#💽Programming Chat v2
1 messages · Page 54 of 1
if it's not 100% done it's like 66% done
wdym nto useable
professionally mounted
rust coded
kotlin*
lol
i wish HTTPS DNS wasn't a thing
nobodys ip on here is actually their ip
its just google/cloudflare services
why do you want dhcp
mak e it for your toy os
CSV
could be but isn't
I'm tempted to say MySQL but I'm betting it's something like Oracle or Postgres
it's actually sqlite
oh that makes sense
running on every android device
multiple instances on every android device lo
- a whole buncha other shit
yeah
sqlite is actually really nice
been using it for a bit
all a basic project really needs
easily scales and isn't intensive on performance
I bet some of those run Rust on them too :3
welcome back java
most of it is Kotlin though
like a solid 85%
if we're talking apps
is the os then it's mostly c
Oh yeah for sure
But it only takes one rust app to say rust is running on it
ok well we got C running on android before Java
true
二次創作、悪意や宗教政治絡まなければ好き放題やってほしい
Lots of kanji…
i cannot show you the post after that
@pallid loom
Jesus that must be bad then
もうこの世にはゲーム溢れてるから自分のゲームに時間を割いてもらえるよう、選ばれるために自分にしか作れないものと価格で勝負するしかない もはやお金いらないから私のゲームで遊んでくれ…
魔魔魔 マジカル・アンド・マラクル
この広い世界へ飛び出そ
HALT
std::mem::forget
This Java tutorial describes exceptions, basic input/output, concurrency, regular expressions, and the platform environment
Gonna look into this for more efficient file and DNS service
why do you need a crate to breathe oxygen
because we can
Oh no
fat fuck
epic 2 minute video essay
#scpsl #scpsecretlaboratory #snake #record
scp:sl time
turn it into Ada time
actually question
considering Ada’s tasks
Consider an http server
How do you spawn more tasks to handle more requests
cause the number that you can have with a subprogram is fixed right
using access-to-task types
I believe the maximum tasks you can have is operating system dependent but the last stat I saw was in the thousands
explain
task type B;
type A is access B;
and then how use
you call the rendevous functions in B thru A
but is that really making new tasks per request?
yes
seems like it's just reusing that task
you're supposed to create a new instance of B with a reference to it thru A
then you can use that access however you want
ad-hoc or store it in a fixed size vector
directory listing ...
client requests directory
check for directory cache
found?
- check directory watcher
- changes?
- GOTO not found
- no changes?
- send back cached list
not found?
- send back cached list
- compute and send back computed directory listing
(same process can be done for DNS record RR computation)
hello yes I would like companion object sandwiches please
"discrete is better" mfs when they see words
we're better than java
arrayOf() or toTypedArray
no
yeah
dont use kotlin async
(false)
word up
oooh that's a good one
@spare quartz "Write once, use a different version anywhere"
not super certain on this working
but gonna try it out
STOP DOING ADA
- Computers were meant to compute, not to be bossed around in Pascal's evening gown
- Ada is a PERSON, not a programming language
- "with Text_IO; use Text_IO;" — sounds like a Victorian etiquette manual
- You can't just "raise an exception" — that's called being rude
- "Hello, I would like
mod 2 ** 5apples please"
They have played us for absolute fools
"RENDEZVOUS"? ARE WE SCHEDULING A DATE OR WRITING CONCURRENT CODE?
lmfao that's great
If Ada Lovelace knew you were using ALGORITHMS to make VIDEO GAMES she would WRITE A STRONGLY WORDED LETTER
LOL
Strong typing? The only STRONG TYPING Ada did was on a MECHANICAL CALCULATING ENGINE
The Department of Defense should DEFEND us FROM Ada not MANDATE it
@spare quartz LMFAO
THAT LAST ONE

this one and the two after it were generated by claude
honestly
pretty good for ai
horrible
oh right
BLS 1.14 and static server 1.3 are incompatible with each otehr
i need to make a static server 1.4..
bls?
BSL**
bsl??
bread_server_library-code
o
i believe our DNS, router, maven, mail, and ws-facilitation services run it
lol
the maven, mail, ws-facilitation are a subset that runs static-server, an executable library for HTTP servers (on BSL)
that looks just as bad as c why would you use js 😭
rust on top
lol I love the amount of unsafe
where?
presumably here and alike repos
I guess its fast though
like, a lot a bit fast
a lot a bit?
no
nope
that isn't real
its real
im 100% sure
u crazy
the person i work with in this mod
is that their name
no
chris from neco
this is why hyphens should be removed from english
he tried to run a python 2 program on python 3
ohhhh
that's your language
AERA!!!!!!!!
