In the end, it's largely put a stop to H-1B sponsorship of workers outside the U.S. That doesn't mean that all these workers can't get visas to work in the U.S. but other - tougher visas - have to be explored.
This kind of cynicism is wild to me. Of course most AI products (and products in general) are for end users. Especially for a company like Google--they need to do everything they can to win the AI wars, and that means winning adoption for their AI models.
This is different. AI is an existential threat to Google. I've almost stopped using Google entirely since ChatGPT came out. Why search for a list of webpages which might have the answer to your question and then manually read them one at a time when I can instead just ask an AI to tell me the answer?
If Google doesn't adapt, they could easily be dead in a decade.
That's funny. I stopped using ChatGPT completely and use Gemini to search, because it actually integrates with Google nicely as opposed to ChatGPT which for some reason messes up sometimes (likely due to being blocked by websites while no one dares block Google's crawler lest they be wiped off the face of the internet), and for coding, it's Claude (and maybe now Gemini for that as well). I see no need to use any other LLMs these days. Sometimes I test out the open source ones like DeepSeek or Kimi but those are just as a curiosity.
If web-pages don't contain the answer, the AI likely won't either. But the AI will confidently tell me "the answer" anyway. I've had atrocious issues with wrong or straight up invented information that I must search up every single claim it makes on a website.
My primary workflow is asking AI questions vaguely to see if it successfully explains information I already know or starts to guess. My average context length of a chat is around 3 messages, since I create new chats with a rephrased version of the question to avoid the context poison. Asking three separate instances the same question in slightly different way regularly gives me 2 different answers.
This is still faster than my old approach of finding a dry ground source like a standards document, book, reference, or datasheet, and chewing through it for everything. Now I can sift through 50 secondary sources for the same information much faster because the AI gives me hunches and keywords to google. But I will not take a single claim for an AI seriously without a link to something that says the same thing.
Given how embracing AI is an imperative in tech companies, "a link to something" is likely to be a product of LLM-assisted writing itself. Entire concept of checking through the internet becomes more and more recursive with every passing moment.
> When they hit a jack pot. They’d document the mutations, throw the engineered strain out and start blasting them with UV. Afterwards you just scan for the same mutations and voila, now it’s classical strain enhancement!
Instead of starting with a fresh gene pool and blasting it with UV and praying that they get the same jackpot mutations, why didn't they start with an entire population with that desirable jackpot mutation and those blast cells with UV and then select for the ones that survived?
They are both about signaling wealth and status. What I don’t understand about the digital items is that the people who own them are often anonymous so why signal? Signaling wealth and status IRL can also carry other benefits that don’t seem to carry over digitally.
Actually with CS stuff there's scarcity too, but since it's digital it's easy to change. Same with designer bags (nothing stopping them from churning out more, but they choose not to do that). Not so with vintage items though, since they're no longer made.
But a lot of Stuff is made with future vintage in mind, e.g. every Ferrari or other high end sports car will only appreciate in value.
I know people that spend gazillions on vintage sneakers. They will literally go and buy some rare designer second hand pair of Nikes or whatever with some scarce design that they only produced a few off. Personally, I wouldn't be that eager to stick my feet into somebody's well worn sneakers. But apparently that's beside the point. Nike actually on purpose feeds that market by coming up with new limited edition designs. These people have enough shoes. They don't buy them because they need another pair of shoes.
The value of money used to be based on gold. Gold has very limited practical value. It actually kind of sucks as a metal because it's not that hard compared to e.g. iron. The main value proposition is that it's pretty and shiny. But people that buy gold don't tend to even look at it. They just store it in a vault. Or worse, they get a digital receipt that proves they own the gold without ever seeing or handling it. The main value of that is that, if you wanted, you could make pretty and shiny things out of the gold bars. And because those pretty and shiny things are valuable, gold is valuable. And therefore people invest in gold. Not to make those things but to be able to sell it to others that might do those things. Of course the vast majority of people buying and selling gold has zero interest in doing that. Most gold ever mined is locked in a vault in bar form and will never be used for anything else than as an intrinsic token of value.
There are a lot of things that have no value beyond subjective esthetics and the group thinking around that. My home country the Netherlands produced a lot of fancy paintings in the seventeenth century. Those are worth a lot now. They are extremely nice according to some. People visit museums to go see them. They are worth tens/hundreds of millions in some cases.
Objectively, most people that visit museums wouldn't be able to tell apart the original from a good replica. And reproducing these things with high fidelity digitally isn't all that hard either. You can find high quality scans of almost any painting for free on the internet. And you would get most of the appreciation/emotion looking at those as you would get by looking at the originals. Of course, most people aren't that into this stuff in any case. But we appreciate these things because other people tell us they are valuable and we take their word for it. The original paintings keep their value mainly because such people keep reassuring us how rare and amazing these things are. That tends to get embarrassing/awkward with forgeries in museums where experts literally have failed to tell the difference.
The value of things whether digital or real is based on social mechanisms for appreciating things. Some things simply are valuable because people agree for whatever irrational reasons that they have value. And then some people buy these things at the market rate because they enjoy having them. Whether that's original art on the wall, some rare sneakers, or a cool skin for a game character that you engage with for many hours while playing the game. The dynamic between the willingness of people to separate with their cash and scarcity is what creates the value.
