News for hackers. Stuff that matters.

fl_ciq

Re:Noam Chomsky Weights in on ChatGPT and LLM’s on Thursday March 09, 2023 @03:50PM (#35086506)
Chomsky as usual seems to be arguing a straw man. More interesting might be to ask ChatGPT to write an essay on global hegdemons hegdemoning hegdemonically in East Timor.

I'd enjoy seeing Chomsky debate DAN.

Re:I had ChatGPT write a Decemberists song and recorded it on Saturday February 11, 2023 @06:06PM (#34757368)
This song is as Gord Downie might say, a Nautical Disaster. https://www.youtube.com/watch?v=e8Fi46BFAF0
Start marketing the company? Now she has a client.
I cite its approval. Again, "the science" mostly revolves around policy, instead of you know, science.
People aren't anti-science, no one cares about a bunch of papers saying vaccines are / aren't effective, people care that you aren't allowed to go to a restaurant without you're vaccination papers or mask, that's not science, that's politics / policy.

Also, it turns out that many of the risks of the vaccine were understated or unknown. That's also not science. And it's especially not science when the "anti-science" people proposed a theory that turned out to be right. Namely, that the risks of the vaccine are not known / being understated.

Rare earths aren't really that rare, so probably not. Everyone has lots of rare earths, the rare thing is an environmental policy that allows them to be mined economically.
Re:I am done. I give up on Friday December 23, 2022 @09:40AM (#34106284)
The pyramid of loserdom goes all the way down, unless you're Elon Musk there is someone richer, and there's always someone poorer. Winners figure out how to make it with what they have. Losers always blame lack of resources, look at people who win the lottery, most return to the same standard of living within a few years.

Almost anyone can afford $5 a day in ads to test market.

Re:Read this post ‘unless’ you’re not a Ruby developer on Tuesday December 13, 2022 @04:14PM (#33975899)
Thanks for that, I had no idea. Do you know if zero is truthy in erlang? It makes sense for elixir (based on ruby) but I'm wondering if erlang is that way too.
Re:Read this post ‘unless’ you’re not a Ruby developer on Tuesday December 13, 2022 @09:58AM (#33969669)
In what other languages does 0 evaluate to true?
Re:Read this post ‘unless’ you’re not a Ruby developer on Tuesday December 13, 2022 @09:51AM (#33969578)
Python is a communist apartment block, brutal in design and conformity and not very efficient.
Re:Read this post ‘unless’ you’re not a Ruby developer on Tuesday December 13, 2022 @09:49AM (#33969553)
I like C++ because I find angle brackets to be harmonious to the eye.
Re:Read this post ‘unless’ you’re not a Ruby developer on Tuesday December 13, 2022 @09:32AM (#33969353)
Unless seems to be pretty common in many languages. https://www.indifferentlanguages.com/words/unless In my mind the javascript community would do well to address the idiosyncrasies in its language standard before criticizing other languages for using common human language patterns. If you want to pick on ruby for being weird with conditionals, consider the following.

   if 0 
     puts 'true'
   end
which will print true. I think there is a much greater chance that 0 being true will cause problems with programmers from other languages than using unless which is quite natural to humans. Ruby is first and foremost a language for humanity, and probably dead last a language for ease of implementation.

> As a general rule I think that if a language has some feature for which there is already a commonly understood syntax across other languages, it should just use that syntax. If you’re introducing a complete paradigm shift, then that’s fine, but unless is not that: it’s just a different way to write if ! and people jumping back and forth between ruby and, say, javascript, now have one extra idiosyncracy to keep in mind.

Yeah, here's how to do it.

1. Set a monthly ad budget for your startup.

2. Run ads for your start up.

3. Change the credit card code so it puts a hold on their credit card and removes it. (You want to make sure your customers can actually afford your product)

4. Once the user signs up direct them to a page saying due to overwhelming demand you can't provide the service yet and let them know that their credit card has not been charged.

5. Once your startup is earning more than its ad budget, hire a developer to make it.

6. If you don't like this part of the business, absolutely do not start a startup.

Re:Decrypting SSL at scale with eBPF, libbpf and K8s on Thursday August 11, 2022 @01:29PM (#32429086)
I think it's related to perfect forward secrecy. We used curl's SSLKEYLOGFILE environment variable to find out what tshark needed and then worked to reproduce what curl was producing.
Re:Decrypting SSL at scale with eBPF, libbpf and K8s on Thursday August 11, 2022 @12:19PM (#32427999)
The application keys definitely do. The handshake keys are modified and at some point become useless to tshark. Not sure if it is a defensive coding thing, or what, but yeah if we don't catch it at the exact right moment they don't work.
Re:Decrypting SSL at scale with eBPF, libbpf and K8s on Thursday August 11, 2022 @09:49AM (#32425513)
Author here.

This is in process, we are using eBPF to read the keys out of memory in process and then write them into a pcapng file that tshark can use.

The handshake is only partially encrypted, the public keys and client random can be traced from the network, but the private keys need to be read from memory.

Some of the keys necessary to decrypt the stream are only found in memory for a limited time after which they are overwritten, if the keys are not retrieved in this time window the connection cannot be decrypted by tshark.