!!!!
what happened
to the messages
that followed
I didn't even get a chance to read them
and they were taken away
ferris ❤️
idk aeras schizo
whaat is goin on
atp is schizo
TVアニメ「てっぺんっ!!!!!!!!!!!!!!!」でEDアニメーションを担当したばやちゃおさんと一緒に歌っちゃいました!
Original:
May'n - あはっててっぺんっ
https://youtu.be/zR99bqxQyw8
作詞:May'n・大石昌良
作曲・編曲:大石昌良・やしきん
▼ TVアニメ「てっぺんっ!!!!!!!!!!...
wrong representation of the lang
I need a slow snake
python problem
import solution
✅ solved
nah bro disabled the GIL
the python vm is just the jvm but for lazy people
ohhhh wait thats genius
message.txt
the directory cache works!
I love mobile
there’s a reason people don’t write microcontrollers in Java
me when I’m flying my jet powered by Java and I die from a GC pause
YOU AND GC PAUSES
im going to pause your life indefinitely
i haven’t really touched kotlin
so then Kotlin scary :3
im going to kill you .
okay dokay 
I’m joking
I’m just trying to prod atp
go develop guardsman or something!!!
There’s a reason that microcontrollers don’t run Java
nope
no!!!!!
why!!!!!!!!!
bepis
getLogger.fine { "Found file for \"$storePath\" at \"${requestedPath.canonicalPath}\"" }
val modifiedSince = request.headers["If-Modified-Since"]
val headers = mapOf(
"Last-Modified" to DateTimeFormatter.RFC_1123_DATE_TIME.format(
ZonedDateTime.ofInstant(
Instant.ofEpochMilli(requestedPath.lastModified()),
ZoneOffset.UTC
)
)
)
var sent304 = false
if (modifiedSince != null) {
val date = DateTimeFormatter.RFC_1123_DATE_TIME.parse(modifiedSince)
val mLong = (date.get(ChronoField.INSTANT_SECONDS) * 1000L) + date.get(ChronoField.MILLI_OF_SECOND)
if (mLong == requestedPath.lastModified()) {
HTTPResponse(
304, request.version, headers, requestedPath.length()
).write(sock.outputStream)
sent304 = true
}
}
if (!sent304) {
HTTPResponse(
200, request.version,
headers, requestedPath.length()
).write(sock.outputStream)
if (request.method == HTTPMethod.GET)
requestedPath.inputStream().transferTo(sock.outputStream)
}
true
file cache experiment
I don’t wanna…
do it
noooo
I Saw.
uh
HTTP/2 looks a lot more different than what i was expecting
time to code in support for it
add quic next
you should plan to drop kotlin and ada :3
I should plan to become a better person
not happening
Java has 9 MILLION classes and NONE OF THEM SOLVE REAL PROBLEMS
FINE: TLSv1.3 TLS_AES_256_GCM_SHA384 "98.10.170.194:43980"
No peer authenticity
Now do it in Ada
do what in ada
DNS
i mean maybe but that'd also require making BSL in ada
and BSL is already moderately complex
ばっか
BSL?
you have the memory of a non ada user
I have fish memory yes this is known
known*
Shut
bread server library
its the thing that holds every protocol specifcation we code for
HTTP/DNS/SMTP/WebSocket/BEG_STDPROXY
also holds stream utilities
might deprecate certbot..
Sure if you can get a cert to become a CA
i mean like
analyse what it does and make our router automatically do that
becoming a CA would probably require tons of money of which we do not have
I don’t understand this
?
what would you be analyzing, what behavior would you make your router replicate, and how
ill answer your questions with a question
what does certbot do? what does certbot give you?
oh you’re saying deprecate automatic certificate acquiring and renewal
not deprecate the entirety of letsencrypt
and do it ourselves
fortunately i have found an rfc on which to start
UNfortunately i need to get myself used to how java manages certs
You and your silly rfcs
ultimately, the goal is to deprecate the internet, and everything within.
and then we deprecate your computer, your lang, your compiler, and then you.