NFTs are weird mainly because they are digital receipts for something (anything) that has value. They are no different than a paper certificate of authenticity for a painting. It all boils down to the trust people have in the impressive looking stamps/signatures on the paper, or the blockchain shenanigans used to ensure authenticity for the NFT. Of course a lot of NFTs are silly. But in game worlds, ownership of skin is kind of limited as you can't really resell them easily or prove authenticity. Which is something that NFTs addresses. Which is why NFTs became popular in games.
The value of game skins is as irrational as second hand sneakers are or the appreciation for shiny metals. Or gems. Or paintings. But as long as people buy those, they have value.
this is not a gender specific issue. You can easily look at the watch-market, luxury car market etc... to see the same issue play out. In cars it's even worse when you have something like the Lamborghini Urus compared to the Audi RSQ8 which are very close to being the actual same car.
People generally don’t take the time to learn the framework and miss out on the tooling it provides. select_related for example. If I had a dollar for every project I’ve been hired to work on that didn’t use it, well, I actually do.
Also, people in general don’t seem to be able to do more than very basic SQL, so creating views is seen as a little known performance “trick”.
FYI you can avoid the manual sprinkling of select_related somewhat with https://pypi.org/project/django-auto-prefetch/, which avoids the lowest hanging N+1 loops automatically. Definitely not a cure all but it helps.
Ah yes. That’s the one thing I need to teach everyone when they’re new to Django. I was wondering if there were other quirks to the ORM beyond avoiding N+1 queries.
I assume they are referring to the default behavior of the ORM fetching all fields for a model by default, and the frequent need to use select_related/prefetch_related to group your queries into larger ones that are (usually) much faster than making many small queries for related tables.
The dark forest is such an obviously false theory to me. Its axioms are:
1. Survival is the primary goal of all civilizations.
Agree.
2. Resources in the universe are finite.
True in the theoretical sense, but false in the practical sense.
3. Civilizations cannot be certain of others’ intentions.
Not obviously true or false.
4. Communication is dangerous.
This is such a strong axiom and is almost certainly false.
Its conclusion from applying the four axioms is that preemptive annihilation is the rational strategy.
As an alien civilization, if your strategy for survival in the cosmos is to "immediately and totally annihilate any sign of life", then that is almost a surely losing strategy. If intelligent life is prevalent, and the cost of annihilating a species is so low that they can just do it willy-nilly, then all it takes is one surviving colony to use the same superweapon against you and you're finished. Oh, you'd also have to be annihilating species left and right across the galaxy without revealing your location. And in the worst case, you've just pissed off all the known alien entities in your galactic neighborhood. Good luck to you.
It makes for fun writing, but I don't understand how anyone can take it seriously.
In simplest scenarios software is not aware of the IP space. Like you bind to 0.0.0.0:443 and move on.
In more sophisticated configs adding / removing IP's or TLS certs requires restarting server, configuring applications. This gets out of hand quickly. Like what if your server has primary IP removed, because the IP space is recycled.
At CF all these things were just a row in database, and systems were created to project it down to http server config, network card setting, BGP configurations, etc. All this is fully automated.
So an action like "adding an IP block" is super simple. This is unique. AFAIK everyone else in the industry, back in 2012, was treating IP's and TLS more like hardware. Like a disk. You install it once and it stays there for the lifetime of the server.
How quickly would the local population of mosquitos adapt to the MosquitoDunks, such that within a few generations, the only surviving ones would be the ones that are unaffected by the dunks? Or is that not a real concern?
I don’t know much biology, but is there a general principle where things that have a shorter reproductive cycle tend to “win” these sorts of arms races? I wonder if we will have to occasionally go out and find better copies of the bacteria, haha.
The arms race might not even be realistically winnable for the mosquito. Most of the time when we talk about resistance it’s to a single molecule or evolutionary adaptation. That mosquitos have not developed a resistance after decades of heavy use (we’ve been using Bt as a pesticide since the 1920s and its the most used pesticide today) implies that the Bacillus thuringiensis used in the dunks/bits is attacking the larvae with multiple different mechanisms of action.
When that happens, natural selection almost always fails to develop a resistance because a resistant organism needs to have a defense against every mechanism of action at the same time. The probability of two beneficial mutations lining up perfectly to create that resistant organism are just extremely unlikely. That’s the basis for modern oncology, which uses combination therapy to prevent tumors from growing resistant to any one drug.
Edit: yep I was right. Against mosquito larvae, Bt produces multiple Cry toxins that bind to specific protein receptors in the microvilli of midgut epithelial cells and Cyt toxins that directly interact with membrane lipids [1]. It’s a triple whammy because there’s six major toxins (Cry4Aa, Cry4Ba, Cry11Aa, Cyt1Aa, Cry10Aa, and Cyt2Ba), Cyt1Aa creates more binding sites for the Cry toxins to attack the cell, and Cyt1Aa itself has two mechanisms of action, one of which disassembles the cell membrane, leading to cell death. That is an incredibly difficult evolutionary hole for an organism to dig itself out of.
Starting at an initial density of air, suppose you descend a distance D such that the air density doubles. Now your air is twice as dense, which doubles the pressure underneath it, meaning if you descend a further D the density will double again. Continue ad infinitum (or at least until the ideal gas law stops being a good approximation).
> Pressure (P), mass (m), and acceleration due to gravity (g) are related by P = F/A = (m*g)/A, where A is the surface area. Atmospheric pressure is thus proportional to the weight per unit area of the atmospheric mass above that location.
What impacts are you seeing as a result of the $100K H-1B fee which took effect on 9/21/25?