HN user
I would guess the price for payment processing drops tremendously when you don't need to handle fraud or chargebacks at all. It's for sure worse for consumers but I do think it has its place (for example adult content that has huge chargeback risk or straight up blocked by other processors).
Your experienced latency will be heavily influenced by which DC you hit, in which case we care about the relative traffic loads because you aren't just hitting your closest one all the time.
By traffic load east coast datacenters dwarfs these.
Most of big tech's major data centers are in Loudoun County, VA on the east coast not the west coast. It's centrally located to be great latency for the east coast and OK latency for west coast and Europe. Plus a friendly regulatory environment and lots of existing DCs (AWS us-east-1, Azure East US 1/2)
If you're feeling any better latency on the west coast it's more likely to be placebo than real.
If the agent you are running really wanted to it could easily find a way to mount the windows folders and read them all. WSL isn't a security boundary, you are only barely more protected than people running grok in their home directory.
Wealth is not in fact finite. If you take a piece of lumber and carve it into a dresser you've just generated wealth. Resources kind of are but not really, extraction takes it from unusable to usable and essentially creates it (also, trees literally grow out of the ground). Similarly Tesla has made a lot of cars that people drive and AI companies have created products that have hundreds of millions of DAU.
Whether that "justifies" the market prices is orthogonal, all the market price says is that people want to own a piece of the pie at the current price being offered. It can stay that high as long as there continue to be people interested in owning shares of the companies.
You are poisoning the context and priming it to say load-bearing. I don't really think there is a good way to actually ban specific phrases other than have it do a second pass, it's something baked into the model from post training.
Within the 1982-2026 span there is an equally negative 3.5 sigmas deviation somewhere (no year labels on the graph). The article doesn't touch on it at all so I have no context as to what it could be. But it definitely suggests 3.5 sigmas is not really 1 in 7000 years.
Your post seems pretty polite. Did you have Claude write it for you?
Most likely the building gets stabilized and then anyone involved gets embroiled in lawsuits and it stays standing half finished for years. One Seaport is a famous recent example of an under construction skyscraper getting halted for structural issues. https://en.wikipedia.org/wiki/161_Maiden_Lane
In Japan, there's a big issue when a snack raises its price 2 cents
No, there really isn't. You're looking at one company that "apologized" as a marketing play but outside of that prices have been increasing with no fanfare for years now. The annual inflation rate has been 2-3% for the past 4 years. It's a lot less interesting to write a news article about that though.
I would wager most if not all of the tokenmaxxing was done on enterprise API priced plans, not subscription plans. You can't actually token"max" if you are limited in the amount of tokens you can use per 5 hours.
That's because you pretty quickly catch downvotes from people with sour grapes whenever you talk good about working at big tech. Quadruply so for companies much more hated than average like Palantir/Meta/Oracle. It's just what echo chamber downvoting culture gives you it doesn't really have to do with what's happening internally. They're very very happy to spill beans and talk about the current environment on Blind which is a lot less hostile.
No not really your perf will only realistically influence 10 maybe 15 percent of your comp for your first 4 years. Vast majority is fixed at signing time. Numbers there are truly what you can expect to get.
You frequently having to tell people about a global configuration gitignore is an obvious consequence of "My general rule is that in-repo .gitignore should only be used for repo-specific things". It wastes less of everyone's time to just gitignore them in every project.
Do you really think the guy branding thousands of people working at Meta as basically pedophiles can really be said to care about "solidarity"? I certainly wouldn't consider someone a peer if they randomly go and call me a pedophile because of where I work. I'm sure 95% of people working there have 0 relation to the algorithm decisions and definitely have no particular fixation on giving teenage girls depression.
Many people do have unlimited budgets because their work pays for it. Just because the latest SOTA model isn't something most consumers can afford for personal use doesn't mean it's not worth releasing or discussing. I would guess that the vast majority of software that benefits from being on the bleeding edge is developed by people working at companies on API pricing.
