awkward

This is just the fallacy. Political groups are coalitions not single monoliths.

I’ve seen a tremendous amount of content about AI water usage, mostly from pro AI sources. The most common type is comparing AI to particularly water intensive agriculture.

The result is that now I think water usage should be taken into account when siting data centers. Great Lakes and eastern seaboard fine, maybe not as much in California or Arizona.

Responding to tone but not to content is what a dog does.

Why would we expect John Deere to have a low AI adoption curve? It's main product line is just as computerized as Tesla's.

Geologically speaking it's just one really cool layer.

Code quality is a tactical concern and products live or die on strategy.

I wouldn't recommend neglecting tactics if your strategy doesn't put you on the good side of a generational bubble though.

There's a reason we had a few years of heavy anti police protest across the US.

Safety and freedom is incompatible with direct political control. Respect for truth is a false value if what truth is is to be put to a vote or changed to match the whims of whoever is in power.

The decline of technic sets is such a shame. There's so little support for anything but representative models of specific cars, despite the platform being able to support a ton of mechanical creativity.

To be fair that's more than a little bit present in most superhero media.

Kernighan's Lever - https://linusakesson.net/programming/kernighans-lever/index....

This article is perennially posted here and is probably the best breakdown of this quote.

You aren't renting walking distance to a butcher baker and candlestick maker for less than $3K for a studio. That's an aspirational lifestyle for a few neighborhoods.

Deeply hate this. Just add a small fee. It's just a couple bucks. What are you, cheap?

Open Source Software underwrites everything. It makes the largest human endeavors work. It makes silly ephemeral games little notes apps and digital art run. Turning maintainers into a kind of digital landlord that charges a fee is both insultingly low bore and enough to squeeze the life out of computing as a hobby.

Folding drying racks come in several shapes and are very affordable.

The article does go into this and gives lip service to the idea that a secure third party could expose age without exposing identity. Ultimately, there's still the problem that even if point of verification can be done in a zero trust way, you are still entrusting very sensitive information to a third party which is subject to data breach.

There are several humans who need to make decisions between bad training data and life or death decisions coming from an LLM.

Games are a zero marginal cost industry driven by hits. The cap is pretty high. The floor is what you should be worried about.

Unions are useful because they are a counterparty to negotiations with management. They have leverage because they are able to represent labor as a single entity. If they are only able to represent labor on one axis, but not on issues that represent quality of workplace, they lose leverage in negotiation that allows them to win larger salaries.

Having a relational database in a gui was great. The problem with access was that it tried to support network applications as a backend but supported a punishingly low number of connections. Having an application that crashed with 10-50 connections put it in an awkward space. Businesses without a strong technical team would build on it, release with N=1 load testing, and get surprised when it crashed out at scale. MS wasn't going to improve it, because they wanted more sophisticated customers to buy SQL server.

It's a myth in the most literal way. Fleming published and promoted his results despite a lack of reproducibility. By the time he won the Nobel Prize, he had backformed or misremembered a folksy story about an open window. That's textbook mythmaking.

It can both be fine to have a glib story to tell schoolkids and important to recognize that the actual intellectual process is messier and more complex.

So many other good details that get to how impossibly multivariate biology research is, like the need to have several days at the exact temperature.

It's not uncommon for results in biology to have this kind of snag in reproducibility even now. Sometimes it's due to attributing variations to something like "steady hands at the bench", but other times it can even be a deliberate attempt to prevent rivals from duplicating a process before it can be patented and privatized.

Chatbots enable self destructive manipulation of others at scale.

I like your angle, but most applications is a big difference from most companies. Serverless comes after deciding whether or not to break up the monolith, and after breaking up engineering into separate teams. It's a good way to manage apps with high variance in traffic while keeping cloud spend down.

It sounds like it's a B2B SaaS product that's gone through multiple pivots with very weak guidance on product on every step.

Not that I'm disagreeing with you, but it's a very common way for things to go wrong.

Tiles aren't just about data selection, they're also about caching. By turning a continuous domain (any part of the world at any scale) into a series of discrete requests (a grid of tiles at several fixed scales), maps become a series of cacheable requests.

This looks like a clever implementation of a useful feature. I enjoyed that the developer was able to link the streaming audio API to a useful visualizer, recognize problems with his approach, debug issues in that visualizer, and find a clever solution using graphics fundamentals like masking.

I'm less than impressed with the general consensus that he's somehow negligent for launching a feature that needed a fix, or that users don't want or need feedback about audio connectivity, or that the poster did something much better sometime back in `02.

The advice is all good. For the specific examples, claims with a lot of concrete numbers should be handled carefully. Unsophisticated users can bounce off numbers rather than content. Sophisticated users might care about the numbers, but are able to recognize spin - 200x speedups are clearly a bugfix rather than an innovative new algorithm.

Of course, if the statistic is actually core part of the product or a competitive differentiator, it should absolutely be highlighted.

There's some domains where the bitter lesson has big impacts on theory. The Peter Norvig vs Noam Chomsky debate on the merits of brute force compute finding a full and complete theory of language is an example. That's a case where the path of "get a ton of data and handle it statistically" competes with the path of "build a complete and abstract understanding of the domain." Lots of resources and lifetimes of work are decided by which path to take.

Agreed that overfitting the bitter lesson often leads slopping piles of compute and hardware at problems that could just be deterministic.

We'll all have jobs in blamefinding software companies. BaaS.

It's crazy to me that almost all new web projects start with two assumptions:

- Mobile first design

- Near unlimited high speed bandwidth

There's never been a case where both are blanket true.