pbasista

> A lot of sites have terms and conditions which explicitly disallow the use of site content as a part of another service.

Having terms and conditions in itself is irrelevant. Because in order for them to have any legal meaning, it is necessary for the other party to agree to them.

An agreement can be implicitly enforced by law. Or explicitly enforced by the website itself before giving access to the data. If neither of those are present, there is no enforced agreement. And agreeing to it becomes optional. Such sites should be considered, in my opinion, publicly accessible.

> If I have a bike and you start renting it out without my permission, surely you are committing theft of some sort.

Of course. But that is a bad analogy. No one is "renting" or "taking" anything from those websites. The bots are just reading it.

Therefore, a better analogy would be that you have a bike, parked out in the public, and people are looking at it. By looking at it they steal nothing from you. And the bike and all of its parts remain yours at all times. That is a suitable analogy, in my opinion, to what those bots are doing.

I do not understand what "theft" they are talking about. Those AI bots were scraping publicly accessible internet.

Publicly. Accessible.

Of course there are some parts of the publicly accessible internet which host content that may be considered illegal or has been obtained illegally. If those AI bots used such content as well, it is fair to call it out as wrong, in my opinion. But that is a separate topic.

Blindly calling scraping of publicly accessible internet a "theft" is, in my opinion, disingenuous. Especially when coming from a company operating a web search engine. Which itself has its own bots scraping the same parts of the internet 24/7.

Exactly. The person holding the camera is the author. Even if they just point the camera and press the record button.

In a similar way, I would not consider AI to be the "author" of the output they produce. Rather, the person prompting it should be considered the author, in my opinion.

The sad part is that the US could very well have had made such a trade and companionship deal with Canada themselves. Or even a better one, should they wanted to.

They used to have an excellent relationship with Canada, the UK, with the EU and with other civilized and democratic parts of the world. Until Trump started to undermine everything. And have selfish and unfair demands against everyone.

Now the US has a "comradeship" with Venezuela, Russia and North Korea.

I know that the above is perhaps an exaggeration. But the fact that the US now has worse relationships with almost all the civilized, democratic parts of the world and is bowing down to dictators and authoritarians does not come from external factors. It is all self-inflicted. Including the Iran war and its consequences.

> Source for this claim?

For example this article: https://www.propublica.org/article/billionaires-net-investme...

The specific mechanism may vary by country. It often involves personal service companies through which the people are paid rather than being paid directly as employees. This usually means that they pay significantly less social insurance and health insurance than what they would have paid if they were regular employees.

To clarify, I am using the term taxes in a very generic way where it covers not only the income taxes but also all kinds of fees collectable by the government. Including social insurance and health insurance.

The Disney World visitors are a specific group of people. The vast majority of them do not go there every week, nor every month. So the behavior you observe in there is not representative of the routine behavior of those people.

In other words, yes, they may pay 5x the price for a meal in an amusement park that they visit once a year. That does not mean they are willing to pay anything close to that for a meal every week or every day.

What does not help, in my opinion, is that the wealthy people often have enough resources to optimize their taxes to a point where they pay much lower rate than their fair share. While the others cannot afford to do that and pay the full rate. That in itself creates inequality.

If a base rate that the society agrees on is e.g. n %, then, in my opinion, everyone should pay that n %. But in reality the poor people pay n % because they have no time to even think about optimizing it. As they often work around the clock to make ends meet. While the richer people hire professionals to take care of their taxes and optimize them to the extent that they pay significantly less than n %. And then there are super rich, some of whom, such as Donald Trump, pay almost no taxes at all.

This is, in my opinion, wrong and significantly contributes to wealth inequality. And therefore I believe that it needs fixing.

Why do you believe that?

There is no "gymnastics" needed. And those "middleware" layers are working just fine.

A user just opens the original X link in xcancel or some of the similar web service. And it will display the actual content only. Without ads and without other user-alienating features.

I think that it is also way simpler than "creating an account".

Well, I do not have an account and when I open any profile on it right now, I cannot read it properly. I cannot click on posts. I cannot read replies. I cannot open images.

In my opinion, that is quite far from "functional".

Such a behavior alone is a sufficient justification, in my opinion, for the services like xcancel to exist.

And if the X social network does not want their content to be publicly available, fine. They are free to gate it behind a login.

