leijurv

Joined 10 years ago

Karma
486
Submissions
71
Member since
May 1
[ my public key: https://keybase.io/leijurv; my proof: https://keybase.io/leijurv/sigs/w7C9Q0ei9POBhzwXcl36pWOTknw0WeEo5yNlqo497G8 ] Verifying my Blockstack ID is secured with the address 1PjWQMkKBD3JJJuYMEXnSYZ9pvpbLTQVqS https://explorer.blockstack.org/address/1PjWQMkKBD3JJJuYMEXnSYZ9pvpbLTQVqS
View on Hacker News

Recent submissions

Comment · 1 year ago
OP did actually host a tracker.

"I then started the tracker. After about an hour, it peaked at about 1.7 million distinct torrents across 3.1 million peers!"

Comment · 1 year ago
Here's the submission that won the Hutter Prize in 2021: https://github.com/amargaritov/starlit It uses a LSTM to predict the next token lossily, then uses https://en.wikipedia.org/wiki/Arithmetic_coding to convert that to lossless compression. Lossless compression can definitely leverage a lossy compressor, such as via arithmetic coding. Also see: https://en.wikipedia.org/wiki/Context-adaptive_binary_arithm... which has a simple "Example" section - imagine if the top prediction made by your neural network was correct, you emit "0", if the 2nd was correct, you emit "10", if the 3rd, "110", if the 4th, "1110". As you can see, this is lossless, but the fundamental prediction is lossy, and the better that prediction is, the better the compression. (In actuality, you wouldn't waste your 1 bits like this, you'd use arithmetic coding instead).
Comment · 2 years ago
I suspect it was, given that they've now deleted their comment.
Comment · 2 years ago
`for char in message: encrypted_char = ord(char) ^ (shared_secret[0] % 256)`

This is not real encryption, it picks only one byte of shared secret and XORs it into the plaintext. Therefore, there are only 256 possible decryption keys to check, which is trivial.

Instead, you'd want to use the shared secret as a key to something strong and symmetric like AES.

Comment · 2 years ago
It was a one-way collaboration, in that we referenced their discoveries and code such as LattiCG https://github.com/mjtb49/LattiCG, but they were unaware of anything we were doing until now. https://twitter.com/admiral_stapler/status/17806748612594609...

Naming Baritone after Fit is actually a coincidence / joke, the repo github.com/cabaletta/baritone was the result of random brainstorming for something untaken. We only later realized it described Fit and thus added that to the readme :)

Comment · 2 years ago
I'm just a regular SWE! Infosec or algorithmic trading - maybe someday.
Comment · 2 years ago
Yes, absolutely :) that's why we went to FitMC to make the video, he always delivers.
Comment · 2 years ago
I believe that may be the spectral test https://en.wikipedia.org/wiki/Spectral_test which I mentioned in the explanation when showing the lattices visually
Comment · 2 years ago
Comment · 3 years ago
No that's not at all how this works. The iterator could "return" a boolean by passing a boolean to "yield". The return value of "yield" just indicates if the generator should continue, or stop because the actual loop has exited (due to break or return).
Comment · 3 years ago
Comment · 4 years ago
One person one vote is democracy, one dollar one vote is web3. Very different.
Comment · 4 years ago
> I wonder how Blockchain explorers deal with more complex scripts. For example on blockchain.com one can look up coins by putting an "address" into the search bar. But how would one look up coins that are not locked by a script that puts the coins under control of a certain key? Or a script that puts them under control of multiple keys?

The reveal of the script is actually time-delayed. See https://en.bitcoin.it/wiki/Pay_to_script_hash

Essentially, you "send" the Bitcoin to the hash of the script, then whenever a transaction spends those coins it must reveal the script (whose hash must match), as well as some data (normally a digital signature or two) that fulfills the script's conditions.

So, the block explorer doesn't get to deal with it until the coins are spent, up until that point all it knows is a hash, which is represented as beginning with a "3", to differentiate it from the simple single-key addresses that begin with a "1".

When it's spent, a block explorer could show the revealed script contents if it wanted.

Comment · 4 years ago
Some mistake or miscommunication happened there, as Wikipedia does have a policy that people can be cited for information about themselves, the policy is called SELFSOURCE. See: https://en.wikipedia.org/wiki/WP:SELFSOURCE

Perhaps the issue was that Philip Roth was unable to sufficiently demonstrate his identity? Of course, Wikipedia can't take a random editor's word when they say "I am this person and this is the truth", then anyone could say anything. There has to be some citation, for example I've seen someone cite a tweet for simple biographical information (e.g. "today is my birthday").

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton