In the footnotes they mention GPT-3.5. Their argument for not testing 4 was because it was paid, and so most users would be using 3.5 - which is already factually incorrect now because the new GPT-4o (which they don’t even mention) is now free. Finally, they didn’t mention GPT-4 Turbo either, which is even better at coding compared to 4.

This is a general tech community, mostly centered around news and end-user technology discussions, so it’s very unlikely you’ll get an answer here. Might want to try asking on Reddit or some dedicated Datto/Connectwise forum.

Considering that predicting the next word from context is the one thing LLMs are really good at, I just don’t understand how none of these developments have found their way into predictive keyboards.

The problem is that LLMs require a considerable amount of computing power to run, unlike the simple markov chain predictions that keyboards use. You could use a cloud-based service like ChatGPT or something, but most people wouldn’t want their keyboards to send all their keystrokes to a remote server… and even if they didn’t know or care, the response time wouldn’t be good enough for real-time predictions.

Now smartphone SoC makers like Qualcomm have started adding NPUs (neural processing units) with their latest chips (such as the SD8 Gen 3, featured in the most recent flagship phones), but it’s going to take a while before devices with NPUs become commonplace, and it’ll take a while for developers to start making/updating apps that can make use of it.

But yeah the good news is that it is coming, it’s only a matter of “when” - I suspect it won’t be long before the likes of SwiftKey start to take advantage of this.

The bypassnro command still works though. Installed 23H2 in a VM yesterday and it worked fine.

Although not the same, this has been going on for about two years now. Jensen Harris, a former MS engineer, criticized the ads as well as the design of the new Start Menu, over here: https://threadreaderapp.com/thread/1564399431545667585.html

Passwords are used in more places than just browsers though. If there wasn’t any need for a dedicated app, why did they bother making one for Windows?

But personally, I dislike Bitwarden as well. I prefer KeepassXC instead, as it works fully offline and I don’t need to depend on a cloud-based provider (or spin up a server). The best part about KeepassXC is that it supports auto-typing credentials, so you don’t need to copy-paste - and it works across a multitude of apps, such as remote desktop / terminal sessions.

all devices

Lies, there’s no Linux app yet. As usual, Proton Inc continues to treat Linux users as third-class citizens, all whilst claiming they care about privacy and security.

Edit: They don’t even have a macOS app yet lol.

It was big among the netsec/sysadmin crowd too, it was the first place you’d hear of 0-days in the wild, or whether a popular site/service was down.

The only Keen I know of (and acknowledge) is the Commander:

Speaking of German privacy tools… Win10privacy is still being updated from the looks of it, so that could be an option too.

Not quite 2025, but: https://lemmy.nz/post/7859245

Nothing beats Linux if you’re want control over your system. As for gaming though, you can check this website to see whether your games are compatible or not: https://www.protondb.com/

And if your games are compatible, then look no further than Bazzite - it’s a gaming-optimised distro that just works out-of-the-box, no need to manually install any drivers and stuff (you can of course use it for non-gaming tasks too).

AFAIK, there won’t ever be a Tiny11

https://archive.org/details/tiny11-2311

Unfortunately Group Policy isn’t bullet proof, Microsoft has a history of sneaking in “features” like this as part of an update, but without any corresponding policy to disable it.

Syncthing, it was made exactly for situations like this. And unlike DropBox etc your data stays local.

Mercury is worth checking out - it’s based on Librewolf but has additional privacy and performance patches.

Promising, but since there’s no mention at all about TDP / power consumption, those Geekbench scores are kinda meaningless.

‏‏‎ ‎

What does surprise me is that people actually pay their own subscription for these platforms. Netflix had a win in profit/revenue. I’m surprised that these people did not go for the Piracy method even though, they complain about:

Most people don’t like change. Maybe they have a habit of switching on Netflix daily during dinner or something. There’s also a big “Netflix” button on their remotes and their TV’s homescreen, which serves as a constant reminder. They probably even have the app on their phones. All of this leads to mental conditioning and addiction, it’s now a part of their daily lives. Humans are a creature of habit, and it’s hard to break out of a routine ingrained over several years.

Piracy could be a option for some people, but it’s still either too technical, or not as convenient, for the average Joe. Sure, there are even websites you could simply go to without installing any app, but most people won’t bother with that - they just want to hit a single button on their remote/TV and watch something, without needing to go to some website, a website which may eventually stop working.

You’ll find that most people would prefer to take the path of least resistance, even if it means paying (more) money. Don’t forget that even pirates may sometimes pay money to make things easier - eg usenet/seedbox/debrid users - and that’s simply because they too would prefer to take the path of least resistance, even if it means being in the ironic situation of paying to watch pirated content. So it’s not too hard to imagine why normal people would just prefer to cough up the extra cash for Netflix and continue with their lives, routines unchanged.