No, that's not what it means at all even if just doing it purely in math terms. Really it is just a reasonable amount to cap at to stop the long tail of super spenders (tokenmaxxers). You could also call it "the amount of AI spend after which Uber has decided there is diminishing returns for the average engineer".
There are quite a few making that much yearly but not the average. Median swe at meta is almost surely >1MM net worth at least and maybe even 2.
Target market for stacked PRs are ICs who don't have much decision making power and let's be real do not care too much about the look and feel of a "launch site" for the feature. It's also something few if anyone is making a purchasing decision over.
Target market for copilot includes people with actual purchasing power and also many new users where this is an actual make or break feature. So this is worth the investment into design while stacked PRs is questionable. I actually question why they bothered with anything more than a blog post at all for stacked PRs (looking at the post it doesn't seem like too too much more than a blog post though).
Claude is so in demand at the moment that there aren't really volume discounts. Anthropic sets the terms and you either accept them or get lost they have that much of a lead (mindshare/desirability wise).
Bitflips specifically may not be; things like network issues, noisy neighbors, row/rack/host maintenance (leading to a downed and migrated host) absolutely are things that happen at high frequency at scale and cause your background level of errors to be more than 0.
This was always a little weird to be because Microsoft internally is actively hostile to cross-org collaboration. If you worked in most of Azure you basically have 0 lanes of communication with someone from the Windows team and vice versa. Triply so for stuff like Kusto or Teams which you'd be dogfooding daily. I guess if there's a horrible stop the world bug it'd get surfaced through telemetry but normal user feedback is not a thing.
Compared to working at other big techs, where I was able to direct msg the engineers on the team for internal protobuf or datalake services in addition to user groups that were generally responsive it was just strange. Also Microsoft doesn't have a monorepo so you can't just commit patches to their service because you don't have access to their repos which I pretty regularly do elsewhere.
Calling #2 more sustainable has no basis in reality, it's just a feeling. It's like saying that clothing before the loom or farming before the tractor were "more sustainable". No, it isn't, it just appeals to yeoman farmer instincts that somehow technology=bad when it's what powers (and sustains) our modern world of 8 billion people.
One of my favorite things to come out of Alice in Wonderland is the Red Queen's Race, which has been used as a metaphor in many many fields for the concept of needing to work as hard as you can just to keep up with others, not even to get ahead.
"Well, in our country," said Alice, still panting a little, "you'd generally get to somewhere else—if you ran very fast for a long time, as we've been doing."
"A slow sort of country!" said the Queen. "Now, here, you see, it takes all the running you can do, to keep in the same place. If you want to get somewhere else, you must run at least twice as fast as that!"
https://en.wikipedia.org/wiki/Red_Queen%27s_race
Also related, recently it's been used to describe involution (neijuan) in the chinese economy https://en.wikipedia.org/wiki/Neijuan
Software that is sold as a service and requires ongoing maintenance like running in the cloud (and people to keep it running in the cloud) is opex not capex. Google Search is most definitely opex.
The best way to use LLMs is via tmux where it's running on a disposable VM. 0 chance of it getting information from your local machine.
A bug-for-bug port to Rust is the first step to fixing that. Assuming the port is actually 1:1 without any behavioral changes, these bugs already exist in the Zig code. The difference is now it's known where effort can be dedicated in order to one day have a memory-safe release of Bun. People have absolutely lost their mind over this and completely forgotten the benefits Rust gives you. I feel like I've gone back 10 years reading threads about the Rust port of Bun these are the exact same arguments we see from people advocating continued use of C++.
It's interesting you call these sites and advertising opportunities absolutely necessary for things like visual arts businesses and yet don't think they are doing any good in the world. If the ad spots and targeting didn't exist then niche internet-ads-dependent businesses also would not.
I'm guessing you have 0 insight into the work that the ads, ranking, apps, and sales teams do to keep the gravy train flowing and even expanding 30% every year. If Meta fired even just half the ranking workers, the recommendation models (both ads and feeds) would very quickly become stale and start shedding many Fortune 500 companies worth of revenue.