One company that I can recommend is Withings. They do have an companion app, but a lot of devices do work on their own, and when not, they work with Google Fit or HomeAssistant, though probably due to that fact the products are pricier.

I’d argue most people just aren’t parasitic enough to willingly exploit both their sellers, workers and customers in the scale of how Amazon did and still does.

But do you also sometimes leave out AI for steps the AI often does for you, like the conceptualisation or the implementation? Would it be possible for you to do these steps as efficiently as before the use of AI? Would you be able to spot the mistakes the AI makes in these steps, even months or years along those lines?

The main issue I have with AI being used in tasks is that it deprives you from using logic by applying it to real life scenarios, the thing we excel at. It would be better to use AI in the opposite direction you are currently use it as: develop methods to view the works critically. After all, if there is one thing a lot of people are bad at, it’s thorough critical thinking. We just suck at knowing of all edge cases and how we test for them.

Let the AI come up with unit tests, let it be the one that questions your work, in order to get a better perspective on it.

I haven’t watched the video yet, but I think TADC has unwillingly joined the “kids” content mill, which is probably what might be referenced.

Even Gooseworx dislikes how those content mill channels have abused TADC’s popularity for their own profit while neither she nor Glitch can do much about it.

Funnily enough, Signal has circumvented the issue by marking their chat window as DRM content, making it invisible to Recall.

I’m not the one who you asked, but I’d still give some feedback of my own. Musk as a person is a difficult character. I would even go as far as calling him narcissistic.

• He got thrown out of PayPal for his incessant micromanagement and disruptions to the flow of the company
• he bought himself into Tesla to replace the CEO with himself
• he tends to depict himself as one of the greatest tech geniuses out there, yet often the plans he presents to the public are often poorly thought out and serve no other purpose than to show his “talents”
• when his proposal to build a tiny submarine for the Than Luang cave rescue was shot down and a British diver was chosen instead he resorted to call the diver a “pedo guy”
• his latest attempts in politics, especially concerning DOGE feel completely half baked and, again, how he presents himself in his position feels more like an ego trip than something more reasonable
• he publicly had talks with the controversial German political party “Alternative für Deutschland”, which are currently legally considered “assured right-wing extremists” and have had a history of having Nazis and Nazi sympathisers in their ranks

I generally can’t trust someone who seems to put himself first at everything to handle anything related to security when the role allows him to exploit it for his own gains. And I do not trust someone who supports political groups known for trying to oppress minorities to defend actual rights for free speech.

The question is whether this actually is E2EE, as it’s easy to fake by using a man in the middle attack and hard to prove. The only real way to prove it for sure is to run a third party security audit, like Signal does.

Taking down the old system doesn’t inspire confidence either, as this downtime could easily been used to interrupt old conversations in order to implement a way to decrypt the messages on the servers before passing it on to the actual recipient, as all keys would have to be re-issued.

Honestly, Telegramm always seemed to me a bit shifty since I learnt E2EE for chats was opt-in.

You are forgetting targeted attacks. A blind attack would pretty much not have much of an effect indeed, however if the attacker knows the machine, then it’s easy for the attackers to exploit these vulnerability if left “out in the open”, and cause havoc, possibly create a lot of damages or leech informations pumped into those machines via old Windows installations.

“working in the LLM server farms makes you wish for a nuclear winter.”

I mean, that’s the empiric method. Often theories are easier proven by showing the impossibility of how the inverse of a theory is true, because it is easier to prove a theory via failure to disprove it than to directly prove it. Thus disproving (or failing to disprove) free will is most likely easier than directly proving free will.

Disclaimer: the article only mentions AI, which I interpret as LLM in my statements due to context.

It feels like this article somehow downplays the effects of AI bias, especially considering how many health insurances already play against their customers. Those companies might push for that tech for those very reasons, simply to save money.

However, I am for AI helping with bureaucracy, as long as one can guarantee its accuracy.

I do think it’s worth the money however, especially since it allows you to cutomize your search results by white-/blacklisting sites and making certain sites rank higher or lower based on your direct feedback. Plus, I like their approach to openness and considerations on how to improve searching without bogging down the standard search.

I should have elaborated on it a bit more, my bad.

While it’s true that DDoS is more of an active technology rather than a CYA thing. It does however also act as insurance when it comes to the “blame game”: if your site goes down it’s not your fault but the provider’s fault, meaning you might be able to recoup lost profits through a lawsuit.

Of course the only way to avoid this for the provider is to provide better and stronger systems, which normally would grow homogenous through more customers and/or growing fees for all customers, which would pay for better capacity and stronger protection by itself.

However here we have a client that is a high value target that others might want to take down at all costs. Even if they didn’t sue, a strong enough attack might, alongside naturally expected DDoS on other clients, not only take down this customer’s server, but others as well, which really isn’t something you want, for the reasons stated above. And rapidly increasing security could be not worth it, as it could devolve into an arms race by proxy with a high risk of the customer leaving if you raise their fees to much, leaving you with a system which’s maintenance will now dig into your profits due to a lost big income stream, or make other customers leave if you raise the general fee.

I don’t want TP convince anyone they are not like jerks, but rather highlight why a corporation would do something like this to a (most likely) lucrative client.

I think the main problem is that people try to shoehorn OOP mechanics into everything, leading to code that is hard to understand. Not to mention that this is basically encouraged by companies as well, to look “futuristic”. A great example of this approach going horribly wrong is FizzBuzz Enterprise Edition.

OOP can be great to abstract complex concepts into a more human readable format, especially when it comes to states. But overall it should be used rarely, as it creates a giant code overhead, and only as far as actually needed.

And insurances provide monetary compensation until you become a common liability, too high to be covered by any sort of fee. DDOS protection is just the same. It’s only feasible if it happens rarely, like they usually happen. However if it’s a common occurrence it will just eat up the profits made by the fees and then some, which just is stupid to do in any case.

Linus Torvals talk at the Aalto University. Specifically a segment where he talks about how hard it is to work with Nvidia when it comes to the Linux kernel.