theanonymousone

My biggest scare in setting up a service where users can upload something, is the probability of "problematic" material. Is there any solution to this?

Of course. My point is, probably a super cheap LLM that does not cut you off after 1500th API request of the day is preferred over the free model that does so, at least for certain use cases.

> There is still Gemini free tier which is ofc basically impossible to beat Is it free free? The last time I checked there was a daily request limit, still generous but limiting for some use cases.…

Given the IPv6 case, I will handle this one with care.

Thanks. Is there some rule of thumb about how much should the article change so that the copyright doesn't apply any more? These points come to my mind: - There is a whole category of software ca…

Does anyone know about the legal implications of building such a "fronted"? Doesn't Reuters own the copyright zo the contents of their news articles? Can't they send a Cease & …

Is the list of the board members or VCs involved in this confidential?

Sorry but is there some source about this for a lame person?

I don't know much about accounting, but isn't shutting down an accounting service three days before the end of the year a huge slap in the face if the clients of the service, maybe even frau…

Does this solution also guarantee IDE support for code completion in the Java libraries in the JRuby code? Such things are a huge deal(breaker) in the Java ecosystem. But maybe not in Ruby?

Is it demonstrably better than Playwright in bypassing Cloudflare measures? I have some scraping projects and the "cat and mouse game" (what's the right expression here?) got so much en…

Why is the repository 95% "HTML" code?

I can't open it since the morning in Switzerland.

Given the comments saying it's performance seems comparable to 4o/4o-mini, is it safe to say that GPT-4 performance can be achieved with less than 100B parameters,in contrary to what previou…

I'm "tracking" the price of if 1M tokens in OpenRouter and it is decreasing every few refreshes. It's funny: https://openrouter.ai/meta-llama/llama-3.3-70b-ins…

Really? I plugged my old phone some days ago and set up a ssh daemon on it, just to see if it stays connected. Should I be worried?

Yes. I mentioned that Heroku also supports them.

I used these so called "build packs" a lot and owe Heroku a great deal in general. However, when I was "forced to" use Docker containers instead of custom build packs (Heroku also …

The "fact" that Java is faster than go caught me with (positive) surprise.

Yes, it worked. But still, I expected that my code should work if it worked last week and I didn't change anything since then :D

No version change, or obsoletion announcement. I only have a hobby project using it, but it should be much more painful for people using it in some production code, no? :-/

Maybe surprisingly, JBang has offered this functionality for Java since 2020. Happy we have it un Python too, now.

The post uses the expression "delve into" :-/