jaffathecake

6 ms·

1.
Controlling when CSS custom properties are computed(jakearchibald.com)
5 points by jaffathecake 21d ago|0 comments
2.
by jaffathecake 22d ago
I don't think this is going to work out. With the way progressive loading works in JXL, you don't really hit "good looking" points aside from at DC resolution.
3.
by jaffathecake 22d ago
I think you missed the point. In cases of poor connectivity, you're not receiving (eg) 200 kB gradually over 10 seconds, it's more likely that you get nothing for 10 seconds then 200 kB lands. This is why progressive rendering doe
4.
by jaffathecake 22d ago
AVIF video does support an alpha channel, which AV1, annoyingly, doesn't.
5.
by jaffathecake 23d ago
Yeah it's the latter. I'm going to look into adding transcoding, but Squoosh is currently set up to pass a bitmap to the encoder rather than the original image, because the original image can be processed first.
6.
by jaffathecake 23d ago
It matters less, but more because data tends to land in 'bursts' when connectivity is poor, so unless the image is massive, you're less likely to see a partial image.
7.
by jaffathecake 23d ago
It often comes down to compat, and developer requests. Personally I find in-browser encoding for images inadequate, because it doesn't expose enough options. When we built https://squoosh.app/ , we exposed the browser e
8.
by jaffathecake 23d ago
Some background on this, and demos of JPEG XL's progressive rendering: https://hacks.mozilla.org/2026/08/intent-to-ship-jpeg-xl/
9.
Firefox Intent to Ship: JPEG XL(hacks.mozilla.org)
25 points by jaffathecake 23d ago|7 comments
10.
by jaffathecake 1mo ago
The solution in the post is 4 lines: <button aria-labelledby="bold-tooltip"> <svg>(bold icon)</svg> </button> <div id="bold-tooltip" popover="hint">B
11.
by jaffathecake 1mo ago
Unstylable, incorrect semantics, doesn't appear on hover, only supports text content.
12.
by jaffathecake 1mo ago
Thank you! I switched from Prism to https://shiki.style/ recently to enable that, and allow VSCode-style wrapping (where it wraps to the indent, rather than the start of the line).
13.
The Goldilocks customizable <select> height(jakearchibald.com)
2 points by jaffathecake 3mo ago|0 comments
14.
by jaffathecake 3mo ago
> So websites can now nag users to allow access to the root of their local disk No, Chrome doesn't allow this. Here's a simple demo: https://output.jsbin.com/kekekac/quiet - note that you can't select
15.
by jaffathecake 5mo ago
> The user agent is the proper channel for the agency Jake is seeking here. Theres nothing preventing the user & their user agent from negotiating what model they use. This isn't how it works. As the developer, you use the syste
16.
by jaffathecake 5mo ago
Yes! It might lie or hallucinate. But also, all browsers claim to be "Mozilla/5.0" in their user agent string. It's a very similar problem.
17.
by jaffathecake 5mo ago
It's fun that I get to be called both "anti-AI" and an "AI shill" by people on the internet depending on the day of the week.
18.
by jaffathecake 5mo ago
Thank you for posting this. On interoperability, time will tell I guess. I've only been working on Firefox for a few months, but general interop issues are way worse than I realised when we worked together at Chrome. Firefox frequently
19.
by jaffathecake 5mo ago
Aww thanks for saying that! I've been doing little videos on https://www.youtube.com/@FirefoxWebDevs (and accounts of the same name, pretty much everywhere). Although they're designed to be short, so they're
20.
by jaffathecake 5mo ago
Agreed. Maybe Google will propose a CSS text formatting property that cannot be used on paragraphs that are critical of the US administration. Like, that sounds daft, but it's not really far from what they're doing here.
21.
by jaffathecake 5mo ago
I think the objection here is unrelated to the love or hate of LLMs. It's about the viability of this particular proposed open web API. I personally use LLMs for coding assistance, and some home automation stuff, but I do not think thi
22.
by jaffathecake 5mo ago
Aww thanks! To be fair I didn't toe the party line when I was at Google (imo). Although, that caused me increasing amount of grief internally, until I left. From what I hear, things have gotten exponentially worse in that regard for fo
23.
by jaffathecake 5mo ago
Sigh, when I posted this, I linked to https://github.com/mozilla/standards-positions/issues/1213#i... (which was posted 11 hours ago). Unfortunately someone changed the link.
24.
by jaffathecake 5mo ago
When I posted this, I linked to the latest statement https://github.com/mozilla/standards-positions/issues/1213#i... , which is the content relevant to the title (the details of our opposition to the API). Unf
25.
by jaffathecake 5mo ago
The Chrome model requires either "16 GB of RAM or more and 4 CPU cores or more" or "Strictly more than 4 GB of VRAM", and "22 GB of free space" (it uses around 4.4GB but it doesn't want to use the remainin
26.
by jaffathecake 5mo ago
The rate of model development is an issue here. Once there are many cross-origin models, it becomes a fingerprinting vector. Also even the small models are many GBs.
27.
Mozilla's opposition to Chrome's Prompt API(github.com)
663 points by jaffathecake 5mo ago|231 comments
28.
by jaffathecake 6mo ago
A lot. My scores: - 0.0028 on my MacBook pro screen - 0.0045 on my Dell monitor - 0.0033 on my Pixel 10 pro And those scores are pretty consistent.
29.
by jaffathecake 6mo ago
The associated deep-dive article is great https://www.keithcirkel.co.uk/too-much-color/
30.
by jaffathecake 7mo ago
fwiw, if you serve your page with: Content-Security-Policy: require-trusted-types-for 'script' …then it blocks you from passing regular strings to the methods that don't sanitize.

More ›