Companies ... and countries. My country (Norway) got cancelled by China after the Nobel Peace Prize was awarded to a Chinese human rights activist in 2010. As a result China cancelled Norway, and it was only 6 years later that the relationship was somewhat restored, after which the Norwegian government has been *very* careful not to upset China in any way, shape or form. I don't like how Apple applies Chinese censorship to its global business, but I can emphatize that it must be a major pain to walk the balance between doing business with China and pleasing China, while catering to the rest of the world.

Hopefully, as the rest of the world may have finally woken up and asks itself "why the h... did we actively and willingly place China's cold hands around our economies and supply chains (simple: because of short term thinking, greed and profits for people who gained who have long since cashed out), and what can we do to get out of this mess", things will get better as countries and companies take more of that power back. Though as long as you want to do business with one of the world's largest economies, you *will* have to keep making moral vs economic choices.

But ... as Chinese companies increasingly take over (wasn't it Huawei or some other Chinese company that just dethroned Apple in their domestic mobile phone market), and perhaps start to go isolationist, maybe that will partially solve itself. And maybe the tables will somehow turn, as Chinese companies require access to Western markets to sell their increasingly competitive products. And the rest of the world will once again have a choice; do we promote the current situation by buying those products (at the expense of non-Chinese companies) because they are cheaper and perhaps also better at the same cost. And ... do we ignore the fact that (going full tin foil hat mode here) those products may potentially be rooted in some way that whenever you type something on your Huawei smart phone someone in Beijing is monitoring, and the latest software update for your new Chinese car contains a nice little piece of code that makes the car do ... whatever they want, insert your favourite evil Hollywood plot here.

There is a lot of Apple trashing here, which I think is somewhat deserved, though I can see why they end up doing it. And ... they kind of made Jon Stewart's episode for him; he is a brilliant man who I have been following for years, and I would very much have liked to see what he would have made - on the other hand, there will be a lot of publicity around Chinese censorship, and he will be talking about it on talk shows, in interviews and on debate panels, for month - maybe ending up giving _more_ attention to the entire thing.

One single drone is not much of a problem. But there is Kant's categorigal imperative, for which a litmus test whether is something is ok, is whether it would be ok if you applied that rule to everyone.

So imagine everyone is using drones to get pizzas, hot meals, legal prescription drugs, whatever ... all times of day/night. If you live in a moderately densely populated area, you will have drones whizzing around all the time. Even if the noise was not that loud, you will constantly be pinged with drone sounds. And there is the visual pollution - you can no longer sit in a park seeing reasonably clear skies, you will have drones everywhere vying for your attention.

It's fine when drones is a little pet experiment of some company somewhere. If everyone was doing it? Then you have the electric scooters in Paris situation.

The airspace - low and high - is a shared resource. And by zipping your drone over my place, you will be annoying me visually and audibly.

This is a little bit like the electrical scooters that popped up unregulated in cities across the world. At first self-service scooters were hailed as a great new thing. Corporations, being corporations, grabbed that opportunity and did their capitalist thing. Then the numbers of scooters littering the sidewalks and the number of accidents, skyrocketed. And now e.g. Paris has banned them

The airspace should be reserved for the stuff that actually deserves it. Emergency vehicles do not flash their lights and use their sirens unless they are rushing to get somewhere in order to prevent loss of lives or crimes. We should have a similar level of respect for the use of the airspace. If we want commercial players to be allowed to use the airspace, we should tax the **** out of that use, as a way to get to use that public good. It should be so expensive that you would only use it to deliver gold-crusted pizzas. And even then, partially because there are people with f*** you money who could fill the air regardless using drones to buy a pack of gum, the amount of stuff we allow to go up there should be very limited.

"We're the best because more people use our search engine as standard", or "more people use our search engine because we have a couple decades of monopoly advantage, during which we have so firmly entrenched ourselves that noone else is able to properly get into that market, and if you are not one of the few global players with very deep pockets don't bother even dreaming about trying".

