I was confused also, I use my Yubikey a passkey, so I did not get the rant.
Yubikeys can be used as a device-bound passkey, that the FIDO2 thing.
Now software managed passkeys are also a thing and can be stored in a (good?) old password manager, of course this is less secure and could be disallowed in the future.
Thus the rant is about people without dedicated hardware keys that would have to use things like device-bound passkeys their OS instead of their password manager.
Yes it's very common on sites, and even somewhat forced to have these digital passkeys. It's almost automatic setup and there's a lot of skepticism about them.
Yeah, I mean currently a lot of people still prefer using 5.6 Sol over Astra for implementation tasks. If 6.1 hasn't fixed output alignment (vs task completion) this is a huge price increase for those wanting to keep using 5.6 Sol...
It was the case in France too, it was called Bloctel; you just had to register your phone number and telemarketing companies had a legal obligation to look it up and not call you.
So it worked for a time, legit companies stoped calling; however most people didn't knew about Bloctel as it was little advertised.
Meanwhile, spam calls turned into scam calls, and a real issue especially with the elderly.
So I think that banning all unsolicited calling kills two birds with one stone, spam becomes illegal and scams as a consequence becomes easier to spot as there are no more unsolicited calls.
We have that kind of list in Italy too. I registered my number and I keep getting calls, so I stopped answering to numbers not in my contacts list. If I know that somebody will be calling me, I ask for the number and I add it to the list.
Have you tried answering and then just saying the name of the list you've enrolled in? Tends to help propagate to the companies that they absolutely should not call the number again. At least here the companies tend to actively self-moderate and cross-reference against the lists, as the law does get enforced. When I worked in cold-calling call centers in the past, I remember they being strict with teaching the "marketers" what to do when people brought up the local list(s).
Yes it’s very codebase dependent, and changes over model versions. In my case codex was so bad, until it suddenly became better than Opus at 5.5, which surprised me. I am sure it can be the inverse for some codebases.
The way I keep an eye on model performance is alternating between models for code review. It shows how much the model understands the codebase without disrupting my workflow.
One could make the case that being SotA in 2026 is very costly and not that important for being SotA in 2030 if much more efficient models indeed happens.
That's an interesting take, however there is no ongoing maintenance related to local models, maybe the only effort is giving more capable machines to the workforce; but yeah I can see how it might feel like a barrier.
The hardware, the power systems, the cooling systems. They need maintenance.
The OS needs updates, file systems get corrupted.
Fans get dirty.
All the things that you need to deal with in hosting your own server infrastructure you have to deal with when hosting your own AI infrastructure (which runs on servers...)
However, you can get many of the benefits of a "local model" by outsourcing all the hardware maintenance but still using an open model. Guaranteed repeatability for one.
A lot of the reason people outsource normal software is its brittle security properties, not sure that even applies to an LLM - it can go and look up the latest security best practices just like an engineer can.
In my experience performance is about the same as native on mobile. On desktop I cannot compare as I thankfully never had to make cross platform desktop apps using native platform SDKs, but Flutter is doing fine. I am a working on a non trivial desktop app, and I am pretty happy about it.
Hopefully the desktop story is going improve as Canonical is now leading the Flutter desktop side.
I'm working on an app that's cross-platform on mobile devices. I do builds for macOS as a convenience for my colleagues demoing the app. Pretty much zero effort to add that as a target. I get some benefits out of it too in the form of being able to see how well the UI responds to unusual screen dimensions, running in a resizable window.
To me the issue is that ad vendors in the app knows what you read, the NYT have over 300 vendors listed in their privacy section, the Economist about 80. But if you refuse in the NYT they nag you over and over again for accepting cookies.
It’s crazy not to have privacy reading a magazine.
Having worked with domain experts I concur on the difficult time they have expressing the rules of their domains.
Once I built a little domain-specific language for them, that was tested against old jobs to see if they contradicted the past; it was a nifty project and since then I am convinced that DSLs are underrated as a way to encode expertise.
I had the experts write markdown files that contained the rules looked somewhat like:
## 1A Rule name
Some prose explaining the rules liking to official documentation.
``` if municipality and inhabitants > 10000 then functionA else functionB ```
Then a trivial parser would extract the rules, the DSL was then handled by Lark[1]. So pretty simple, but it made collaborating with experts easier as simulated results would also output some markdown they could read.
Sounds like "Literate Programming", where the code and comments are reversed: instead of everything being code except what's marked as a comment, everything is a comment except what's marked as code.
Yubikeys can be used as a device-bound passkey, that the FIDO2 thing.
Now software managed passkeys are also a thing and can be stored in a (good?) old password manager, of course this is less secure and could be disallowed in the future.
Thus the rant is about people without dedicated hardware keys that would have to use things like device-bound passkeys their OS instead of their password manager.
reply