Uber, Lyft etc: No we are not a taxi company at all, we just help people that was going somewhere to take someone else along. Our service just by chance looks and functions as a taxi service, but no, regulations protecting customers or drivers do not apply to us ...

AirBNB: No, this is not an apartment hotel at all, it's just a service allowing people renting out their flat for a while when away (or buy several flats they never live in and drive up prices while dodging regulations)

"Prediction markets". Not gambling at all. It just ... well, that's just plain dumb. Same thing, new name.

Please stop.

Night trains with sleeper cars does that, if you are in a location (several parts of Europe) where such services are available and serves major travel routes on suitable distances.

Yes, obfuscation may help a bit. Even if people probably want a notification within - perhaps 30s?

And note that this isn't about Signal metadata, but from the Signal app to the OS. if you have the notifications for *single device* and they match the time stamps of a sender within 5 min, with no other notifications (if you only communicated with that sender), it may be indicative, even there is some p for it being a coincidence.

Even non-preview notifications can be a compromise. Metadata, as usual.

Let's say some malevolent agency got the unlocked phone of your friend, with messages (perhaps "let's meet at the docks at 9:00", "are you there" ...) sent to a Signal username at specific times.

And then you have OS notifications "you have a Signal message" done at exactly that times + processing delay. That would be a bit revealing.

Possibly by flying. How so? As an IT nerd, you are probably aware of the process of "bootstrapping" - using the current environment with it's limitations to try getting something better running.

Now on the other hand, flying for leisure, without your activities at the destination even trying to help resolving important global issues, is a bit harder to justify. Even harder to justify is action (or inaction) that outright blocks people from living more sustainable - for example blocking construction of bike lanes, cutting down on trains instead of improving them or putting fossil fuel lobbyists in charge of environmental agencies.

For intra-Europe travel, we will do quite fine, yes. Intercontinental travel will be much harder, but at least travel to US had already declined, in part because people don't want to risk getting randomly detained.

But yes, we already had similar situations - for example when the EyjafjallajÃkull eruption closed air space. Trains will be crowded, long-distance buses will have a boom time. The night trains will be very sold out. Some people will probably drive. Some conferences may have to be re-scheduled because of longer travel times, or go digital. But we will manage. Travel within Europe can work quite OK without air travel.

Fossil gas dependency is another matter - some countries still rely on it for heating and industry, and we don't want to buy Russian gas for obvious reasons. It was replaces with LNG, imported from Gulf countries or the US. The best we can do now is probably lowering temperatures a bit in homes, starting insulating better for the winter, and in general help central Europe getting heating less dependent on gas, as we already have in Northern Europe. When we can cope with only gas from Norway / the UK, we would have better resilience, but of course zero fossil gas would be even better.

Text adventures could have some UI similarities with speech based navigation for visually impaired users, whether the navigation aid is dog-shaped or not.
"You are in a dimly lit forest groove. A squirrel sits on a tree stump one metre in front of you. The squirrel looks confused. There are exits to the north, west and east." (or perhaps 9,12 and 3 o clock)

Also, I could imagine some scepticism on the chipmakers part on whether most current "AI" is a viable business, or if they are mostly driven by investment money with disproportionally less actual revenue coming. If the current "AI" craze has elements of a bubble, chip makers won't hesitate selling to AI companies for inflated prices, but will be far more reluctant investing in capacity that may or may not pay back.

If affects us to some degree. Petrol - for those that use it - is a bit more expensive, and it probably will disturb still fossil-sources industrial supply chains after a while. There is still natural gas dependency, we don't want to buy from the invading dictator in Russia, and the supply may also diminish.

But we generally have more sensible cities, where it's rather easy to ride a bike or even walk to services, kindergarten and often to work. And public transit is much more abundant, even if of course people have issues with quality or punctuality in places.

I'm not very updated on "TDS", but I assume it's about the mindset that makes people elect a convicted felon and general con artist with a high score on false public statements?

How come blatant large-scale download of online and offline copyrighted works and reuse in derivative works is OK if "AI" corporations does it, but download by private persons of similar works for reading / viewing is classified as piracy? (OK, besides the "AI" hype that it's somehow not a derivative work, "corporate" may be a hint ...)

Interesting reading on the exuberant tuition fees in the US, and how it hinders people from getting higher education. I live in Sweden, and we have *no* tuition fees, at least if you are a citizen. College / university is free - you just have to be accepted to the programme / course, which is usually done on either high school grades, standardised test results or previous academic merits. Even travel (for example excursions when you study biology, geology, urban planning ...) is usually free.

Every student also get a small state-funded grant (around 3000 SEK / 300 EUR each month) for at most 5-6 years of studies from CSN, the central study grants board. There has been proposals from student's organistions to increase this part as it is mostly not enough even for rent, but currently most students also take a loan from the same agency or take part-time jobs.

Firefox too it seems, even if I haven't tried it.

Honest question: Could local text translation be a good use case for local ML models? Nice not having a remote translation service getting to know what text you read or write.
That said, such things can probably be done OK enough without specific NPU:s.

(otherwise, agree it sounds like buzzword of the year, with usage of rather overestimated)

No, it does not have to be a laptop - it can still be a phone with typical form factor and use cases - staying in the pocket, casual chat, phone calls and video, outdoor navigation ... Case in point: I had a Nokia N900 in 2011 - it was phone-sized (a bit thicker than modern phones but not much), and could do all phone things. With wonderful two-thumbs text input. In addition, it had a resistive instead of capacitive touch screen, so I could use it even with thick gloves - Finnish design considerations?

Yeah, you could use it as a "mini-laptop" - it even ran Libreoffice and gcc. However, that was not how you typically used it, except for exceptional cases.