You’ll never deprecate rust
Nuh uh
Not even Ada comes close to the glory
The magnificence
...of ばやちゃお!!!
acme time
HTTP Router Microserver <-> ACME Microserver <-> DNS Microserver
probably will do something in this type of setup
finally good server design
Huh
self realization
so efficient our memory use actually went down overnight
how is it continuing to drop
we're at 875 mib use
@spare quartz dang, kotlin can target jvm 8?
insane that it supports jvms that far back
javac and by proxy kotlincs compilers accomodate everything supported when they were made
although im trying to make ktor run and uh
logback is too recent
oh logback 1.3.0 works
@spare quartz instead of deprecating matrix you should deprecate irc or xmpp, matrix is a doomed protocol
what makes you say this now
魔魔魔
but yes
well conduwuit which was like the only decent actively developed server is now abandonware
the maintainer quit within the last week
bad design
protocol itself isn't really that secure because it leaks metadata like a sieve
whole thing is basically centralized on matrix.org anyways
(there's a lot of metadata with each message and if any of it is encrypted at all it's a very very small portion)
i need to get started on this
I kinda want to try making an irc server
and UNFORTUNATELY I'm open to using Kotlin or Ada instead of just Rust
@spare quartz yeah here's what a matrix event looks like https://web.archive.org/web/20210131022939/https://snippets.serpentsec.1337.cx/matrix-metadata-leaks
you can see lots of it is unencrypted
could it protect me against inviting a journalist
uhh
{
"type": "m.room.member",
"sender": "@babba27:fantasycookie17.onederfultech.com",
"content": {
"membership": "join",
"displayname": "DriftNotSkid",
"avatar_url": "mxc://fantasycookie17.onederfultech.com/uLqqZhJdYHRNvNWWIFeJbDlP"
},
"state_key": "@babba27:fantasycookie17.onederfultech.com",
"origin_server_ts": 1611901567921,
"unsigned": {
"replaces_state": "$xEr7yGLdOVOxF7NxEEFHK-ToDTDMAW2PqPqhwA6XqFg",
"prev_content": {
"avatar_url": "mxc://fantasycookie17.onederfultech.com/uLqqZhJdYHRNvNWWIFeJbDlP",
"displayname": "DriftNotSkid",
"is_direct": true,
"membership": "invite"
},
"prev_sender": "@babba27:tchncs.de",
"age": 724
},
"event_id": "$AsXIuazCSM7liww5v_ZUntbAFDx8kXDH0-11QaS3UGQ",
"room_id": "!ADDxogcQLsbRpVWTLR:tchncs.de"
}
probably not
if the feds have a link to the journalist's matrix username then
lmao
lmao
stop saying lmao‼️
lmao
i just finished my food
now i gotta go fix a shader for a website
jvm could do that faster
wdym a sunfire
Like a Sun machine
probably
Running on SPARC
the primary reliance is on if GCC has support for it
Might be able to use SweetAda
Since it says it’s tested against SPARC
@spare quartz how would I go about trying to compile Ada for SPARC
GNAT is bundled into GCC so no
as long as the GNAT on that GCC is designed for the system it should work correctly
otherwise if you're say, building for just a CPU architecture, and not any system, you have to either use a special runtime of that GNAT (e.g. ZFP) or make your own RTS
bah this project makes me irrationally mad
it is this easy to make a port scanner
urgh I might have to build gcc myself to get SPARC support
added dns
i mean
aeddded tcp to dns
気張っていく路線にしようか考えたけど旅行中のありのままをお届けしようとこんなのになりました。
またいっぱい行きたいな!シャ!あと、わたしがぽ氏さんと結婚した理由がよくわかる動画になった(狂人)
#熱海旅行 #伊豆高原 #きらの里 #花火大会
▼...
which is a prerequirement for dns over http
let me watch my vlog of
http doesn't run over udp?
bayachao making the ghost sounds from mario 😭
http/3
ok buddy that's quic
DNS OVER BAYACHAO.
i cant get over this
😭
ARE ALL JAPANESE PEOPLE ON METH OR SOMETHING??
so true
RIGHT
ACME DEPRECATION
thankfully certbot is already gone
adding SSHFP
@timid quartz sooo
apparently... there isn't actually any costs
we'd probably not be eligible cause we're not even a company yet but
bread CA...
new low for JVM
52 M
impossible
oh interesting
yeah we just have to get WebTrust or ETSI approval
(or government if we were making a government ca)
compiled lang like rust or ada would be lower
thats what a queer would say
it's the truth
its false
imagine it uses like 200 mb
waow 51 M
heres DNS stats rn
the lowest mem is static_server
is that btop
all servers have a maximum heap size of 250 MB
yes
ZGC is configured on each to remain under 100 MB
The Java Remote Method Invocation (Java RMI) is a Java API that performs remote method invocation, the object-oriented equivalent of remote procedure calls (RPC), with support for direct transfer of serialized Java classes and distributed garbage-collection.
