andutu
Joined 7 years ago
Karma
179
Submissions
26
Member since
Nov 5
https://andutu.dev
View on Hacker NewsRecent submissions
Comment · 1 year ago
Pretty cool, sound similar to what Barrelfish OS enabled (https://barrelfish.org/).
Comment · 1 year ago
On the one hand writing logs can be tedious. Essentially logs are breadcrumbs signifying when significant state changes have logically taken place. From what I've seen, logs are added after every few lines and I always fantasized about creating a language where logging is "automatic".
On the other hand, writing logs is a skill worth mastering. Wrote too many and when a service crashes you have to shift through lots of noise and potentially miss the signal. I once went down a rabbit hole trying to root cause an issue in a Gunicorn application that had custom health check logic in it. An admin worker would read health check state from a file that worker threads wrote to. Sometimes a race condition would occur, in which an error log would be emitted. The thing was, this error wasn't fatal and was a red herring for why the service actually crashed. Of instead it would have been logged at the debug level a lot of time would have been saved.
Fine let LLMs write code but take logging seriously!!!
Comment · 1 year ago
There is an excellent paper and talk on how Google's TPU cluster is managed: https://www.usenix.org/conference/nsdi24/presentation/zu.
Comment · 2 years ago
I've skimmed some literature on Quantum Crypto and from my understanding the outstanding issues currently are 1. How to make these work over long distances and 2. How to implement features found in PKI authentication (though QKD schemes are theoretically secure against MITM attacks, there still isn't a quantum cryptography scheme yet to ensure that you are talking to a non-adversary). There have been advances with the 1st problem, but the 2nd is trickier. Tbf, you don't strictly need PKE to have secure communication between 2 parties (look at Section 6 in https://signal.org/docs/specifications/sesame/).
A lot of real world implementations of quantum crypto have been with respect to satellite communications, which makes sense. The satellite is usually built by the same actors who set up the communication links from the ground the satellite and quantum particles can be transmitted by laser. But as the article points out, it probably won't see widespread use for a while. There was a paper that came out recently called "How (not) to Build Quantum PKE in Minicrypt" (https://arxiv.org/abs/2405.20295) and from my limited understanding of it, a quantum PKE system will likely have very little components from classical crypto incorporated into it. Not to mention that specially built devices have to be installed at ISPs, data centers, etc. for this to work.
Work in this space is valuable as a hedge against a world where all conventional crypto is broken. It also helps advance work in quantum mechanics more generally and other fields in physics and it's generally very interesting :)
edit: grammar corrections
Comment · 2 years ago
Somewhat related, but there is a great talk by Andrew "bunnie" Huang on hardware supply chain security:
https://m.youtube.com/watch?v=RqQhWitJ1As&t=3s&pp=ygUZQnVubm...
Comment · 2 years ago
My boyfriend woke up one day to find out they couldn't pee. After a few ER visits and an MRI, they determined he had Stage 2 bladder cancer. However, because they had the audacity to seek immediate medical treatment for a life threatening health issue, they have now been saddled with thousands of dollars in medical debt. I had to personally dip into my earnings from my Big Tech internship to help pay some of it off. Luckily through their job they have some of the best medical insurance in the state and have been getting treatment at UT Southwestern. They pay little, but there was an instance where one of their physicians had to appeal to insurance to get a scan done (thankfully the insurance acquiesced). It doesn't make sense and all this debate around single payer healthcare is just obfuscation and distraction from investigating actual solutions. The vast majority of doctors genuinely want to help people and are more I interested in practicing the skills they've honed for decades rather than deal with faceless automata at health insurance companies who deny claims upon a mere glance. As long as people like Rick Scott can not only get away with introducing inefficiencies in the system, but defraud people and get away with it with no consequences other than personal enrichment, we are doomed.