> Your ability to continue to perform this work can be withdrawn at any time for any (or no) reason.

Thanks to the fact that there are no widespread stories about this actually occurring in practice, at least not yet, people do not take it as a relevant risk at the moment.

> You have little ability to validate that the work is being performed as expected and isn't being silently nerfed or outright subverted based on competitive considerations, bribes, overactive 'safety', or cost management.

Yes, I agree that this is a real concern that many people might rightfully have. And I am unaware of any way to mitigate this concern while using black box AI models. Because the only thing that their creators can do is to tell their customers: "trust us". But there is no way to objectively verify whether they serve tainted AI model responses or not.

> CDU has been trying to adopt more AfD positions

> this keeps only strengthening the AfD

I am unfamiliar with the details of German politics but if what you say is true, I am not surprised at all. I do not think that one can "win" against those parties by using their methods or adopting their positions. Because they are genuine at it. While the others would only fake it.

If the voters have a choice between a fake populist and a genuine populist, they would, in my opinion, naturally pick the genuine one.

I think that one of the reasons is that the "fakeness" is not very appealing to anyone. So they prefer the "real deal".

> that makes the far-right parties grow stronger

That issue is just one of many that they use and abuse to gain popularity. They have also spread lies about vaccines. They promote antisemitism, xenophobia, hate against the LGBTI people, historical revisionism (e.g. they say nice things about the Nazis), etc. They are even fighting against the wind turbines (or, as Donald Trump calls them, the "wind mills"). You cannot satisfy their demands by "fixing" one issue. Because they will always find another. The more absurd one, the better, it seems. I think that the reason is that those absurd causes are supported by people who are easy to manipulate. They know it. And they abuse it.

> Why wait for these parties to use this issue to fuel their growth?

No one wants to wait. But at the same time, I have not yet seen a successful recipe on how to root out their populist practices and render them ineffective against the general population at large. Because, unfortunately, there is still a considerable number of people who are falling for it. And it seems to me that no politician or party so far has been able to figure out a reliable way to protect those who are most vulnerable from being manipulated this way.

I assume that these populist, extremist or dictator-willing-to-be politicians, not only in Germany but also in other European countries, in Russia, in the US, in Canada, as well as in China and elsewhere, are now all spending countless AI tokens to perfect their psychological manipulation and brainwashing techniques.

So that they can secure more and more fanatic followers who will vote for them no matter what. And so that they can remain in power for as long as possible.

To them, the "illegal immigration issue" is just a vehicle to gain power. Nothing more.

Because, from what I can see, their only and actual goal is to get in power. No matter the cost. Then abuse it to the benefit of them and their friends. And never ever give it up ever again.

> Now its likely soiled and I have to throw it away.

Could you explain why you think so?

> X got a letter saying they should stop.

I am unfamiliar with these matters and I am also unsure what specifically you are referring to.

But in general, if big companies get away with wrongdoings by receiving a gentle slap on their fingers, figuratively speaking, it will, in my opinion, lead to no change whatsoever.

Because if there are no material consequences for wrongdoings, they probably would not care. And there are very few kinds of actually material consequences for big companies. One of them being a really hefty fine.

Maybe all of them use the same datacenter. Which has just experienced issues.

I think that it would be sufficient to realize that no one is forced to use a specific AI model or model harness that has anti-consumer features built in.

People should just walk away if they see something like that. This is where the actual, practical, consumer rights start. Not with the regulation. But with the customers being determined to stand up for themselves. And not just fold.

There are plenty of good enough models that are open weight and whose use comes with almost no strings attached. And plenty of great harnesses such as various flavours of Pi.

If you just say that you run out of tokens, it does not mean anything about the token quotas themselves being reasonable or not. That depends on how much you use it.

For instance, if you had 10s of agents running all the time, it is not that unexpected that you run out of tokens quickly.

> Spotify is in the process of killing the librespot project

How could they possibly do that? They cannot prevent people or AI from reverse engineering their APIs.

> Are you absolutely sure you would catch any malicious payload in some obscure single-purpose Python script?

Of course not. But I can at least fairly easily review it. And that is the point.

A review cannot be done as easily with a minified JavaScript webpage or a WebAssembly binary that the WebUSB loader tool would use.

Also, with regards to binary Windows blobs that would require installing Windows drivers - yes, I agree that it is considerably worse than WebUSB. But we can always find something worse..