The original implementation depends on Java Virtual Machine (JVM) class-representation m...
depends on what protocol you use for it
Oh so you could do it over HTTPS?
my instrumentation RMI server should be running now
just gotta write a client for it
very cool
What will you use it for?
instrumentation and updating primarily
Human Instrumentality Project!!
question mark
also also ACME
Evangelion
just realized i wont need UDP since RMI effectively already creates a command structure
RMI lookup (ACME) -> request through function
RMI lookup (DNS) -> request store path
write stores
<do acme stuff>
delete stores
return certificate through ACME RMI function
rust does have some instrumentation capabilities (via crates of course)
namely like
logging
logging and instrumentation are two separate things
but i expect you to have both
in the JVM instrumentation is mainly done using "beans" which are just little tidbits of data about an operation in the jvm
beans
With Rust you’d probably have to manually implement this all
What could make it easier is a proc macro that you decorate a function with to add in stuff
in ada most instrumentation is done with GNAT stuff
So like #[crate::instrument]
like the custom allocator that tracks leaks and stuff
ah ic ic
so much
its also possible to view MXBeans remotely
theres a tool like this in nearly every JDK distribution called jconsole
funnily enough this lets jconsole instrrument itself
Zig has a thing like this as well
The Zig standard library also has a general-purpose allocator. This is a safe allocator that can prevent double-free, use-after-free and can detect leaks. Safety checks and thread safety can be turned off via its configuration struct (left empty below). Zig's GPA is designed for safety over performance, but may still be many times faster than page_allocator.
__gnat_malloc copycat
I’m guessing Ada doesnt have as much instrumentation as Java though
somewhat
instrumentation in compiled languages is a little harder since
it's not like it's running in a VM like the jvm
but ada has certain attributes that let you view what compiler version was used for something
as well as whatever GNAT gives you and System
intellij has so many threads
i wish you could discern from virtual and physical threads
Typical Java app tbh
i wish it were possible to reduce them
biased
not biased
biased
not biased
based indeed
lol what
the data dont lie
because its dumb
if js is dumb how come 9 quintillion developers use it for their servers
Mmmm get this guy a 
shut up nerd and buy me bayachao merch

okay nvm
she has brain worms
Twitterに載せたもの再うpしました。
声!ノイズ多くてすみません!!!
収益化したらいいマイク買います。
よろしくお願いします。
kill me
and it's not even package.json or anything
pure code
I really don't know why in the everliving fuck I've decided to seperate backend and frontend
would've been so much simpler if I did it all in next.js
Did you see the Next.js vuln
BRO KEYBOARD
ok
Mm idk about that
It was built into next and couldn’t be controlled
Next did it automatically
@spare quartz The JVM was good for its time but its time has passed
jansoekioejfcooske
unfortunate you have to build GCC like that
I have no clue if this works or not because I gotta wait on terry for the damn sysroot
I hope I don't have to do --disable-libada
it depends if the rts for sparc-sun-solaris2.10 is distributed with gcc
I wonder where I could find that
ah, gcc/ada/libgnat/ has system-solaris-sparc.ads so I'm assuming there's a RTS for it
may or may not be full
solaris is a quite mature os though so
i expect it to be
well
that's the file
looks standard
this type of system file is required for every RTS
baremetal or not
in the future if you need your own RTS it's not super hard to make one yourself, as lot of the code is just copyable from any other RTS
just gotta wait on egg boy to send me his /usr and /lib folders from his sun workstation
but it can be a bit strange in some places where you gotta actually make your own impls (like the AFCB)
Air Force Code Base???
ada filesystem control block
air force command base
oh yeah i forgot you also have to make your own delay impl if you make an RTS
(hopefully your system has a clock or they just immediately return)
I would imagine a system with no clock is rare
even the embedded systems I've worked with have a clock
with more modern computers it's just "harder" to get to a clock
cause back then you could just use CPU I/O to dial the RTC but now it's uncertain if you really have one
so now you have to go thru PIC (antiquated), or APIC (hard)
unless you run on an os where you can just link to a native function
it's not strictly required you use a clock for delay though
as long as you know you can stall the processor in some way it'll work
busy loop :3
i use a loop of billions for early setup
GCOBOL
The COBOL compiler, gcobol, first appeared in GCC 15. To build the COBOL parser, you need GNU Bison 3.5.1 or later (but not 3.8.0). To build the lexer requires GNU Flex 2.6.4, the current version as of this writing, released on 2017-05-06.
