tomatbebo

Your comment is definitive but your reference is to a Wikipedia section labelled as "speculation".

What counts as a project in your case? Would you oppose the idea of modules within a project mixing languages? I believe curl is over 500,000 lines of code so in the case of forks a progressive rewrite, module by module seems a lot more achievable than porting everything all at once

I strongly disagree. I don't know the rights around one's own voice, but the idea that you suddenly lose ownership of something because you shared it online is the exact thing that many people take issue with when it is written in the terms of service for social networks, creator tools (adobe), etc.

Potential abuse of such powers should not be tolerated regardless of whether one is part of a protected group or not.

Cool, now I can name my internal packages keygen to avoid dependency confusion /s

I assume the bandwidth was too costly. It seems to be archived on Wayback Machine https://web.archive.org/web/20230509042606/https://worthdoin...

Did you fill in the form with super compelling use case or something?

They updated it but as soon as you click it the old icon and UX is there?

That would be nice but at least they published their research

Reading this makes me wonder if there is room for something similar to "The Global Chubby Planned Outage". In addition to good comms, they could gradually degrade your service up until suspension as a way of getting your attention.

Embeddings are often used as features for these LLMs so before they were paying to generate embeddings and doing inference with these large models. Now they pay to generate embeddings, fine-tune them and do semantic search (probably approximate k-nearest neighbors). The hardware requirements for most LLMs make them much more expensive than approximate KNN with a vector database.

It looks like the feature was released in September last year but has not been well publicized.

"any and all testing?". Are you aware of the flat earth movement? For some of their members, no amount of evidence is enough. The goal posts always move. I suspect the same applies here.

Imagine JPMC paying off your debts as part of merger when those debts include the cost of generating fake customers

I'm inclined to agree with the premise of this article but I'm skeptical of the numbers.

Not saying one is better than the other but there were other costs too. Like the people displaced from their homes, the environmental clean up costs, etc.

Pressumably things will start to snowball if these smaller scale trials demonstrate adaquate safety

Playing devil's advocate in Twitter's defense:

Is there any public information about how twitter handled the two examples you provided at the end there? Do we know for sure they actively chose not to take any actions?

If not, I'm not sure it is fair to claim they completely ignored them. They may have limited their reach via some kind of shadow banning. Action like that is harder to notice than an account ban.

Also, since your examples predate the particular ban we are discussing, do we know for sure there weren't policy changes in between. Perhaps even because of examples like the ones you list.

Personally, I wonder if Twitter might have a more American-centric view of the world and a higher sense of urgency for "events at home", so to speak. That could go some way to explaining what seems to be selective enforcement of their terms of service.

A less sensational take is also provided by https://www.bmj.com/content/378/bmj.o2303

So unnecessary too. Having experience in the domain, like teaching react, lends more creditability than "trust me, I am old".

With the upcoming Apple event, I wonder if this is another case of crushing the competition before Apple releases its own app?