varispeed

with active sanctions against.

Wouldn't that be breaking the law selling to them?

Half of the time it doesn't work as well.

Sounds like something security services would love people to use. Instead of using wrench to extract the password - and distressed person may lose memory, they can just locate the passkey.

I find it difficult to explain how to use password manager to non-IT person. Whatever I say, they say it is not secure. No amount of explanation will change their mind. They prefer to keep their passwords in their physical note book hidden in the safe (yes, they open the safe etc each time they need to log in somewhere when they get logged out).

As someone with ADHD a passkey is something I can lose easily and I don't want my accounts to be tied to any specific device. What if I have to upgrade my laptop tomorrow because one I use got bricked? Sounds like an absolute nightmare.

Password on the other hand I can remember for dozens of services, each very long.

I think you need one more zero.

Never had an issue. Filmed during holidays plenty of times, even under water.

This is the same in the UK too. Governments everywhere are the same. It's just humans motivated by greed and easily corruptible.

Phone

I don't see a use case for these cameras. Phone takes amazing pictures and videos and is always on hand and if I need something more polished, I just get DSLR. Sure DSLR is more expensive, but if I want to do something well, I'd rather go all in.

Both are entirely useless on complex problem, because they have a bias in training data they can only partially detect in their own output. The answers are getting worse and worse as you dive deeper into the topic you are working on. I thought I could sharpen one of the documents I worked on using Opus 4.8 and GPT-5.5 together. Fable 5 and GPT-5.6 completely destroyed it. Not only it is not human readable anymore, but also doesn't make sense.

They should make a phone with completely open bootloader. As censorship and surveillance is going to be implemented across the West, that could be a great differentiator from the brands supporting loss of liberty.

Filament snaps at 1am and then you have to run print again. 10 hours turn into many days potentially.

I watch tokens to see if it goes in right direction. If model goes off the rails, then it is time to stop and adjust prompt.

I found that Claude often has classist bias and produces answers that favour corporations or e.g. regulation that favours big corporations. It often belittles small business in subtle ways. Only apologises when get called out and then does it again.

Won't this require citizen to provide ID and therefore citizens asking difficult questions could become a target for retaliation?

Indeed, gdpr was created for corporations to have legal basis to process and sell data. Before gdpr it was a gray area. It was never about privacy.

Samsung should be fined out of existence for this.

As someone who couldn't afford it, I always felt excluded when developers have not implemented any alternative, like if someone had plain sound blaster clone.

I noticed this works today, but for at last few days when I clicked "continue" it would not do anything. It couldn't even "compact" the conversation, was getting stuck at 95%.

Yes, I meant "to". Thank you.

Part of me hopes that we got back to the late 1990s dedicated websites/forums. That seems to be the Discord model but let's see.

Well, thanks for lobbying, that's got regulated away. One must be mad today to run a forum.

I would say GPT-5.6 Sol is miles better at least in areas like legal work. Fable is still making stupid mistakes and display no understanding of what it reads and writes. Fable often makes logic and grammar mistakes that change the meaning of sentences, so you have to read everything carefully.

Fable is cooked. So many sessions ends with you have reached max token limit (despite having credits), sometimes mid way responding. So you have to start again losing all context you built. Looks like another way to scam people.

Peak vibe code.

Should we believe that OpenAI is not stealing secrets of companies using their models?

You can always buy multiple Macs. I think Apple's great differentiator could be making frontier class models local. I you want more threads, buy more Macs.

I don't want to run any workflows on someone else's computers.

It's good to have option in case your own government turns rogue.

How about "non-professionals"? It could be useful to check device before sending for pre-compliance / compliance checks and save money - that would avoid very expensive iterations.

These apps should be banned or be forced to only show feeds from accounts you follow in chronological order, nothing more. Perhaps with some basic search if you want to discover something new, but cumbersome enough so that you don't spend hours on it.

I know someone he keeps scrolling the minute they wake up. They are behind on chores, life stuff, but keep scrolling.

That wouldn't be too bad in itself if not for sheer amount of misinformation that is being served. Especially health related. Thanks to AI enhanced videos showing parasites everywhere, person now has developed eating disorder and thinks their poor health is due to parasites and not the fact they spend whole day on Instagram. Now they started ordering "supplements" and "courses" those people peddle. I tried to report these accounts, but Meta insist they are no breaching any T&C.

If you don't have RAM you swap, so it is still better to at least ensure swap is a little bit faster.

If safety is extremely important, why there were bugs in the first place? Surely these should have been caught before code would get into main, no?