The gcobol documentation is maintained as manpages using troff mdoc. GNU groff is required to convert them to PDF format. Conversion to HTML is done with mandoc, available at http://mdocml.bsd.lv/.
Because ISO COBOL defines strict requirements for numerical precision, gcobol requires hardware with 128-bit computation instructions. This requirement applies to both host and target. For integer and fixed-point computation, gcobol uses
Data type: __int128
meaning 16 == sizeof(long long int). For floating point, gcobol uses
Data type: _Float128
(On some architectures, GCC supports 128-bit floating point in software.)
gcobol has so far been tested on two architectures only: x86_64 and aarch64 with little-endian encoding.
gross
should note the other keywords in ada are uh
much, MUCH harder to make in your own RTS without a complex understanding of them though
(for example, OOP keywords like tagged)
oop 👎
even mathematical operations (div/mul/mod) have to be added by a custom RTS
so you get to deal with assembly
it would be neat to get ada running on the sparc station
but also...it would be so apt to have kotlin (a jvm language) running on it
depends
do you want a modern JVM or any JVM
if you want a modern JVM on a SPARC it'd require a lot of work, but if you don't care about that, old JVMs are available
yeah terry said like
jvm 8 💀
which was why I was like "oh wow kotlin can target jvm 8" earlier today
yea lo
bring me bayachao art
6 errors reported by DNS overnight
resource use sustained
of course its the router maxing out memory use
but everything else is under 70 MB
that should fix -1 and len < 0 errors..
NOT extended labels though
at least the TCP portion of DNS should no longer crash as a result of socket failure though
router screen got corrupted again somehow
the pace of development of these servers is incredibly quick
it isn't uncommon for maven/static to be running on BSL 1.16.2 while the latest is 1.20
thankfully the majority of BSL updates are protocol or stability specific, so it doesn't incur many errors
2.8
@timid quartz @pallid loom @rustic vine ... and to anyone else interested ...
https://www.bleepingcomputer.com/news/security/mitre-warns-that-funding-for-critical-cve-program-expires-today/
pretty bad news
So it's real
yeah
Love the us of a
hopefully this isn't sustained cause.. i think we all know how bad it can be
Worlds first third world country
fourth world country
damn
xmpp core rfc is yuge
and there's 5 "basic" rfcs
dns specification
unless overridden by EDNS(0) dns responses must be 512 bytes or less over UDP (excluding IP/UDP headers)
there is chaos on every screen
you
bayacghao
I wanna cross compile ada NOWWW
cross compile deez nuts
nah
yes/
no
i just use regular youtube videos as music
so good
BACK TO HTTP/2 DEV
unfortunately debugging whats wrong with my http/2 is magnitudes harder
how so
encryption
y not
self-signed cert requires a lot of work and makes me have to redo our JKS
mitmproxy means i have to scp essentially 4 times per connection test
since termshark doesnt support ssl master keys
what is even going wrong
mm
are you omitting trailing = in the header value
the what
The content of the HTTP2-Settings header field is the payload of a SETTINGS frame (Section 6.5), encoded as a base64url string (that is, the URL- and filename-safe Base64 encoding described in Section 5 of [RFC4648], with any trailing '=' characters omitted). The ABNF [RFC5234] production for token68 is defined in Section 2.1 of [RFC7235].
no multiple of 6
A SETTINGS frame with a length other than a multiple of 6 octets MUST be treated as a connection error (Section 5.4.1) of type FRAME_SIZE_ERROR.
yes i already do that
a byte is an octet in modern computers
yes die
stupid rust
not a rust problem
let me draw my furry art,
no go back to debugging http/2
im not sending anything yet thats the thing
just reading rn
HOW to draw... Thigh Highs
and...you're erroring on read?
yes due to misalignment
just dont get misaligned
use IntelliJ's wonderful debugger to help you
again not usable
bro h o w your code that is erroring literally must read the decrypted value
to OBSERVE
I CANT USE AD EBUGGER
WHY
IDIOT
it's MISALIGNED
the remote server is sending u correct data
omg
(unless you also wrote the server)
so
you have a computer
that is sending requests
and you have a router that is reading them
and erroring
yes
(id need to make a JKS and host another server so the router has backing)
debug it and watch it prepare the data to send
and you call me lazy
nah my website would never have such terrible design
yeah cause it already is terrible..
