment of american monopoly number 3 so they deserve it and are also all dead now so its fine
Replying to @nabijaczleweli
syscon.. these nuts?
she let me hit 'cause im fun, but in the sense that the stupidest song you've ever heard is "fun" because the phonotactics are inoffensive
Replying to @atax1a
never, at least never as described. in the single-user PDP-7 proto-UNIX, chdir was a regular program /but/ it had no fork, and no exec per se – the shell had a loader that overlaid it with the program (like ld.so hehe), the program exited, exit(2) overlaid it with a new shell[1]
Replying to @atax1a
since fork, chdir has always been a builtin[2]; this is a constant in UNIXes as we know them today[3](1), alongside login as soon as init was split into init/getty/login in V2[4](2) (the reasoning for the latter is horseshit[5](3) and poisoned by the proto-UNIX but hey)
Replying to @atax1a
[1]: https://www.bell-labs.com/usr/dmr/www/hist.html, Process control, list 2; to the precision of PDP-7 proto-UNIX being super fucked up, cf. down from The PDP-7 Unix file system
[2]: ibid., list 1; ibid., para. "Although the multiple-process"
Replying to @atax1a
[3]: https://www.tuhs.org/Archive/Distributions/Research/Dennis_v1/UNIX_ProgrammersManual_Nov71.pdf, chdir (I)
[4]: https://www.tuhs.org/Archive/Distributions/Research/Dennis_v2/v2man.pdf, login (I)
[5]: ibid., sh (I)
Replying to @atax1a
would love to know your thoughts on this
Replying to @dazabani
yes, and it's waiting for the cum shell to die, you cd - by ^Ding :)
Replying to @atax1a
this has stunlocked me. what the fuck is this guy /on/ about. s6 was odd but haha funi, this is straight-up detached from reality
Replying to @dazabani
it is, of course, awful. truly horrible idea realistically. BUT it would be funny and you'd get a (if you remove login) builtin-free shell, which would be pog
Replying to @nabijaczleweli
bublik no jutsu
Replying to @atax1a
a more extreme variant of this take but one that I'd agree with would be to make the CWD not special but rather a conventional userspace notion of fd -1 (like AT_FDCWD, actual fd -1 would be stupid), making all opens openats; and while at it also do that for / instead of chroot()
Replying to @atax1a
in that case there is no different between a "container" or "chroot" and any process; "chdir" is just freopenat (+- sp., you'd want a better primitive that did something like openat(x, ...) = y; dup(y, x) probably; AT_REOPEN?) and you get ch{root,dir}ing for free always ex def
Replying to @atax1a
init just gets created with fd 0 being A root directory (the system "global" root or whatever else root for a "container"), everything else flows down from that normally (dup for CWD, &c.). of course linux-style modprobe upcalls would be harder but eeeh you could arrive at a bett
Replying to @atax1a
fun fact: it always has been but just namespaced differently with ε different rules – paths starting with / substitute the CWD fd with a global (later process-specific) root fd: in that way openat is free; you could say that this would be a massive simplification of the system
Replying to @atax1a
as for the latter bit, well, I've been in the Austin Group part of my WG11 paper (late december 2021) since May, and it's a wording change that would make extant Linux x32 systems conformant and no changes to already-conformant systems :)
she propagate on my nan til i signal
Replying to @nabijaczleweli
https://twitter.com/nabijaczleweli/status/1379046876540977153
i love the bootloaders tweet
Replying to @nabijaczleweli
votrax
occupied by a hostile force
[two girls one cup voice] two systems one country
all water and no salt makes jeff a very dry boy
society if
if character-cell terminals did half-line-feeds
Replying to @nabijaczleweli
guess who doesn't pick up the phone or hasn't replied to my on-platform posts in the past two weeks :)
Replying to @nabijaczleweli
finally: a reason to have more than three gigabytes of ram. also, why is this so slow. i could recognise characters on a thousand pages very quickly, and i need less than 300W to do it. it is easy, too me.
Replying to @nabijaczleweli
108`607.5J for just the OCR bit. that's 26kcal, but cant feed computers milk (yet). i stay winning
great. love that. zero notes on why this is deprecated, or what that entails. no mentions in {NEWS,changelog}.{,Debian}. you telling me they got rid of SHA256?
Replying to @nabijaczleweli
apparently there's OSSL_DEPRECATEDIN_*_FOR() macros with replacements too but they're unused here so it must just be that they cracked SHA256
Replying to @nabijaczleweli
there's a migration guide, not distributed, of course, which says that "low-level digest functions" "have been informally discouraged from use for a long time". as opposed to what. how do you make this fundamental API high-level
Replying to @nabijaczleweli
EVP_DigestInit_ex() isn't any higher-level, and the original PR that introduces this says absolutely nothing on why you'd want this except repeating the same meaningless spiel. why!!!!!!!!!!!!
Replying to @nabijaczleweli
"being informally discouraged" means nothing. there's no reasoning. why would you force everyone to pay a string lookup (or not, I guess, but the no-lookup functions aren't advertised) and fucking /allocate/ for the hashes
Replying to @nabijaczleweli
im sorry im sorry im trying to delete it
@greenTetra_ umm..... found your alt? https://www.youtube.com/watch?v=5ipscwyz3Zk