anachronist @ anachronist @midwest.social

Posts

9

Comments

518

Joined

2 yr. ago

"Self driving cars will make the roads safer. They won't be drunk or tired or make a mistake."

Self driving cars start killing people.

"Yeah but how do they compare to the average human driver?"

Goal post moving.

Look up "interurban railways". Most towns east of the Mississippi used to have frequent rail service with whistle stops at every farm and crossroads. In addition to passengers these railroads also transported the harvest, Sears purchases, kit houses, even hearses!

When I first started using DDG I would use the bang to get me back into google very often. Now whenever I use a browser or device (mis)configured to use google I feel like a guy who accidentally launched and tried to use IE8.

I switched from Alta Vista at Google in the early 2000s because the Alta Vista index was stale and full of spam. Google search tools were comparatively primitive (av let you do things like word stem search) but the results were really good.

Unfortunately Mozilla is being run by a McKinsey consultant.

The planned goal of the mission was to achieve orbital velocity but not orbital trajectory. This was because they had not yet demonstrated the ability of their vac engines to relight in space. If they go into a stable orbit but can't relight they can not deorbit and they become space junk.

They initially claimed that this was a success (they achieved target velocity) but subsequent analysis was they were quite a bit off. Also because their engine relight test was failed/cancelled they will also not be allowed to attempt a stable orbit in IFT4. They have to demonstrate relight/deorbit capability before they will be allowed to attempt stable orbit.

Which part of the video is wrong? The fact is that it failed to reach planned velocity. This is public record. If it did not reach planned velocity then it did not reach the non-circualized suborbit that they intended. They were not "just a circulization away from orbit."

The CSS channel was created when Musk and Shotwell were making bonkers claims about their Mars plans, as well as other crazy bullshit like the suborbital rocket airline stuff. The point of CSS is that none of their claims pencil out if you do even basic math, and they proved that by doing the math. They've also gone after other space grifters like orbital assembly.

The appeal to google and friends is that it's even less obvious when you're being advertised to when a LLM tells you something than on their existing SERPs.

Google was already going downhill but when they fired Matt Cutts and replaced him with an advertising person was the point where it was obvious they weren't interested in search anymore.

Yeah that was the joke. 🙃

Edit: Also none have made it to orbit or even near orbit. They initially claimed that the third one made it to the non-circularized suborbit they had planned, but later analysis was that it did not actually reach the planned velocity:

https://www.youtube.com/watch?v=fQ03eVRgiZ4

Spite and pettiness seem like a poor way to run a business but what do I know? I'm just a guy who's gotten zero starships successfully to orbit.

Interesting how most of the points about how "nobody wants to live here" boil down to the housing crisis.

It's a bit like that old joke about the trendy restaurant: "Nobody goes there anymore, it's too crowded!"

Alternate theory we'll look back the same way we looked back on the claims that IBM watson was intelligent, or the claims in the 60s, 70s, 80s, 90s, 2000s, 2010s, that

<insert technology x>

A simple path forward, is to go from classifying single elements of training data, to classifying multiple elements and their relationship in the training data.

Training data already has multiple labels.

Slightly less simple, is to gather orders of magnitude more data, by just hooking the input to an IRL robot.

An entire point of the paper and video is that massive increases in training set size are showing diminishing returns.

Another step, is for the NN to control the robot and decide which parts of the data require refinement, and focus on that.

🤡

I wanna see Sam Altman reenact the "make it work" mirror scene from always sunny.

https://www.youtube.com/watch?v=kBXNa0ipFPQ

Pretty crazy to think that it is actually not sure whether spending less than 500k on a supercomputer is worth it.

Has more to do with the market for supercomputers. They are monsters to keep fed so it's not a question of if you can buy it but rather if you can run it. But customers for supercomputers are in the market because they need the most raw power that the technology is capable of supplying, so buying and installing a decade old supercomputer (which is going to have the same operating costs at a lower capability than a new one) doesn't make sense.

You also have to consider that the downtime's going to be a lot higher on this equipment as you're going to start having components hit the end of their useful life.

The biggest con is the industry's war to make Kei trucks illegal in the US.

shit conscription is evil

This is kinda a weird take. War is evil but conscription is a universal aspect of every total war. Every country on earth reserves the right to conscript its citizens in the case of national emergency.

Tictok is horrible but this entire ban is being driven by the swamp being upset that young people have come to the wrong conclusion about Israel. I think that the idea is that as long as people are using facebook or google-owned properties, people can be shown only information that will lead them to the correct, approved opinions.

These are almost certainly saleried, exempt employees with no "timeclock".

They were fired for expressing a political opinion and doing so in a way Google did not like.

It is certainly legal for Google to fire them for this because it is legal for Google to fire them for almost any reason. But it's also pretty certian that there is no way in America to protest your employer in a way where the law would protect you from retaliation.