The point I am trying to make is that WebUSB, although being better than Windows binary blobs, is still less transparent, in my opinion, than e.g. a simple Python script.

Edit: To clarify: Whether the user will actually be able to find potential issues during the said review is a separate topic. But I think that the goal should be to enable users to do so. And make it as simple as possible and user-friendly as possible.

WebUSB is deliberately unsupported on Firefox due to potential for abuse, which seems reasonable to me. Therefore, delivering updates to USB devices likes that is, in my opinion, suboptimal because it requires a specific browser feature which not everyone is willing to use.

It is also less transparent and more difficult to review than, e.g., a Python script. Which would do just the same job. While being even more portable.

I use Linux Containers managed by Incus for working with Claude.

I have a dedicated container for that. It can run its own Docker daemon and other system services if needed.

Apart from the Claude login token, it has no SSH keys or other credentials. I push everything I need to it from the local machine. And I pull the Claude generated outputs from it.

Of course, this kind of setup requires a stack which can run or at least be tested without any credentials.

> I'd not accept to be treated like children.

What is your expectation from such a class? What are you hoping to get out of it?

If you merely want a tutor that will help you out navigate a topic you are interested in, with no strings attached, then sure, yes. No need to take exams.

But if the goal is to get a formally recognized degree in some topic, then you will need to accept that there will be oversight over how you understand that topic. Because the institutions granting those degrees need to sign them off. And if they care about their reputation, they would be interested in signing off only on those students, who actually understand the topic.

So there will need to be exams. In the AI era, it is close to impossible to guarantee that a remote exam will accurately evaluate your understanding of a topic. So the exams would need to be carried out in person.

There is nothing children-specific in using such approach. It is merely one of the few still somewhat objective ways to actually test people's understanding of some topic.

EU does not want that.

Only a few politicians, some of whom happen to be currently in power, in the EU or elsewhere, are pushing for those agendas. But fortunately, they do not represent the majority.

> what should be done about open weight bioweapon and cyber-offense capabilities?

In my opinion, the governments should deploy open-weight AI countermeasures. Because it seems to me that it is impossible to efficiently fight AI-powered criminals without AI.

When only AI-restricting regulations would be put in place, the criminals would, in my opinion, just ignore it. We as a society have a difficult time tracking even the illegal gun or drug dealers. I cannot imagine how could one hope to "regulate" something that can be downloaded as a file and run on a computer.

These AI countermeasures should be open because it provides transparency as to whether the countermeasures actually work. Independent testing, tuning, refining or retraining is then possible.

If the closed models were used instead, their provider could at any point in time shut down the entire operation. Or sabotage it under the hood.

The important part is that with the closed, black box, proprietary models, one can never know what they are being served.

No. No restrictions should be put on downloads. The main point is that the distribution channels should always be free.

What should be gated and limited is the commercial use: reselling, relicensing, using for business, etc.

> the optional payment (I promise you nobody's gonna do that)

Such "promises" seem to be based on no facts. Moreover, they contradict my personal experience.

I would be very interested in using that business model. In fact, I have already used it with other types of services. I have paid for software that is open source, free to download and free to use. Multiple times. Simply because I like the software and I wanted to thank the authors for it in some tangible way.

Of course, the music should still be protected against the commercial use or reselling.

The users usually not paying is a problem introduced by those platforms, in my opinion. They make the entire experience too generic for the regular users to care.

It seems to me that the bands or musicians should distribute their music via their own websites. I think that then the users would care much more.

If I was a musician and produced music nowadays, I would not induce regular copyright rights on it. Because, in my opinion, it makes no practical sense. It mostly alienates the users.

I would put the uncompressed flac files of my music directly on my website for everyone to download.

That does not mean that I would not be interested in getting paid. But I would approach it differently. I would charge for broadcasting it on YouTube or Spotify. Or for playing it in venues.

But I would not charge a regular Joe for it. They would be free to download it, play it and redistribute it in any noncommercial way they see fit.

The most important part is this. I would encourage the "buy after you like" model. Everyone is free to listen. And who likes what they hear, is welcome to buy it.

In my opinion, many small bands and solo musicians would benefit from that model. And I think it would also create some goodwill among their fans.

It also would not shut down the mainstream delivery channels. So if someone still wanted to listen to it over Spotify and pay for it that way, they would still be able to do so.