My company's site has also been getting hammered by Brazilian IPs. They're focused on a single filterable table of fewer than 100 rows, querying it with various filter combinations every second of every minute of every day.
I also have this issue. What I do is work in a separate branch and cherry pick commits over to main then rebase.
I bet that what you describe could be implemented with stashes and scripts run on hooks, but that feels like fighting the tool too much. There's probably something we're missing?
What statements about properties of randomly generated code snippets would be useful for coding LLMs? You would need to generate text explaining what each snippet does, but that would require an existing coding LLM, so any IP concerns would persist.
>In order for an AI to evaluate the effect of a small molecule on the brain, it would have to... simulate the operation of a human brain in a simulated environment.
Humans aren't capable of doing this, but still make useful drug discoveries. AI can be empowered to conduct research in the real world, it doesn't need to simulate everything.
What's not mentioned in the comment you're replying to is the postgraduate-level type theory you need to learn and constantly refine to work with these languages at a practical level. So be aware that even if you put the time into learning them, it's unlikely you'll get to use them at work.
Look up Coq, Agda, Lean, and Idris. I would start with Coq, it's the most used. Idris is more like Haskell and programmer-oriented.
Edit: Nevermind, apparently they were just talking about Haskell...
Chronic post-vasectomy pain syndrome negatively impacts quality of life in 1-2% of cases. Even a small chance of chronic pain should be weighed carefully against the benefits and alternatives.
The person you responded to accounts for uncurved classes in their argument, that cheaters in uncurved classes "inject false signal into the difficulty feedback loop".
If the average assignment grades are very high the professor will be inclined to make the assignments more challenging. Cheating disrupts the professor's ability to accurately determine how well their students are grasping the material and adjust accordingly.
> Lean to me is not only obscure but unapproachable by someone from a non-mathematical background. ... If we want to democratise formal methods and theorem proving ...
For whom is that a goal? Are there really that many individuals at the intersection of "I need to use a theorem prover" and "I don't have formal mathematical training"?
Well some calls absolutely should invoke network or disk activity, so you would additionally need to define what constitutes good and bad activity for each. Moreover unless the package is a collection of pure functions it would be easy to hide the malware trigger in state that won't be initialized properly by the automated method calls but would be in the standard usage of the package.
A lot of government agencies maintain social media presences.
The White House has a Twitter account, for example. So there are presumably federal employees with Twitter installed on their work phones for the purpose of operating that account.