Respectfully, I don't think you understand the concept of "no reason." Just because there are other ways to do a thing doesn't mean there's "no reason" to do it a particular way.

Some things need to work when most everything on the network is broken. Think: out of band access to the DNS server (DRAC, ILO, IPMI).

No, you know that some machine out there responded to that IP address. You don't know whether it's the one you meant or, say, the hotel's captive portal.

About SpaceX in particular? I'm not. About the sheer number of companies that behave similarly? Yeah, I'm bitter.

Actually, I declined the interview. This was in the pandemic before the vaccine. During the phone screen the recruiter told me all work was required to be on site and asked if I was okay with that. I said: sure, but only if I have an office so I can set up an air filter and generally control my working environment. The recruiter said no one gets an office, not even Musk. I said thank you and goodbye.

No partitions. No walls. No doors. Just desks and chairs.

The complaint wasn't about being in the same physical location. It was about the compulsory open-office configuration, even back in the middle of the pandemic.

Correct. Even in the middle of the pandemic he demanded on-site work and would not allow private offices in the building.

NASA hired women as scientists and engineers when that wasn't a thing. If her talents were worth it, that was that.

Musk won't hire people unwilling to work in an open office. And forget about telework. It doesn't matter what skills you bring to the table, Musk having his way is more important.

That's how NASA landed people on the moon while SpaceX's rocket keeps blowing up.

Not just identical twins. If you haven't found pictures of your doppelgangers online, it's only because they're not famous enough and haven't committed any crimes. With 8 billion people on the planet, statistics demand that a lot of them look enough alike that a casual observer wouldn't be able to tell the difference.

Except this wouldn't actually solve this. You'd be able to share the business logic, which would be a benefit. But you wouldn't be able to share any of the UI, system, or OS interaction code which is where all the incompatibilities come. If you just wanted to share business logic, there were already ways you could do that (write it in C would be one way).

Also, if they really wanted to do that, they should consider going the opposite way and bringing Kotlin to Swift. Kotlin also has a significant server side use that's growing (mainly by replacing Java), Swift is iOS and Mac only. They'll find a lot more people willing to learn Kotlin than Swift. Of course Apple won't consider that due to NIH and control issues.

Oh. You don't know how a search engine works. No wonder you don't know how copyright and computer software interacts.

You know what a search engine is, right? It takes all the words in a document and stores them in a database along with a link saying that this word was found in that document. The search engine has stored every word in the document, but it has done it in a way that it's not possible for the search engine to reproduce the document. The legal precedent is crystal clear that this activity does not violate the document's copyright.

Now you have a baseline for storing every word of a copyrighted book without violating its copyright.

When the LLM "trains" on a copyrighted book, how does it store the data? Has it saved the original data, in order, where it can spit it back out on command? Or like the search engine, has it stored relationships learned from the data which allow it to reason about the work but not reproduce it verbatim?

That's the correct question to ask when determining whether an LLM violates the copyrights of its training data. The plaintiffs failed to offer a credible answer to that question.

TOSLink's attenuation budget is something like 4dB. That's why even with glass at the proper diameter you can't get a TOSLink cable much longer than about 10 meters.

When the judge said the plaintiffs failed to litigate effectively, what he meant was this:

The defendant said that the LLM is not capable of reproducing the training materials. Instead, the information derived from them is able to summarize relationships, identify contained information, maybe even mimic the style of the book with new writings. In other words, it can do the things a normal human being can do after reading a book.

To prevail, the plaintiffs would have had to offer evidence that it was likely that the LLM stored sufficient data about the books to reproduce them exactly. If it can exactly reproduce the original works then it's not transformative, it's derivative.

The plaintiffs failed to offer any such evidence.

It's not just the high attenuation, you also suffer a high bit error rate as the optical signal interferes with itself. And that's when you move from 65 microns down to 9 microns. TOSLINK would be 1000 microns down to 9, two orders of magnitude more change in diameter and the square of that in area.