- > I witnes far more people screaming against AI.
If you shove it into people's faces, they will have knee jerk reaction and hate it.
If the AI industry didn't desperately try to push it in every possible way in desperate bid to be profitable and it was just a thing that slowly gets better and is value added, not a nagging push, there would be far less of that.
But companies like MS have idea of consent of average rapist and will not even give option "no, I don't want copilot in teams", there is only "add it now" or "remind later"
- There was no thought process
- So they won't pay you and just scrape pages that have it public, and you will never get traffic from search again until you let them scrape
- the AI will just run chrome instance
- it's few orders of magnitude harder given the amount of SEO spam prevalent, and that just gonna get worse with AI
- My experience with it is that it somehow took 20 seconds to load (site might've been hn-hugged at the time), only to "protect" some fucking static page instead of just serving that shit in the first place rather than wasting CPU on... whatever it was doing to cause delay
- > USB-IF are the prime examples of "don't let engineers name things, they can't"
Engineers don't make names that are nice for marketing team.
But they absolutely do make consistent ones. The engineer wouldn't name it superspeed, the engineer would encode the speed in the name
- less worse.
It would be like having Quadro 6000 and 6050 be completely different generation
- Intel is allergic for making consumer stuff good. Remember how in consumer range like half of the chips had fucking virtualisation disabled, long after competition had it on everything ?
- For Apple, datacenter stuff is low margin business
- > “why is I/O in docker slow, and how would you improve it” is pretty esoteric knowledge now, but would have been considered basic knowledge (for other applications, not specifically just docker) only 12 years ago.
you could've used docker for 12 years and never hit it if you used it on Linux, and followed sensible practices (mount the data dir from outside so it can be reattached to upgraded version of the container)
- I don't think many orgs learn all that much from coaching their juniors, at least after first few.
Juniors are just... necessary in the balance, have to little of them and the mid and senior devs will get more and more expensive, so you hire a bunch of juniors, they get trained on job, and it balances it out.
Hell, if company does it right they might underpay junior-turned-senior for decade before they notice and look how industry pay looks like!
- I see that you know nothing about anything involved. The cost of cable is essentially irrelevant in both cases, the "everything else" is the expensive part.
Power can be had from the closest high voltage line (and maybe even easier to get outside of the city)
The fiber have to be dragged from either nearest point of presence (a building with many fiber connections coming to it from multiple companies to exchange traffic), or to whatever dark fiber infrastructure is available nearby.
In city, that's usually not that hard, ISPs already "plumbed" most of the bigger cities with fiber infrastructure.
Middle of the boonies, where we want the datacenters to go ? Dig, dig dig, get permissions for digging, get permission from everything around the ride from city's fiber infrastructure to the place in middle of nowhere, months or years in getting permissions, and red tape anywhere. There is probably some power close enough, or at the very least you can find a location close to power, but location close to fiber will if anything be some existing industrial centre, and even that might not be a sure bet. You might get lucky and get a permission to use existing poles to drag some fiber on them, but it's still PITA
- They are but as with everything else in last 10 years they are insanely incompetent at it so it will take a while
- Yeah intern contaminated by samples is easier to replace than the samples ;)
- > * In theory, you can test this with an anemometer. In practice, I've not found a cheap one that will so much as _turn the prop_ under such a light load.
Hot wire anoemometer ? Given it's rapid response you could probably even measure the turbulence in the flow. Probably possible to DIY too
- > I honestly don’t know why these things are so expensive. I’m guessing it’s a premium from the brand name, as well as quality assurance.
Being used mostly by medical (a small market with relatively rich companies) is the reason. It's a lot of testing to make proper one for not too massive market so the price is "eh, they can't be arsed to try to DIY that" basically
- Better comparison would be prices in various 3D printing sides.
But either way, margins for some companies like mentioned GW are huge
- Also resin printer would be better comparison, far better for printing small detail models
Precisely because A|b|c is simple and clear to write. For anything more verbose (and far more saner) we have Python already, and for other stuff Go is there too