incorrecthorse

Plot twist: It's not actually scalable because no amount of tools and buzzwords can compensate for the lack of experience in proper architecture for scaling.

For the uninitiated, Linear is a project management tool that feels impossibly fast. Click an issue, it opens instantly. Update a status and watch in a second browser, it updates almost as fast as the source. No loading states, no page refreshes - just instant, interactions.

How garbage the web has become for a low-latency click action being qualified as "impossibly fast". This is ridiculous.

And compute-intensive hash functions. Computers this day are powerful enough to hashcat each individual pwd+salt if a fast hashing function is used.

I think the common confusion boils down to:

The moment of capturing a measurement is known as a metric event

Which suspiciously reads like a log.

In practice, a metric is an aggregate of events (the "metric events") when you're not interested in the individual event but, but in the aggregate itself. For practical reasons this is not implemented with logs but with more primitive technical events emission.

This is not fundamentally incompatible notions. If you do an electrocardiogram, you might be interested in your BPM, but it is deduced by the full log of each beat. The segregation we do in computing is more practical than fundamental.

The bad ID naming `d1` is convenient for the argument...

It blows my mind OpenAI wouldn't be able to build a Windsurf alternative for orders of magnitude less than $3B.

we form thoughts at 1,000-3,000 words per minute

I would like to know what this measures exactly.

The reason I often prefer writing to talking is because writing lets me the time to pause and think. In those cases the bottleneck is very clearly my thought process (which, at least consciously, doesn't appear to me as "words").

It _is_ equivalent to a back door, that's the point. The UK demand can be accessed more rapidly and properly by disabling the feature than by implementing a backdoor, since it is the same thing.

So, what exactly has generative AI actually done? Where are the products?

The product is ChatGPT, actually.

If LLMs are a bubble, then you should expect most of OpenAI's revenue to come from its API (which is used by startups which have raised money to do "magic AI stuff", and the bubble would pop when investors would stop giving the money). But according to https://futuresearch.ai/openai-revenue-report, revenue from the API accounts only for 15%, the other 85% being the different subscriptions offers, including 55% of ChatGTP Plus subscriptions -- that is, _direct consumers_.

This doesn't prove that it isn't a bubble (the consumers could realize it's useless and then leave some time later), but it makes it less likely IMO.

Do they have an incentive to learn? If the reward process in the company also consider it "out of scope" (meaning the remaining work is scheduled as such), there is nothing to really learn.

*Cinephiles took the DVD boom era for granted

The reason it ended is because the general public doesn't care.

Those people seem to have a passion for developing package managers (instead of just seeing it as a tool that needs to do the job), and as long as it is the case, I don't see how we wouldn't end up with one new package manager every year.

Aren't you screwed from the moment you have a malicious user in your workspace? This user can change their picture/name and directly ask for the API key, or send some phishing link or get loose on whatever social engineering is fundamentally possible in any instant message system.

It can take an arbitrary amount of time. Modules are just code executed top to bottom, and might contain anything beyond mere constants and functions declaration.

Code coverage proves that the code runs, not that it does what it should do.

Unless you want an empty test suite or a test suite full of `assert True`, the reward function is more complicated than you think.

The update affected less than 1% of all Windows machines. [1] Although maybe the biggest software failure in history, far from the biggest possible one. The level of cloud connectivity in the world could basically break the world if we didn't have diversity.

[1] https://blogs.microsoft.com/blog/2024/07/20/helping-our-cust...

Those two flight deck pilots had breathed-up all the oxygen in their breathing packs by the time they hit the sea, something confirmed by the empty breathing packs that were recovered. Which means they were alive when they hit the sea!

I don't understand how this follows. The best scenario is that they had their last drops of oxygen around hitting the sea; in other scenarios they died from lack of oxygen before hitting the sea.

Board state itself becomes unique pretty quickly, so you would just end up with a gigantic lot of "moves" played only 1 time.

EDIT: so you could define "rare" moves as the biggest difference of occurrences between state N and state N+1.

Looks like this missed the opportunity to load the board in JS from the URL to be truly static.

After a candidate's defeat in an election, you will be supplied with the "cause" of the voters' disgruntlement. Any conceivable cause can do. The media, however, go to great lengths to make the process "thorough" with their armies of fact-checkers. It is as if they wanted to be wrong with infinite precision (instead of accepting being approximately right, like a fable writer).

-- N.N. Taleb

they've gone from barely stringing together a TODO app to structuring and executing large-scale changes in entire repositories in 3 years.

No they didn't. They're still at the step of barely stringing together a TODO app, and mostly because it's as simple as copying the gazillionth TODO app from GitHub.

Excellent advice.

On the technical side, I believed waiting was due to the lock queue rather that having acquired an ACCESS EXCLUSIVE lock. The ALTER is specifically _waiting_ for any lock lower than ACCESS EXCLUSIVE to be release.

I'm not sure information theory deals with this question.

Since this isn't lossless decompression, the point of having no "real" data is already reached. It _is_ inventing things, and the only relevant question is how plausible are the things being invented; in other words, if the video also existed in higher resolution, how close would it actually look like the inferred version. Seems obvious that this metric increases as a function of the amount of information from the source, but I would guess the exact relationship is a very open question.

I guess 1080p was the big shiny edge thing back at the time. 35mm can supposedly be scanned beyond 8K, so you could theoretically consider 4K filming not good enough neither.

Most modern navigation apps continue working in tunnels and other places without GPS. It's more like GPS augmented with dead reckoning.

Many comments are missing the point here (although the article doesn't properly explain neither); it's not about resolution, but about fixing imperfections in filming:

The recent Cameron restorations were based on new 4K scans of the original negative, none of which needed extensive repair of that kind. [...] The A.I. can artificially refocus an out-of-focus image, as well as make other creative tweaks. “You don’t want to crank the knob all the way because then it’ll look like garbage,” Burdick said. “But if we can make it look a little better, we might as well.”

The only movies which would require upscaling to 4K are those released between about mid-2000s to mid-2010s, the advent of native digital cinema, but filmed in 2K. Everything before was filmed in 35mm film, which can be scanned to 4K with information to spare; everything after is filmed in native digital 4K or more.

Moreover, upscaling which deal only with resolution has absolutely no need of AI. Any TV will decently upscale in _real-time_ a non-4K movie, and more sophisticated techniques can give basically indistinguishable results. 2017's _Alien: Covenant_ was voluntarily produced in 2K but released in 4K through upscaling and the image look just great.

In some sense this is almost tautological.

Yes. The interesting property would be the reverse proposition: what percentage of victories are granted by not blundering?

In amateur level chess, that number is very high. That's the point the author was trying to make.

No. Use integers to store the smallest money decimal, and store the currency name alongside.

The vast majority of humanity is Poor Brain and therefore actual money is what matters.