We may wake up one day and Riemann hypothesis or any other unproven hypothesis would be proven by AI or with AI help. Probably it's just a matter of time (GPT10?). Theoretical advances may also count.
@zeddotdev@maroffo i had such situation year or so ago, and there is long-term github issue with people complaining about such memory consumption, https://t.co/MFTscmy5I2
In my case, removal of autocomplete suggestions and run time code analysis helped (+ migration to 64G RAM from 16)
We’ve identified a security incident that involved unauthorized access to certain internal Vercel systems, impacting a limited subset of customers. Please see our security bulletin:
https://t.co/0S939n3qHC
first time I tried electric a while ago, it just crashed on "tutorial-like flow" and they accepted the bug on their github. So I didn't try it since then (while it seems resolved, not sure https://t.co/ss49FZHwSk). Since when I try something new and it crashes on simplest cases, it is not production-ready. But in general, you need to be very careful in multi-tenant apps about sync engines to guarantee RBAC, for example as well as in B2B corporate networks websocket may not be open by corporate policies for some customers. As well as stream design should be done carefully to avoid unnecessary traffic volume to the client-side.
it's not correct. in "da net" "da" does not have a meaning of "yes", it's translated as "but" "well" or "indeed", because "da" has several meanings - "yes" and "but"/"well"/"oh" and as well as even "and" (Ivan da Maria = Ivan and Maria). So "da net" in English has a closer translation as "well no"
@Scobleizer@openclaws recently (without openclaw, but based on claude) built such a tool for a niche market monitoring (websites, blogs and github-based), for myself, really works well. X as a new affordable datasource really opens up a new stream of data
besides login with these systems, Anthropic and OpenAI could offer a transparent revenue-sharing model for products that are built on top of their APIs. So developers of the API-based AI-powered apps could build a monetisation model that is tied to PAYG token-based consumption for every end user, so that business model could be transactional per each API call, focusing on features for exact use case and not thinking about the average consumption of tokens per user to stay profitable
I did it 26 years ago and it was quite tough to pass exams for high-ranked STEM university. These days, educational system has a country-wide "graduation test" in school and every university should accept the result. Not sure about the complexity of this test, but probably it's now more standard than those old time "specially curated exams"
@MatthewBerman environment is a pretty wide term. in every language it may mean different things and ways of installation. Do you mean environment including dependencies/packages or only code version?