I would be happy with a fediverse-spanning login.
When you stumble upon a link to any instance apart from your own, you basically can’t do anything, except manually copy/paste/edit the link so that it’s opened via your home instance.
Yeah, that seems like exactly the kind of selfserving virtue signaling that makes the left look like a bunch of edgelords.
But I guess, it’s easier to start such “actions” to feel like you’re doing something than actually doing something. (not that I’m actually doing something, but I’m not pretending either)
In German, there’s a relatively new phrase for that: Gratismut, free bravery. On the surface it looks “brave”, but it’s actually 100% riskfree and has no consequences whatsoever. But you do look brave.
No, it’s there other way round. You gain users/customers by selling domains and then upselling there.
I get your opportunity cost argument, but that only really makes sense, if Google plans to use the money/resources freed somewhere else, and it doesn’t seem like Google is doing that much investing lately. This small branch could have churned on, making a bit of profit, but bot causing and issues either.
Ironically, you’re manually doing a bots job on a post regarding bots doing human jobs.
Using that argument, domains could be used to upsell customers. Oh you want a domain? You know what? We’lll give you a year of gmail business (whatever that’s called) for half the price! Or maybe you fancy a 100$ Cloud Services voucher?
That way you can lock in customers.
Well, science is expensive. Even more so if you want to combine it with a metric buttload of technology transfer.
That’s actually standing law in Germany. At least for people and their homes.
So, they think they can find contracts worth 3b selling bullshit to customers.
I mean, let’s be honest here: AI will not be primarily used to find out new truthiness about the universe, but order butter at the right time. Or write basic essays, code, explain known things.
That kind of knowledge could easily be at categorized.
I’m surprised, these models don’t have something like a “ground truth layer” by now.
Given that ChatGPT for example is completely unspecialized, I would have expected that relatively there’s a way to hand encode axiomatic knowledge. Like specialized domain knowledge or even just basic math. Even tieried data (i.e. more/less trusted sources) seem not to be part of the design.
If you’re using Alexa to simply flick a switch, that is indeed a bit overkill. I’m using a cheap Trådfri (or something like that) light switch from Ikea. I have a dimmable lamp with three different light temperatures and can simply set the lighting from my couch. That’s really convenient. But it’s also point to point, since it’s just ZigBee. And it’s pretty cheap, 20€ all in all.
Well, of course, that’s just good engineering.
You see, kerpow!s scale much better than kaflooeys due to cache invalidation problems in the ooey inductors, that’s like first semester knowledge.
They’ll find some people.
The reality is, not having (good enough) mods will take a while to really hurt the bottom line. Subs will slowly deteriorate.
But I’m 100% sure, within a few weeks you can establish a new order of more servile mods.
This seems to be more about finding instances, less about actually posting localized.
BTW: why is feddit.uk hosted in Germany according to this site? I thought they brexited?