Show HN: We built a terminal-only Bluesky / AT Proto client written in Fortran

145 pointsposted 5 days ago
by FormerLabFred
(github.com)

81 Comments

lzhgusapp

4 days ago

Love seeing niche Show HN projects like this. The choice of Fortran is wild but that's what makes it fun. As someone building small Mac utilities, I appreciate any project that proves you don't need a massive stack to ship something useful.

FormerLabFred

4 days ago

Thanks for the feedback!

We got a spec for Assemblersky. Will be a weekend project. Cancel Easter :) And probably Midsummer too…

We got image composer and decoder plugged in in dev env, but will let this first open version breathe first. ASCII or early Apple algo

Fortran-fast feed builder probably in next open version. It crunches fast

Morning, 9 o’clock in Sweden, coffee and check-the-feed-on- Fortransky-time :)

bombcar

4 days ago

COBOLsky. It’s enterprise webscale.

cat-turner

5 days ago

out of curiosity, why fortran? no disrespect. I wrote a lot of scientific software in the earlier days of my career and I learned fortran to update ocean modeling software.

FormerLabFred

5 days ago

You are not the first one to ask :)

We built Cobolsky. Will go public soon. Parallelly too curious on Fortran. The world is better with a Fortran-based social network client in it :)

When we are building the feed composer, in next version, Fortran will be great for the algorithm etc.

Keeping the ancient languages alive. I built some Cobol stuff many years ago. Back at it again. Rusty.

Both Cobolsky and Fortransky looks great on Swordfish90’s cool-retro-term, but we are building our own terminal for Fortransky too. There is a blog post with screenshots over at Patreon/formerlab

Can’t get enough Fortran

embedding-shape

5 days ago

> The world is better with a Fortran-based social network client in it

If you don't mind me asking, why is the world better with more Fortran-based software?

FormerLabFred

5 days ago

Our modern languages are built on it, and it’s incredibly fast,

so it deserves to be kept alive. We owe a great deal to the people who wrote it in the 1950s I guess

mountainriver

5 days ago

This thread makes me happy

johnisgood

4 days ago

I know right! I was pleasantly surprised. It has to be one of the greatest news I have heard in a while.

embedding-shape

5 days ago

> Our modern languages are built on it

It's part of the lineage, yeah, probably started with Algol though? Fast I guess is always nice, but I'm not sure that's enough to keep it alive solely for that, at least to me.

mathieudombrock

4 days ago

I think the best answer you're really going to get here is that it's cool and fun to learn and use old languages.

embedding-shape

4 days ago

I'd agree with all of those reasons! I do so myself as well, was just specifically curious about the "The world is better with a Fortran-based social network client in it" part. Don't get me wrong, I've spent too many nights learning "dead" languages too, but never with the idea that the world would be better if I published more code in these dead languages, it's just for my own gratification and learnings.

hedora

4 days ago

I came here to suggest COBOL as a better fit, then saw your comment a few levels up in this thread.

Out of curiosity, does your implementation use CODASYL?

(For people that don't pay much attention to historical software systems, most CODASYL implementations were similar to JSON document databases, so going that way isn't as crazy as it sounds.)

FormerLabFred

4 days ago

Great comment! Thanks!

No CODASYL, the JSON parser is hand-rolled Fortran with a depth-tracking key scanner

CODASYL not a crazy direction for the feed composer

Got more depth to explore here, still early :)

hedora

2 days ago

CODASYL is deeply integrated with COBOL; I'm not sure it ever even worked with fortran. (I was asking about the new COBOL version you mentioned.)

cat-turner

2 days ago

Ya know what? Rock on Fortran Friend. Keep doing what you are doing. We need more nerds passionate about Fortran.

Melatonic

4 days ago

How do COBOL and Fortran compare for something like this ?

FormerLabFred

4 days ago

COBOL is more painful, Fortran better.

Good you raised the topic, can write a blog post on it when we ship Cobolsky. Will be a proof of concept repo. Fortransky is the one

(If AT Proto did fixed-width records instead of JSON, COBOL would be formidable)

bee_rider

4 days ago

Which version of the language is it? It looks like you used Fortran90 at least (modules are used), which is pretty old, but not totally ancient like Fortran77.

Anyway there are also 2018 and 2023 versions…

fortran77

4 days ago

I still make a living from Fortran77 work, though much of it is converting it to Matlab.

bee_rider

4 days ago

Building interfaces or full converting? Either way sounds like a dream job, haha.

johnisgood

4 days ago

Thank you for choosing Fortran. Seriously, I mean it.

Someone else already said this, but it is awesome because it proves that you can write useful software using languages others consider ancient or dead and therefore "unusable".

