anchpop

Joined 11 years ago

Karma
2,639
Submissions
860
Member since
Nov 17
[ my public key: https://keybase.io/anchpop; my proof: https://keybase.io/anchpop/sigs/wyyyPQEZu9nQAtEteuBWP1QFnPvqqXEo-xMB1g-hGdY ]

my blog: blog.andrepopovitch.com

View on Hacker News

Recent submissions

Comment · 1 year ago
I did that (distracting subtitles) on one of my videos and it had a very negative response. I won't do it again, but I was puzzled because I find it much nicer than the traditional subtitle format personally. It's easier for my brain to focus on. (And no one in my test audience minded.)
Comment · 1 year ago
I love the liquid glass look and have helped multiple non-techie people install the developer beta just because they were excited to experience "the frutiger aero update". (frutiger aero is a bit of a meme on tiktok right now so that might be why.) It's fun to hate on apple and the ridiculous level of glass on the control center was pretty funny, but I do think they knocked it out of the park with this one, in terms of mass appeal.

The one issue is that everyone on the beta says their phone is slower now. Which is probably not because of the liquid glass effect, since I think that should be doable with just a couple of texture lookups. (One of those funny things about computer graphics is that often the most visually impressive effects are the simplest computationally - compare this to a "proper" gaussian blur, which is quite expensive.)

Comment · 1 year ago
Comment · 2 years ago
I think you're right. I produced a small rust program to test it by iterating over all 32-bit floats: https://gist.github.com/anchpop/30f119efd8d4c29a6a99202d57a5...

Output:

    Count between 0 and 1:    1065353215
    Count between 1 and +inf: 1073741824
    Ratio: 1.0
But a more theoretical approach will probably be needed to see if the same ratio exists for 64 bit floats.
Comment · 3 years ago
Who lost? Everyone who got laid off wouldn't have had at a job at all if they weren't hired, and they probably got severance packages for more than the average American makes in a year.
Comment · 4 years ago
On Ethereum you can you decentralized tumblers like Tornado Cash
Comment · 4 years ago
ZK rollups are not anywhere near 14 years old
Comment · 4 years ago
They shouldn't even all be on the same computer. Ideally they would be engraved in titanium and inside people's safe deposit boxes
Comment · 4 years ago
In this case, the weakness was that the keys that controlled the bridge were somehow stored insecurely. When attackers gained access to the keys, they were able to steal from the bridge. In a properly-implemented rollup, there are no keys to secure, so this attack vector is ruled out.

But more broadly, there is really nothing else with the same security properties as a smart-contract-enabled cryptocurrency. Paypal will delete your account any time they want, Visa and Mastercard will blacklist whatever industries they feel like blacklisting, etc. If you want a system that's decentralized and where these attacks aren't possible, you have no alternative. The problem is that current blockchain-based systems can only handle a certain number of operations/second while remaining decentralized. The appeal of scaling solutions like ZK-rollups is that they give us the same security properties as the main chain without any security compromises (relative to the main chain). That's all conditional on their code being correct, but given that there's such a large payout to hacking e.g. bitcoin or ethereum or zksync and it still hasn't happened, we can guess that the coders have done their jobs well and such problems are at least very difficult to find.

Comment · 4 years ago
Right. It's possible to conceive of a rollup, particularly a zk-rollup, without anything like a master key. But current rollups do have those keys. ZK-sync for example has two, one used mostly used for upgrading the smart contract that has a 14-day withdrawal delay (or something like that) and one for use in case of emergency that has no withdrawal delay. If the second were compromised, it would lead to all the money stored in the rollup being stolen. But there's no reason in principle that either of these are necessary.

ZK-rollups are awesome because they don't introduce any trust assumptions (except for the master key issue, which is just an implementation detail). The only risk is current zk-rollup designs is that they could censor certain transactions by never including them in a "batch" (the rollup equivalent of a block), but with unpermissioned rollups like the one I think Polygon has even this issue is mitigated

Comment · 4 years ago
Right now they're immature, but I'm hopeful that advancements in ZK-tech will allow practical ZK-rollups. ZKSync already has a zk-evm testnet running (which I believe is based on zk-llvm), so we're close. Currently all the big rollups have master keys which can be used to steal all the money deposited by them, but there's no reason in principle they have to have this. Polygon has permissionless rollups, so I'm quite hopeful that they'll be a viable trustless permissionless scaling solution soon.
Comment · 4 years ago
what kind of volume does tornado.cash process? If it normally processes e.g. $1M/day, it'd take a while to use it as a mixer right?
Comment · 4 years ago
I’m curious about handshake vs ens. The thing I like about ens is that it’s on ethereum which should make it harder to 51% than handshake (which is a relatively small PoW chain iirc). Of course with ethereum you need to pay gas fees which are out of this world (I paid $55 gas for my ens name). I’m hopeful that in the coming years ens will be able to reduce fees by partially transitioning to an L2 once zk-L2s mature, but we’ll see

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton