Responding to myself to add a point.

Businesses that raised their prices likely lost sales. Basic Econ 101 that most goods have lower sales when prices go up. So a business simply repaying the tariffs to its customers is likely to be a net loser. These businesses will not ever fully refund the tariffs paid.

How are you going to pay "the consumers" back? Everybody an equal share? How about those people who intentionally buy American and hardly buy anything else? Should they be reimbursed? Do you have receipts that show how much extra you paid? Without a doubt, prices were raised. But quite often a lot of businesses chose to reduce their own profit - i.e. the costs of the tariffs were shared and prices weren't raised as much as the costs. Sorting this sort of thing out is just about impossible.

I don't understand it could work any other way than the people who directly paid the tariffs (businesses for the most part) should be the ones that get their money back. If they in turn choose to reimburse their customers, good for them.

I do get annoyed at all these modern programs that treat CR and LF identically. CR should just go to the beginning of the line (and not scroll down!). The text that follows should overwrite the line. Just like typewriters do. /s

AI translations still need LOTS of human help to have any sort of quality. AI tends to mostly stick with word-for-word translations, which often screw up idiomatic expression and often is very stilted in the target language. Yes, it is much better than nothing - and can even be the starting place for a good translation, but I wouldn't count on it for anything important.

It is an interesting question what Apple will do. They likely have long-term contracts as well as much better than average negotiating power. Apple might be able to hold the line on pricing and in doing so gain market share. The pricing for Neo was set well after the spike in prices, so they must have some confidence that they can keep the price steady. Maybe the same for their other products.

Be aware of unexpected correlations. Maybe DUI will go up because people now have to drive to another state to get their fix.

Remember, a user's goal isn't to learn a new piece of software, it is to do the task at hand.

I don't think it is so much fear of change as annoyance in having to learn something new which has no direct relation to the task at hand (e.g. writing a document). Why spend time retraining when you can just get your work done? This is especially true of power users who have been using the software for decades, who know every little trick on how to format a document, how to get nice looking tables, get the page numbers in the right place, etc.... Even if the other software can do each and every thing, the way to do it will be different. It can literally take years to achieve the same level of competence in another piece of software (for no net gain in productivity and a net loss for a period of time).

People will of course continue to go into work to collect their paycheck, but they will be annoyed and be less productive for a good long time. Eventually they'll learn or retire.

The reasons a country would be motivated to try to move is that despite this loss in productivity is that they believe that the gain in sovereignty will outweigh other considerations. The US has shown it is willing to abuse sanctions and nobody wants to be vulnerable to extorsion. In the end it will cost US business a lot. But these things take time. Probably 10-20 years (at which point there will be people expert in the new software).

There is a vast difference between city-level and country-level resources. There is also a different motivation: fear that the US will abuse its power. This isn't an abstract fear. The US sanctioned members of the International Criminal Court - forcing them to lose access to their e-mail, credit cards (Visa/MasterCard), social media, app stores, ability to transfer money abroad, etc. In principle they could sanction a whole country. Trump has shown himself to use whatever tools he can use as leverage in negotiations... and probably the next US President will continue (these things tend to be "sticky"... once one president acquires a certain power, the next president will use it too). Countries don't want to be at this sort of disadvantage in negotiations and I think many are waking up to the fact that they need to act.

The basic problem is that you are likely throwing away decades of experience on average per user. For example, I have been using MS Word since the late 1980s (ok, I am older than average; average is probably about 20-25 years). Yes, it has changed over time, but many concepts remain the same and I've had continuous re-training. Could I use another word processor? Yes. Will I be as effective using it? No, not for 4-5 years of using the new program. Knowing the new way to use styles, format pages, etc, just takes time to learn. Of course, I could make basic documents using LibreOffice or whatever alternative program is proposed, but it just wouldn't be as good or as fast. Learning to use a new program isn't the primary task... getting a document written is.

Excel is probably a much bigger problem than Word. There are a very large number of power-users of that program who just won't be as effective using other programs. It is used extensively in the corporate world and likely in governments as well.

When you're talking about an OS, it is also file management, etc. These are probably faster to learn as most users probably don't do sophisticated things with their file organization, etc.

If you change all at once, you are asking for a lot of user complaints, a lot of confused people stumped on "obvious" problems. I am not arguing that it shouldn't be a goal to use open source, I am just pointing out some problems. For anyone based outside the US, I would think that migrating off of US-based products for anything critical would become essential given the way the US has recently abused sanctions (e.g. sanctioning members of the International Criminal Court). The US just has far too much leverage.

Well, we don't yet know whether they will be allowed to re-enter the country.

Well, it could be that progress is limited because the scientists keep sampling their wares.

The problem is that the people who are trusted to make changes got hacked by a sophisticated, well-funded, highly targeted campaign. There is no technical solution to this. Somebody has to have the right to make changes and these people are human and will make mistakes.

I think this sort of problem could happen with any sort of package manager. A developer targeted by a sophisticated, high-budget fake is not a problem specific to NPM. I don't really know how the protocols for publishing Maven, NuGet, PyPI, etc. differ ... but I suspect all are vulnerable to the same sort of attack. Maybe NPM is more vulnerable than the others, but it is only a matter of degree.

Amongst (a lot of) other things, they have a successful line of cameras.