Keep up the great work!

enriquto

5 days ago

> why fortran?

why not? the language is straightforward and loops are fast. It is portable and your code will work unchanged for the next 50 years. It may be a bit verbose, but that's not a big deal with today's tooling.

FormerLabFred

5 days ago

Fortran will survive the cockroaches even, when the world 404s

foxglacier

4 days ago

Yea na, Fortran is pretty compiler dependent and there are a lot of compilers. Already old Fortran code used all sorts of now-dead proprietary compilers and can take a huge effort to get it to compile on modern compilers or even modern computers. Modern code might use Gfortran which sometimes makes breaking changes so that's not an option. Perhaps if everyone uses the latest shiny new Flang or whatever, then it'll finally last 50 years? Not likely, given the history.

pklausler

3 days ago

For a standardized language, Fortran isn’t very portable across compilers. GNU Fortran has done a great job supporting legacy features, and I hope that our work in flang-new has made it easy to port to, as well. I basically ignored the zealots who wanted flang-new to be a strict compiler by default. The hobbyist project LFortran is quite the opposite, and will yell at you by default for perfectly conforming variations in keyword spelling. For those who like that sort of thing, that’s exactly the sort of thing that they like.

pklausler

5 days ago

Your code will work unchanged until you try to change compilers or your compiler adopts a J3 breaking change to the language.

kergonath

4 days ago

> your compiler adopts a J3 breaking change to the language

Like all the 3 of them they added in the last 30 years, and that compiler vendors are not enforcing anyway because they don’t want to annoy their users?

Windows’ backward compatibility is a joke compared to Fortran.

pklausler

17 hours ago

We're definitely enforcing some of them. But the latest one from F'2023 isn't going to be.

deadbabe

4 days ago

Great, can we see some benchmarks?

FormerLabFred

4 days ago

Cobolsky holds the record for most surprised looks per line of code :)

Fortransky benchmarks pending

the feed scorer will have real numbers worth reporting

Whenever time allows in future: Fortran vs Go vs Python

pklausler

5 days ago

maybe they weren't really concerned about portability or a decent standard?

FormerLabFred

5 days ago

It’s keyboard navigation only, and we got the Bluesky firehose raw straight into the Rust decoder. Or we switch mode to Jetstream with m+EnTER :)

You hit l+ENTER to like a post. If anyone replies from Bluesky, we hit n+ENTER and see the notifs. And so on

Fortransky is 70% Fortran, rest is Rust, C and a tiny Python helper

zoom6628

4 days ago

Brilliant in every possible way. Fortran was first language I learnt at high school in its "PORTRAN" variant.

FormerLabFred

4 days ago

:) We tried Abacus Fortran on the C64 back in the days and were allowed to stay indoors during breaks doing Basic stuff too on school computer.

One of the guys wrote 8xxROM some 10 years later.

Today known as U−Boot :)

jrmg

4 days ago

It seems you’ve written an extensive blog post about this, but I don’t see a direct link anywhere!

https://www.patreon.com/posts/fortransky-we-in-153457794

FormerLabFred

4 days ago

To the repo? It’s in the blog post head, in section under first screenshot,

https://github.com/FormerLab/fortransky

jrmg

4 days ago

No, I mean the other direction.

The repo is what’s this HN post links to, and I don’t see a link to the blog post in it or in other comments here. The blog post is interesting!

FormerLabFred

4 days ago

Ok got you. Good point, added to repo README

worldmerge

4 days ago

I’ve never seen fortran before. Wow, that looks more readable than a lot of modern languages.

bee_rider

4 days ago

Modern versions of Fortran are reasonably pretty.

It’s a fairly nice language. You can probably get better performance out of C/C++ with unlimited effort. But, it is really nice for allowing computational scientists to get, like, 95% of the way there.

I think it actually suffers from the reputation as this ancient/super hardcore performance language. The name comes from “Formula Translating System,” which implies… it was written for people who speak human languages first!

sitzkrieg

4 days ago

ive maintained a simulation software where the core is written in fortran. its using some intel math library that is expensive that i cannot recall, does immense calculations and makes faster binaries than c on every compiler we tried

extraduder_ire

4 days ago

Have you tried using the restrict keyword everywhere you can in c?

In Fortran, arrays are not allowed to overlap which allows some optimisations. c has rules in the spec about how memory accesses must occur, and overlapping arrays being possible prevents some compiler optimisations. The restrict keyword is you promising that the memory at some pointer won't be accessed through another pointer.

You can compare two implementations in Fortran/c using godbolt to see how each of them compile.

pklausler

3 days ago

Dummy arguments may not alias anything else, including other dummy arguments, in conforming Fortran if either are modified, with some exceptions. But there are many ways in which arrays can overlap that don’t involve dummy arguments. GNU Fortran optimizes ASSOCIATE names as if they were free of aliasing, too, but that is a bug.

bee_rider

4 days ago

Interesting.