(HTTP/2, Frame) SETTINGS, 0x00000000 [], SETTINGS #: [3]
(HTTP/2, Setting) SETTINGS_MAX_CONCURRENT_STREAMS [100]
(HTTP/2, Setting) SETTINGS_INITIAL_WINDOW_SIZE [65536]
(HTTP/2, Setting) SETTINGS_ENABLE_PUSH [false]
0x00, 0x00, 0x12
0x04
0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x03 | 0x00, 0x00, 0x00, 0x64
0x00, 0x04 | 0x00, 0x01, 0x00, 0x00
0x00, 0x02 | 0x00, 0x00, 0x00, 0x00
(HTTP/2, Frame) WINDOW_UPDATE, 0x00000000 [1048510465]
0x00, 0x00, 0x04
0x08
0x00
0x00, 0x00, 0x00, 0x00
0x3E, 0x7F, 0x00, 0x01
first time using java, kinda nervous
time to add Range support
mm @spare quartz I might not be able to add runtime support
idk I keep banging my head into the wall with "Configuration sparcv9-sun-solaris2.0 not supported"
source ?
what is
compiling this damn thing
to make ada run on a sparc
@spare quartz if you have any ada resources it'd be great if you could ask
i can help you im just a little preoccupied
oh?
do you have a cross compiler from your architecture to SPARC
what do you know about this I'm curious
uh no
that's what im trying to build
I have the sources for gcc-12.3.0 and binutils-2.38
the binutils built fine
but gcc refuses
yes GCC is uniquely horrible to build, i can't actually help much there (i thought you already had a compiler ready)
i would suggest https://wiki.osdev.org/GNAT_Cross-Compiler, but
last i tried it it didn't work
and the article is extremely antiquated
Installing GCC: Configuration
so many thigns..
they should've just made a compiler in kotlin
no clue if this'll work
this builds it without runtime
ig I could try removing the flag that does it but
im not exactly certain how runtime builds work if im being honest
i suppose it depends on your GCC source...
this is a massive pain
cross compilers are great if they're provided OR if they're for systems that arent like decades dead
sob
why doesnt this work
theres gotta be like a mailing list or a forum I can post on
uguyggyuh
im probabluy gonna die of exsanguination whilwe proigmming
no success..
im crashing out
I followed the steps on the osdev wiki thing
STILL "not supported"
told you they were antiquated
the stuff on there is like
from 2012 unironically
im not sure on wjhat to do next though
I give up on trying to make Ada run on SPARC
would you like a consolation prize
no
pleaseee
the only realistic thing now is....getting onto the sparc directly and probably having to build gcc with ada support with the toolchain that's already there
do you think you're working too hard for a project that you probably won't use
BAYACHAO
PSD
FILES
AQUUUUIIIIRIRRREEDDDD
Yes but like
hold on bayachao lore just dropped and this is crazy. but continhue
Your thing is making a server for every possible server under the sun
My thing is getting things to run in unorthodox manners
multiple use of "server," clarify
did you mean protocol for second use of server
but that's going to drive you mad and you know it..
Ok well more generally your thing is protocols
IT ALRWADY IS
SEE
eg rust on teensy 4.1 no std, ada on SPARC
I just wonder where I could ask questions about compiling gcc
the gcc mailing list?
urghhhh boomers
and I doubt anyone would respond
“ew some kid is asking how to do X, I remember the answer to this on a thread from 25 years ago, if this kid can’t find it then he shouldn’t be compiling my gcc”
okay well i imagine they're not like stack-overflow user petty
the guy who emailed me back about a GCC bug report seemed nice
hmmmm maybe I’ll try then
part of me really wants to write a compiler for sparc now, but
im unfortunately not dumb enough to know what that entails
first: custom mail server to receive mailing list replies? :3
with your 9 quintillion crates thats easy
Yknow I was gonna say “who said I was doing it in rust” but then you reminded me you have one in Ada and Kotlin so I gotta complete the trifecta
i think i found out where japanese people get mega freaky
DO NOT DISTRIBUTE THE FOLLOWING TEXT FILE
im actually gonna go lay down soon
but ... hopefully ONE day ill be able to understand that text file withuot needing a translator (probably not)
with rust’s superior development speed, by the time you wake up, I’ll have made SMTP 2
i make servers in single days
you're gonna have to be a little better than that .