It's kind of similar to dictators who can argue that the people love them because they win "elections" with 109% of the votes.

there's ChatGPT4 which is confidently hallucinating and making stuff up

Fits my experience perfectly, and from now on this will be my go-to description of what is an AI language model: It is a computer program that hallucinates and makes stuff up. Perfect!

Copyright laws set limits on what is allowed. The U.S. Copyright Office FAQ has info on works published from 1978, and before that it seems things get more murky. But ... if the record companies actually hold copyright (which my hunch is they do, based on how U.S. copyright law seems to be about sc***ing the consumer whenever possible), then it would seem reasonable others are not entitled to distribute the copyrighted works.

We may not appreciate that. I think it is perfectly unreasonable. Don't like it? Engage in politics and/or lobbyism to change the law. Seems like an impossible quest? Engage in politics to try to improve the system somehow.

When it comes to _preserving_, that is something which is possible to do without distributing. A country can put in place a system which requires copyrighted works to be provided to an archiving service, and provide distribution of content at the time when it enters the public domain. There are obvious risks to this approach - it takes only one non-democratic administration to make an executive decision on a good old "book burning", to lose content forever. Non-profit organizations could do the same, which mitigates that risk somehow - though it is hard to do as systematically if there is no formalized system for handing them copyrighted works.

It is much harder to verify something which someone (something?) else has written, than writing something yourself. At least when _you_ write something, you know why you are writing it - even if that is "man, at at this point I am just inventing stuff". My strong recommendation is to _never_ have ChatGPT or another language model write something for you if the factual content of whatever that is, is somehow important, and you are not willing to check _everything_. At this point, language models would be better for some sort of "I wrote this text, take that text and polish its structure and form, keeping the contents".

I got to hand it to the AI language models ... they lie so credibly I am not surprised a lot of people get charmed by their eloquence and professional language, sort of like the lure of the alpha demagogue. I have on occasions asked medium-complexity factual quesions to Google Bard (yeah ... it's Bard ... I did not bother with registering with ChatGPT and handing over my soul, but testing Bing which uses ChatGPT also did not impress me), and it got it absolutely wrong - every - single - time.

AFAIK, at least for offshore production in my country (Norway), natural gas was originally a byproduct of oil extraction. What did they use to do with it? They had huge torches on the oil platforms where they simply burned it off. If you are going to be producing natural gas anyways, and have to burn it, then might as well get some use out of it. These days it is treated as a valuable product and is shipped off to Central Europe in huge pipelines.

Natural gas is obviously not clean - it is fossile fuel and releases captured carbon into the atmosphere - any other claim is ridiculous. But as long as there is oil production, there will be natural gas. Moving to a sustainable energy economy means dialing back those two energy sources at somewhat the same pace.

Dropping use of natural gas without dialing back on oil, simply means that gas will be burned off elsewhere, with the same impact on the environment, but getting no benefit from it.

A password manager (though necessary) can be a huge risk. It is a single point of failure "steal all your credentials and take over your life" type of product. I would never put any truly important passwords into a widely used syncs-stuff-and-magically-integrates-with-browsers product. Because if that stuff gets hacked, you are seriously pwned.

Personally I use PasswordSafe with some old binaries I have used for years; I figure if those executables were compromised, I would have been hacked already. Decided to take some risk and use StrongBox for accessing the PS database on Mac. For the most sensitive passwords (the "we can log in to your bank account and take all your money" or "access government services and do whatever like change your legal name" kind of passwords) I do a little obfuscation of the stored passwords, which prevents them from being stolen as-is if the software compromised, plus passwords are never copied in their actual form into the clipboard. Such obfuscation will not protect your nuclear launch codes, but for an Average Joe who does not have a big target on their back, it should hopefully be enough for most hackers to not bother and move onto an easier target.

Some non-sensitive passwords for stuff I need access to frequently, I keep a copy in Google Keep, with some obfuscation - plus allow Firefox to keep some passwords like Netflix and such.

