
I was planning on holding off on this announcement until the project was farther along, but I'm excited to announce that I've been working on a new programming language called Phil, and the team at @Higherdotfun_ is supporting its development.

Parsers and parsing accessories. Para-academic. Rootless cosmopolitan redneck. Writes for @Status451Blog. 'Juvenile but not white supremacist' -- @splcenter

I was planning on holding off on this announcement until the project was farther along, but I'm excited to announce that I've been working on a new programming language called Phil, and the team at @Higherdotfun_ is supporting its development.

today showed exactly why Phil matters, and why the philcoin community keeps building for it $15.5M left Lighter today and the protocol wasn't even hacked. what's being reported: two market maker API keys leaked. those keys aren't supposed to be able to send money out. so the attacker made a pool from the victim's own account, bought 99.99% of it for $10k, and moved the funds in $1M at a time every step was allowed. the rule was the bug. it checked where the money went, not who owned it after Phil is built for exactly this. you write the rule once, in one line: after a move, the owner still holds all of it. and the language refuses anything that breaks it so within hours of the news we wrote that rule in Phil and ran the same drain against it refused Phil doesn't stop a key from leaking. it stops a leaked key from reaching the money Meredith Patterson is building the language that refuses this kind of loss. the philcoin community is the one showing it, live, the same day it happens still unconfirmed by Lighter and our page says so. will update when the post-mortem is out philcoinhigh.github.io/Phil-… $philcoin Ea5dBrJZN58yV1nm1Q4DybJNYKeTvwpb2GEkXyW64bUH

This is really turning into the Java of proof-carrying code: specify once, verify independently, implement anywhere, run on anything. Except we have generics from day 1. And hopefully less boilerplate. 🙏

Similarly @highwiredotfun submitted an example PR today. CI is running on it now. We're taking it as non-normative and it will become a normative example as the features they need land. A Phase 2 tracking issue is already in place.
We wrote one round of The Line's wager protocol in @maradydd 's Phil! The checker even bounced our first draft 👇

I am amazed and delighted that Phil users who want to use exotic backends can already use the language in consequential ways before that backend even exists. Seriously check out what @fewinputs and the Philcoin community have been up to. Great stuff!
philcoin community doesn't wait around. we build today I got "my coin" working and here's why it's big paste any solana coin, connect your wallet, and from one page you can: - lock supply until a date. no cancel, no early unlock - open a burn vault for the fees. the SOL only leaves when someone burns the coin - run a giveaway where a future block picks the winner, not the dev then you share the link and anyone can check all of it without a wallet every one of those actions goes through code that Meredith Patterson's Phil compiler produced. 3 days ago Phil on solana was a program that returned 42. now it's a launch kit any coin can use no team allocation, no roadmap promises, just a community shipping real tools for a real language every single day devnet for now, unaudited, mainnet is next philcoinhigh.github.io/Phil-… $philcoin Ea5dBrJZN58yV1nm1Q4DybJNYKeTvwpb2GEkXyW64bUH

Gm, for certain values of morning! I slept in a bit, but the hamsters stayed busy. 29 PRs landed yesterday on full autopilot. The grammar correspondence proof is on its fourth day of debugging, but that's not surprising because it's enormous. We'll get there.

The weather outside is pretty meh today, so that makes it a great day to stay inside, snuggle the cats, and watch netflix while the hamsters chew through auditing, refinement, and correspondence.

Gm everyone! Another 30 PRs landed yesterday on the way to Phase 1 closeout. It's Sunday, so I won't be working today -- but the hamsters never sleep and never get tired.

I also built some more robust air traffic control architecture in anticipation of expanding the number of runways during phase 2. The Rocq refinement hamster has had a hard day, but it's working through a very big proof, one of two remaining for Phase 1. It doesn't block 0.1.1.

Gm! Phil Phase 1 closeout marches on, with 35 PRs landed in the last 24 hours. Most of this has been audit reconciliation but there's also been a good amount of formal semantics documentation.

Updated the assets library with some more fun drawings of Phil. Also working on hamster air traffic control but the mascot pics are way cuter. github.com/abiggerhammer/phi…

Early phase 2 is already looking like I'll be able to do a 7-lane fanout, which is ludicrous and I'll probably end up blocked on github actions runners again. C'est la vie.

Gm everyone! Looks like the hamsters had a good night. 13 PRs landed while I slept, which is a pretty typical cadence for 2 simultaneous lanes. Audits are difficult to parallelize, so they aren't the Mecha Hamster Ball's finest performance, but that's ok. github.com/abiggerhammer/phi…

It's been a good day. Phase 1 progresses steadily toward closeout and the hamsters have the wheel while I knock off for the night. (I promise I will explain the hamsters thing relatively soon. It will be a blog post.) github.com/abiggerhammer/phi…

It's coming up lunchtime and the Mecha Hamster Ball is rolling. Hope everyone's having a good day!

Thanks for all the interest in Phil, everyone. I'm going to bed now and will see to other questions in the morning. Good night!