I'll be disappointed if this isn't named AbstractAiInfrastructureFactory

These kind of complaints get tiresome, here for some Objective-C fun,

https://github.com/Quotation/LongestCocoa

AbstractAiInfrastructureComponentInitializationAndConfigurationFactory

Ofcom knows it's barmy. They are pursuing foreign sites because a) they are required to by law, and b) they are under political pressure to do so. But I suspect they are quite happy that it has backfired, to help relieve some of that pressure and bring into the open the barminess of the OSA enterprise.

This. The regulation organisations can sometimes try to make a point when they're tasked with enforcing an unenforceable or otherwise unworkable rule.

Be careful what you wish for. They may start learning to code.

If that will happen, they may end up as the only species which knows how to code

I would gladly teach them.

Nice work. This is a good approach IMHO, and (if you need validation) also the one adopted by Templ in Go, which has a 'generate' command to do this.

Odd that the FFI list didn't include Java given the JVM target. Typo or oversight?

I believe it's just not FFI towards Java byte code, since it's the "native".

But of course as a Truffle language it gets "free" java interop

Java FFI can be done, it is just on the messier end of the spectrum. I do a little bit of it as needed, but I've focused in on Polyglot languages for now, and deferred a lot of the hooks for nicely handling it to later work.

Talking about Java FFI invites you to say 'well then smart-ass how do you extend this Java class/interface over here with code written in your language' and frankly, that has never had a satisfying answer for any language that isn't Java.

But today you can pass Haskell Data.Text out through the FFI binding framework and it shows up on the other side as a valid TruffleString, in UTF-8 with zero copy semantics and if any Truffle language passes in a TruffleString that is UTF-8 encoded it gets unwrapped into Data.Text using the same.

That and some array support is pretty much enough to talk to everybody else.

X is the same. "Check out what [politician/celeb] said on X!!" And it's the login page.

The answer is not to sign up to X, but to learn to suppress your FOMO. Chances are it was drivel anyway. Especially true for LinkedIn.

Just replace x.com with nitter.cf

nitter.cf is a controversial one because (a) it's not real Nitter - it's an AI translation into Rust and (b) - it's run by a person who also runs a child porn site

You can use twiiit.com or xxcancel.com to redirect you to any nitter that's currently working.

I see. Had no idea. Thanks for the alternate links. More the merrier.

that's neat thanks

Can't escape the feeling that the original vulnerability was deliberately planted as a mechanism for law enforcement (or maybe criminal orgs) to "expedite" investigations.

She wasn't in power very long though.

And then Elizabeth had the example of Bloody Mary to avoid (not least of which: to not marry is better than to marry badly). "I have no desire to make windows into men's souls" is one of the wiser things any English monarch has ever said.

Reading articles in advance of commenting usually helps

Good article.

One of the great things about having an open platform like the Xteink is that it should uncover loads of novel and unsatisfied use cases. Eg, this could be paired with a mobile phone to act as a secondary display much better suited to outdoor applications in bright sunlight.

> open platform like the Xteink is that it should uncover loads of novel and unsatisfied use cases

+1.

And it's worth celebrating the underlying technologies that enable this: ESP32, and RISC-V.

Anyone know what's needed for eink to become cheaper?

working on an daily life ePub generator (because I want to stick to CrossPoint) with the opposite goal, namely use the phone as little as possible, but I admit I would also like to tinker with an additional temporary display, at least for typing