Some bonus thoughts, relevant to the general topic, but not to the question of OP:

Bonus idea - I never made an app for this, but ... it would be very simple to make a program which generates any password for you by taking a single master password and concatenating with some service-specific text (like the domain name of a service plus your username), feeding that as a seed for a cryptographically strong pseudo random number generator, and extracting a password from it. Bonus: only need to remember a single password. Risk: that password unlocks _all_ your other passwords. But ... that is what a master password for a pwd management product does anyways, isn't it?

Second bonus idea, and this one I have actually done - I was involved in an initiative a while back in which we thought about ways to use public key encryption for logging into service, in which the key pair itself is what you use to identify your user and authenticate. This would have lots of advantages for obvious reasons. The big problem? You cannot remember or meaningfully enter all that keydata, and there is the huge problem of managing your keypairs and copying them around.

So we did the analogue to what I suggested above for generating per-service passwords from a single master password, but the procedure instead generates a keypair, and it always generates the same keypair from the same set of input data. This stuff you can easily do in python using the versile-python3" library which is available from PyPI (pip install versile-python3). Here is all it takes to do it,

from versile.quick import VCrypto, VX509Crypto
purpose = 'Slashdot demo'
personal = 'CptJeanLuc'
password = 'GreatestPwd4Evva'
identity = VCrypto.lazy().dia(1024, purpose, personal, password)
x509_privkey = VX509Crypto.export_private_key(identity).decode('ascii')
x509_pubkey = VX509Crypto.export_public_key(identity).decode('ascii')
print(x509_privkey)
print(x509_pubkey)

and here is the output.

-----BEGIN RSA PRIVATE KEY-----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-----END RSA PRIVATE KEY-----

-----BEGIN RSA PUBLIC KEY-----
MIGJAoGBAKZkw/LpyfeodeaaaZlqd1C4aOnEHYLRijmaWbgFP0teD/KQNQhAmto1vcjKr5nt0ZSm
vKSIk+4sXW98g+aKz2ZiIYD/Axoip0TL+RXe59bmseUmtvyqQiLWTbtpKprCtmVfejQnEnz064yT
z0TK/GE37Rlu0Xgk59+jiCfUuqtNAgMBAAE=
-----END RSA PUBLIC KEY-----

Versile Python was written some 10+ years ago and is not maintained, but it is LGPL so you can just run with it and do something fun. Documentation is not currently available online, but you can download the Versile Python source from github and build it using Sphinx (pip install sphinx and make html).

Versile Python does lots of stuff, and the cryptography/X.509 bits is only small part of it. From PyPI: "Versile Platform is a set of open protocols enabling object-level service interaction between heterogenous technologies. The protocols are designed to enable simple yet flexible and powerful patterns for interacting with remote services or running services." There was also an implementation of the Versile platform for Java (also available on Github), but that probably does not work with the later java runtime versions.

The real reason for this AI is to answer the question "why on earth did I buy Twitter"

They can only make it hard to leave the service, if they have made it at least as hard to sign up.

• Hard-to-find unsubscribe option? Equally hard to find sign up button.
• "Please state your reason for subscribing"
• "Are you sure you want to subscribe? Here are 5 things you could do instead - play yahtzee, go for a walk, ..."
• Here's all the stuff we will _actually_ be doing with your data. You really ok with this???

TLDR: uncool to change this for already sold printers, would be ok for new printers. Buy another type of printer if that is what it takes to bring your cost per page down.

I find this to be an unacceptable change of practice for existing printers which have already been purchased. Consumers have bought those printers in an environment in which they, and HP, have been aware of the existence of alternative supplieres of cartridges - and that has been part of the decision to buy - as well as to sell - those printers. Not cool, HP.

I think it would be ok to do this for new printers, if they made it very clear to buyers that only cartridges from HP will work with the device. Then consumers can make an informed decision. In a way, this also makes sense. My understanding is those printers are sold cheaply (too cheaply?) because part of the model how those printers are sold, is the supplier will recover that loss from cartridges.