MKL is Intel’s famous numerical library (it includes things like BLAS “basic linear algebra subroutines” and fast Fourier transforms). It is availible for free, but IIRC they had some support plans maybe, maybe that’s what you are remembering?

It is closed source, but you can look at the source of the best open source competitor libflame/BLIS, and see that most of the performance comes from C and assembly.

It is difficult to beat “unlimited effort” C, but not many program really justify that treatment.

foxglacier

4 days ago

MKL used to not be free. You had to buy it from a local reseller for a few hundred dollars per developer.

cmrdporcupine

4 days ago

Modern dialects of Fortran, sure.

Old school Fortran from before they added cleaner and more structured elements is terrifying.

hk1337

5 days ago

Are there any other AT protocol apps that aren’t derivatives of bluesky? By that I mean, not social media feed related, twitter clone.

FormerLabFred

4 days ago

Don’t know if people are building GitHub-like systems, offline-first app sync etc

Many devs reuse schema and write some twitter/bsky clone

Kind of search engine for my Blueaky likes

Gotta get off the timeline, germ has a messaging app, then there is Tangled and some more

We do Fortransky and Cobolsky now, got more ideas for the protocol than time :)

linolevan

4 days ago

https://tangled.org/ <--- GitHub on ATProto

that's all I'm aware of

(edit) Oops, just saw that you mentioned it, confused by your first line then. Tangled is awesome!

tjuene

4 days ago

i am currently working on a dropbox alternative: https://dropb.at

Hamuko

4 days ago

How do you plan on differentiating it from just a trivial setup of an FTP account that's mounted locally with curlftpfs and has files in a SVN or CVS on the mounted filesystem?

tjuene

4 days ago

good point, i havent considered this at all. its basically dead on arrival

vissnia

4 days ago

I'm curious if you used AI during development and how it handled Fortran

FormerLabFred

3 days ago

Great question!

This was quite straightforward to code ’manually’, but we got stuck for some hours when trying to send the first message. CPU topped, fans spinning, no message through. Analyzed, and after a while when we got same block the nth time, someone proposed Claude.

So Claude got the repo, prompted Claude:

”You got one job, why isn’t message through and why CPU usage high?”

He didn’t solve it. Probably could have if we kept Clauding.

But we found the problem. The issue was that the code was hitting the terminal 300 times trying to send the message :)

For clarity reasons Claude got the job to write the README, personally I don’t like that approach, but we got many other projects so time management basically.

We are building Cobolsky too.

Right now we are building the part to get images through using Apples old algo. Will be 576x720 using the Föoyd-Steinberg dither, 1984 Apple algorithm :)

Uptrenda

4 days ago

Whenever I hear about this AT protocol I think about the text-based commands to control cellular modems. I can't be the only one, r-right guise... r-right?

h4ch1

4 days ago

It's always nice to see production codebases in languages that you've never used but are interested in.

Tangential, but to the author, are there any FORTRAN codebases you feel are well designed?

FormerLabFred

4 days ago

Good reminder to dig around. Will check. We picked up the Fortran manual and just went for it.

The original Manual exists as a PDF. Was it in a Stuttgart uni URL? Just a search away.

Late in Sweden, gotta Fortran tomorrow. Happy to continue discussion here tomorrow.

fortran77

4 days ago

Finally! Some relevant news here. I wondered where all the other Fortran people were hiding. I thought I knew all the other BlueSky Fortraners.

FormerLabFred

4 days ago

Forecast: Fortransky on Tangled tomorrow Sunday

jrmg

4 days ago

There seems to be a bunch of Rust in there too?

FormerLabFred

4 days ago

It’s early version yes. It’s 70% Fortran, next version more Fortran and we will add x86-64 assembly decoder for AT Proto firehose frames

uberdru

5 days ago

The world is a better place for this app. Wonderful!

FormerLabFred

5 days ago

:) I mean there are still some people alive out there who never saw a web UI when beginning dev. They get a bit nostalgic, missing their 286

It’s fun and it is appreciated by them, and the young ones who are curious

isodev

4 days ago

On ATProto: it’s funny how we never learn the lesson:

- VCs band together to fund something shiny.

- Devs love shiny, helping spread the something.

- VCs enschitify it to get their coins back.

youhai

4 days ago

From my experience building browser automation tools, the biggest challenge with most Chromium-based solutions is that their TLS fingerprint is a dead giveaway. Firefox-based approaches tend to fare much better against JA3/JA4 fingerprinting.

The key insight is moving fingerprint spoofing from the JS level (which is itself detectable) down to the native C++ level. It's a fundamentally different approach.