I can agree that stealing ideas is pretty universally considered bad. I can't agree that neglecting to fully disclose all the diverse inputs and sources of ideas that go into a mathematical paper will necessarily be interpreted by the community as evidence of theft. It has not been my experience as a member of the mathematics community.

Especially when the paper is proof of a counterexample, as in TFA. You ignored my example of Fermat's Last Theorem, maybe because it was fictional? Here's a related paper from an actual mathematics journal disproving Euler's conjecture for k=5. https://www.ams.org/journals/b... It looks to be about 40 words long.

It does say "a direct search yielded" (easy today but I wonder how much compute time that took, in 1966.) Is your assertion that if they had left those words out, we should treat their paper with suspicion?

This seems a bit accusatory and needlessly combative so I'm going to tune you out.

Your comment was about soundness not credit.

But yes TFA is about credit, so did you mean to say

I think I'd have to disagree with this too, in mathematics. For scientific publications that rely on empirical data collection and experimentation I can see how citing the sources of data and the established experimental methods you relied on would be important.

Math != empirical scientific research.

Fermat's last theorem was open for a long long time. Someone eventually proved it, but what if it had turned out to be false? What if a grade school kid had run a bunch of random numbers on her computer and by pure luck found an a,b,c so that a^3+b^3=c^3? She wouldn't have to cite anything. She could just publish the three numbers in a journal (they'd be falling over each other to publish it) and be famous. Trivial to verify. Mathematics would have been advanced by this event, even though the counterexample itself and the method for discovering it wouldn't have been deep or insightful in any way.

OpenAI is asserting that an important, long-open conjecture is false. The way they found their counterexample may or may not be interesting to mathematicians. Even so, it may be very interesting to AI researchers. LLMs, for all their limitations, may wind up being very useful at finding counterexamples to open conjectures, through a sort of semi-optimized brute force. We shall see.

I don't know about scientific publication, but there is no need whatsoever in mathematics to cite all inputs to be regarded as sound. It just needs to be verifiable from known facts. If what you've discovered is a counterexample, as is the case here, then it is often *much* simpler to verify that it is indeed a counterexample than it is to talk about how you discovered it. Disclosure of the "how" is interesting but not required. It could be outright plagiarism and still be mathematically "sound."

That line's been blurred already.

Why? "lose-lose" means your side is losing too. Many of us would want our negotiators to establish "wins" for us, and would not be specifically opposed to the "opposite side" also gaining something potentially. Often that's the best win in a transaction.

As for referring to Canada as the "opposite side," that's a very transactional view of things and I guess it's the position you'd have to take if you're going to shill for Trump. The Trump administration is not renowned for being strategic in their negotiations or their communications. But strategically the fact is that Canada is a ally, a friendly border, and an economic partner, not to mention the social, political, and cultural connection. Shitting on them and creating "lose-lose" situations is hardly laudable. It's not good business or politics.

It's not practical for us to try to solution here but a contingency backup of 50 TB of digital assets might be as simple as making a second copy and dumping it onto media in a box somewhere. It's an archive not a live system. Or whatever. Point is, it's not that hard to have *something* beyond just blind faith that a vendor will keep their end of the bargain.

Other non-technical contingency plans would include making sure the contracts have provisions for this common-enough scenario so you don't have to litigate in court to get your stuff back from Iron Mountain.

It's impossible to plan for every obscure scenario but this is not an obscure scenario. Business partners go out of business all the time.

Just this spring. Anything for my Professor T.

If you're suggesting that the mismanagement of this contract and related contingency is just due to lack of resources, well, maybe that contributed, we don't know. If so, it would be fair to take that into account when deciding whether to fire the managers. It doesn't change my point though, which is that this type of planning is standard management activity, and you generally can blame managers when they have zero contingency plans on the loss of such an important asset.

Actually we can and we do. You need to plan for contingencies. If you've got an asset that's vital for your business and want to store it with someone else, you need to figure out up front what you're going to do if they disappear, or damage the asset, etc. If the Nine PBS management didn't do this then they should be canned.

Similarly, if you put your money in a bank, you need to know what will happen if that bank fails. In many places, there is government-backed insurance for exactly that. If that didn't exist, we'd all need to manage our money differently. (And likely in a way that is not generally advantageous to the economy. That's why the government provides the insurance.)

The point isn't victim blaming. Rather, I strongly recommend that you take a lesson from this incident if you're ever going to be in charge of anything important.

What's wrong with the original definition?

https://en.wikipedia.org/wiki/...

Yeah, that's why they did it. But it remains true that the noise cripples the airtag as an anti-theft device.

Also the beeping collars upset my dogs.

Under $20. My alert threshold has been set to $16 for years.

I woke up this morning to an alert that my tiny static web site had incurred $3 million in S3 storage charges. Is that what "going viral" feels like? No thanks.

It was a good reminder to review what protections I have from liability should some script kiddie ddos me.

Interesting because I googled it and multiple sources say a few hundred a month is typical. Sources vary but I don't see anything remotely close to $1800/month. Anecdotally, I live in a similar sized house as you and also in the great lakes, a 120-year old house with spotty insulation and we broke a window frame this winter which I have yet to fully repair. Still it didn't cost us $1800 for the whole winter let alone a month.

Where did you get this number?

Zuck should rename his company from "Meta" to "Slop." The ticker is available.