If consumers don't like that model, they can look for an alternative. I don't like the "pay lots every time you need to print a little" model personally, so I went with an inkjet printer (from HP, as it happens) that has tanks for refueling ink. The printer was more expensive (and sadly had to be returned twice before I got a model with acceptable print quality for black ink, so not sure about HP's quality control), but the cost per page (when the volume of printing starts to accumulate) is lower. Not sure whether I can buy ink from alternative providers, wasn't a big part of my decision.

There is some silver lining, though, to the "buy expensive cartridges" model. If you only need to print every very rarely, then that model provides you with a very cheap printer, and as you never need new cartridges, it is way cheaper than alternatives.

A colleague of mine posted how he had asked ChatGPT to write a piece of python code to do a specific task, and behold ... it did! Except when I had a look at the code, it did not do what it was supposed to at all. And this is not limited to coding (for which ChatGPT was not designed, as a language "auto-complete-the-next-word" tool, so there is no reason to believe ChatGPT should be able to code at all - yet it will still tell you it can solve your problem), it will answer questions with responses that are plain false or half-true.

The only worse thing than having no AI, is having an AI that lies to you. It is like [insert favourite wannabe dictator here] who will deliver big lies with a straight face, while packaging the lie inside a solid posture and authorative and uncompromising language, to make you believe.

It is a useful tool for what it does well. But using it to replace a search engine? Depends. If I am using it for something serious, like "I will go to jail if I get this wrong" (like not paying a fine in time) or "there is a non-insignificant amount of money on the line", then would I trust ChatGPT? No way. I will happily waddle through 20-40 old-school search results and figure it out myself, thank you.

A lot of strong opinions on this topic, I see, and somwhat harsh words here and there. Thought I'd pick this one to give a response.

I listen to a lot of science podcasts like Science Friday and a big one we have in my country called Abel's Tårn (my country is Norway, we are high up in the Northern Hemisphere, so we are quite affected by DST during winters when daytime is short - if you have not experienced it, try spending a couple months in December/January getting up when it is pitch black and the sun does not get up until an hour after you got started at work, and then I will listen to opinions on how DST does not matter - and I do not even live up north in our country where they get hardly no daylight at all).

Something I have picked up, is that we evolved to follow the day/night cycle. There are biological processes that are attuned to "follow the light". Before electricity, people would get up and go to bed largely based on the time of day. Then electricity started messing with things, though.

When you get up in the morning and it is dark outside, you are working against biology. When you are working trying to stay performing and it is pitch black, you are working against biology. DST tries to even things out a little. Without DST, you are working somewhat harder against biology. It is as simple as that.

Everyone to their opinion. As a mathematician dabbling in physics, I like the "purist" perspective that everything is simpler (in maths we like simple) if we just count oscillations of a cesium clock, and use that to define the time. period. But recognizing I am a human being, and it makes my life easier if we organize things so that society tries to work with rather than aginst human biology, I can accept the "anomaly" that is DST.

Alternatively, why not go the other direction? If we are going to be purist, let's define a single time zone for all of earth. Or get away with common time alltogether, seeing that general relativity makes the notion of a common time frame on earth useless anyways.

And yes, as someone mentioned, electricity screws up biological clocks. But how and whether you use electricity, that is on you. If you want to browse Facebook until 1AM with blue light tuned up to the max, that is your choice. Whereas I do not have the choice to come to work whenever it suits me.

Nuff said. Strong opinions on this topic. But ... it seems the anti-DST people are out in force, whereas I do believe the pro-DST people are a silent majority. So makes sense someone voices a few opinions from the other side.

There are two options. Adapt technology and societal systems to fit human biology, or adapt human biology to avoid having to adapt technology or societal systems. Or a third option, do some serious geoengineering and fix earth's "broken" tilted axis.

Human biology is adapted to follow the day/night cycle. The inconvenience of adjusting the clock twice per year is small compared to the pain of living for months every year with a misaligned biological/physical clock.

If we remove daylight savings, I guess human biology will adapt in some thousand years, if we are still around by then. Which at the rate humanity is making stupid decisions at the moment, I would not bet significant money that is going to happen.