Luke

Somewhat goes to prove the point that perhaps it's time to retire this technology from the web platform. If it's both not interopable and has security issues.

They should have spent an extra 10 years in R&D and they might have had a semi successful business out of it.

For one thing fixed focal length just doesn't work for that kind of device. The verifocal tech should have been finished before they launched a serious consumer product.

If you access the website on desktop you can use a basic user script to bypass the restrictions. I'm probably breaking some kind of law by suggesting this but oh well.

gist.githubusercontent.com/mary-ext/6e2...

RSS in HTML: A Follow-Up Writing about the big beautiful mess that is making things for the world wide web.

blog.jim-nielsen.com/2024/rss-in-...

Or just use www.jsonfeed.org for the feed metadata and use html for a fancy view on it. But +1 to anything that isn't XML.

Mason wrote a new one that compiles libxslt into wasm, which is probably the simplest approach to this nowadays.

Though obviously just not using xslt would be even easier for most 😅

A title card which says "Get up, come on, get down with the WebKit" featuring Alejandro Garcia. Behind it are soundwaves visualized and emanating from Igalia's logo

🎙️ New Episode of Igalia Chats: Get Down with the WebKit

@bkardell.com and @meyerweb.com
chat with Alejandro (Alex) Garcia about the WebKit project and Igalia's @wpewebkit.org port

www.igalia.com/chats/GetDow...

Published on: Sun, 03 Aug 2025 Hitchcock Retrospective: To Catch A Thief <p>Today on Episode 220, the guys watch the final Hitchcock film in their retrospective, To Catch A Thief. Will the final film have the guys going out on a high note or leave a bitter taste in their mouths. Tune in to see who has the early lead and more on this episode of The Cinema Geeks!</p><p>Be Sure to Follow The Hosts on X!</p><p>Kevin<a target="_blank" rel="noopener noreferrer nofollow" href="https://twitter.com/OptimusSolo"><strong> “OptimusSolo”</strong> </a>Thompson and Dan <strong>“</strong><a target="_blank" rel="noopener noreferrer nofollow" href="https://twitter.com/_ComicConcierge?s=20"><strong>The Comic Concierge</strong></a><strong>”</strong> Clark!</p><p>#UNLEASHTHECINEMAGEEKINYOU!!!</p>

In Firefox I see the below rendering. In Chrome Canary it just renders as raw XML but it seems to work in Chrome Stable so that might be a wonky flag on my end.

Fyi there seems to be a bug because I'm seeing lots of raw html tags in the rendered output.

Ah okay that makes sense... Yeah that's not ideal...

How do you mean? I just meant something like if the request has accept text/html serve pretty html (which presumably you could pre-render) else serve the current raw XML? I'm assuming that works but I don't personally use RSS so it's possible I'm missing something?

I'm guessing with a bit of server side logic you could make it actually serve a html version instead of the raw XML right?

If that's the case you could make use of the polyfill that Mason provided in the GitHub issue (or something similar)

Alternatively just serving different html entirely.

Should we remove XSLT from the web platform? #11523

github.com/whatwg/html/...

Weigh in with an emoji or more complex thoughts

Definitely feels like a prime candidate to unship. At the very least it should be obsoleted such that new browser engines don't feel the need to support it just because it's specced.

Meanwhile everyone supported arm macOS pretty quickly (especially relative to windows efforts).

It's a shame windows on arm is so unloved it's an annoying circle of it being a pain to develop for so no one supports it, so in turn no one uses it and there's no point investing in it 😅

The fact languages like kotlin don't even support it shows the issues.

Can't be as bad as Oracle Virtual Box (and just oracle in general)?

My issue is that I can only easily run windows arm, and kotlin doesn't actually support that as a target 😅

Yeah for JS especially I can see it just not coming up often enough, for other languages that aren't as cross platform symmetrical I guess it's probably more applicable. Like for the node runtime itself I guess?

How do people do cross-platform testing of their code nowadays? I know CI can help but if you want to actually get the full GUI running? Is there a good VM setup? Or is it a case of just having a spare machine with that OS? I have a macOS arm64 host.

At least in this case it's just platform sniffing so it's far less brittle. In fact don't bother parsing the UA string at all just access the platform value directly.

As a native speaker, I also can't 😂

The hardest problems in computer science are cache invalidation and printers

Buttons are laid out very weirdly, so it's possibly just shenanigans to do with that.

It's to ask them to move in so you can squeeze past obviously! Because of our tiny houses.

Did they get forced to allow WebAPKs to other browsers yet?

Next.js Web Component Used by some of the world's largest companies, <next-js> enables you to create high-quality web applications with the power of web components....

I'd forgotten, I turned this into a web component to expedite Time To First Error
codepen.io/dbushell/ful...

Is it just the audio API? Or is it any user activation gated API?
There's the JS user activation API that might let you detect if it's a more general issue. It might just be that canvas swallows the interaction so it never switches the flag. Is there a WebKit bug tracker issue?

I'm assuming web compat, someone's probably done some prototype pollution somewhere?

Does this handle the origin binding on the codes, or does it just suggest them in any app and the user can decide?