This year, LLMs have been involved in a number of interesting proofs of conjectures. But that is not even half of mathematics. Has anyone tried to use an LLM to generate a mathematically interesting conjecture, on the level of Conway's refinement conjecture? If so, what happened?
With all the talk of mathematicians possibly being obsolete, I'm wondering where the future conjectures that future LLMs would prove might come from.
I would not be surprised if an LLM could have generated this conjecture, since it's such a natural extension of the integers, does it hold for surreal integers is not a big leap. It's totally predictable / probabilistic and thats what they do at a high level.
My read on the current sitch is that the drama is more around the humans behind the LLMs, like the OpenAi *people*, who are dishonest, thieving, technocrats, stealing human mathematicians work and claiming credit for it. Since the LLMs are only trained on finite amounts of data on the internet I don't think anyone expects them to replace mathematicians in a serious way. But surely the field is changing dramatically, people will adapt.
I'm guessing you located are in the bay area? If so, that is one thing under your control that you could change. The early SV VCs didn't build the industry by staying in NY (where the capital concentration at that time was). They had to get away from that scene to do what they did. It might be easier to cut your own lane if you weren't sharing the same turf.
It is a series of science fiction novels in which a future China conquers the world and re-writes history as if the Tang dynasty never ended. The rulers present themselves as being the legitimate inheritors of a continuous Tang dynasty and all of western history from 900 CE onward is secret banned information.
Sort of the exact opposite of this conspiracy theory.
I'm wondering if anyone in charge at Netflix is going to wise up and realize that their biggest problem is their UI.
Years ago, if you asked a programmer why the Netflix UI worked the way it did, the answer was that it was designed to hide how small their catalog actually was. Then, Netflix became quite successful and their UI became an industry standard, with Amazon, Youtube, etc. all copying it or aspects of it. So why mess with success?
But now Netflix is in a situation where they have a dearth of new good content, but now they also have a fairly substantial catalog which has been built up over more than a decade. And they are still sticking with a UI designed to hide the size of their catalog. I often talk with friends and family who have Netflix and they are not aware of shows which they might like, simply because the Netflix algorithm hasn't shown it to them.
I suspect that making one simple UI change could have as much effect on their fortunes as adding $hundreds of millions of new content. Simply only display the icon for a given show once in any given session. In other words, if say "Rebel Moon" is shown under Sci-Fi movies, only show it there, not also under Action movies or some other row. They've already seen it once and clicked past, so showing it again is wasting a space for some other show they might like.
People are leaving Netflix because they click around for too long trying to find something they want to watch. The problem is perceived as lack of good new content, but the real problem might just be the clicking around.
They've also made it harder to go to content in that catalog. You used to be able to get to the sidebar relatively easily, even if you'd swiped right a dozen times, by just going up|down, then left. then you were at the menu and about 2 clicks away from search.
Now? Now the search bar is at the top. If you've scrolled down through categories while looking for something, then decided to just use search, or to just look in movies or shows, you now gotta scroll up & up & up.
Everyone likes to gatekeep what urban is, to the point that some only consider New York City or even just Long Island to be an urban area. The areas in the linked graph are all recognized as urban areas by the US Census Bureau, and whether you're in Odessa Texas, Provo Utah, Vancouver Washington, or Pasadena California, you're in a similar population density with similar services, except maybe that going to Salt Lake or Savannah airports only takes 45 minutes, instead of the one to two hours it takes to get to LAX from Pasadena.
I don't disbelieve the cultural influence myself but I'm pretty sure that urbanization matters.
I think the master narrative of human civilization is that almost all of us had ancestors working in subsistence agriculture a few generation ago who have moved or are moving into urban areas. This transition is largely complete in the US and Europe, happened explosively in China post-1980 and is about 2/3 complete there and is less far along in Africa and South America. Birth rates go down in this process.
It would be helpful if the event result message remained on-screen until clicked. At least on Firefox, the event result message disappears faster than it can be read (example, what happens after selecting an option for the inspector).
The Great Pulping is the name, or rather names, that modern historians have given to two legendary events which were supposed to have occured early in the Confusion Era. The first of these events was when libraries converted from a primarly paper-based form of information storage to a digital form. A majority of library book collections were scanned, digitized, and then once the information was stored in digital forms, the books were sold to be pulped in order to free up room for library patrons to use the new system more effectively. Of course, the librarians could not know what was going to happen in the future to all this information, what with the viruses, cross-copying, AI hallucinations, archival rivalries, meme wars, and all the other diverse information-distorting causes of the Confusion.
The second event refers to what happened to the librarians who had been responsible for this decision, when the full effects of it became evident.
- A legendarium of the Confusion Era, by Franklin Duane, 192 PCE Edition
You used the phrase "our industry". Personally, I'm not a huge fan of the 'tech industry' concept, simply because a lot of startups are not in software/computing, and a lot of new technology isn't either. But I get what people mean.
I notice that the companies you mention like Costco and Patagonia are not in the tech industry. Does your new book have any examples which show how to stay incorruptible in the face of the network effects that drive monopolization in the tech industry? Alternatively, have you seen workable ways to split network effects amongst networked affiliates, to spread out the market power?
I know that most founders aren't exactly looking to make a startup with a lot of competition (I'm sure not), but it would be nice to know if someone is fixing problems specific to the 'tech industry'.
Yeah, I did my best in the book to find that balance between tech and non-tech stories. You'll find a few tech stories in there, for example Anthropic, GitLab and Cloudflare. And if you want all the horror stories, you can read Cory Doctorow.
I wanted to have a wide variety of stories in the book from many different industries to show just how pervasive these forces are that we're doing battle against. While there are some things that are unique to the tech industry, I think we have more in common than what sets us apart.
FWIW, I wish people would talk about the "startup movement" rather than the "tech industry" for what we do. But maybe that ship has sailed.
reply