Terrible article. Don't waste your time.

Design trends coming and going isn’t really CSS’s fault.

CSS evolved a lot over the years, not sure how you could argue otherwise. 20 years ago we didn't have flexbox or grid or variables. Those are all huge for the impact they had on creating responsive layouts. Not to mention things built on top of CSS like CSS modules and LESS.

I also disagree that CSS is poorly documented. There's great documentation for it on MDN, but much of it won't make sense unless you invest time in understanding the browser rendering pipeline and how layering works. And you need a pretty deep understanding of to really master CSS.

Most devs don't bother because you can get CSS to be "good enough" even if it breaks in some edge cases. But that isn't very different from other forms of style sheet languages or real programming languages in general.

I wonder what we would use if we could start again with all we learned the last 3 decades.

Real enough language, just a fucking awful one to my first language C brain, I enjoy it as much as forcing Word to format somewhat sensibly (i.e. I never invested the time to get good). I'll use it the bare minimum to get a job done that it does best. Then again I feel the same way about javascript, but Python is fine. However, if you're gendering code you're at best a bloody idiot and probably a misogynistic asshole.

Thank you!

https://archive.is/jhm1S

Paywalled. If anyone has access, I’d appreciate a quick summary.

@codeinabox Biais de confirmation, mais oui je partage : typescript est mort.

Fair comparison, and I can see why you’d make it from the 100,000,000 target.

The idea isn’t really to recreate the Million Dollar Homepage though — there are no pixels or advertising spaces being sold. Each person contributes one photo and one message, with one qualifying contribution per person/account, and the collection is meant to be browsed as people, photographs and messages rather than as ad inventory.

The scaling/UX side matters to me precisely because the collection itself is the product rather than a one-off page that eventually fills up.

That said, your reaction is useful. If your immediate impression was “another Million Dollar Homepage”, then I may not be communicating the distinction clearly enough. Out of curiosity, was that impression mainly from the 100,000,000 concept itself, or from how the homepage explains it?

People have been trying to reinvent the million dollar homepage for the last 20 years, to no avail.

Realistically, that era of the internet is over and I can’t say this brings anything novel enough to break through. With that said, scaling and UX are kinda moot points.

But how is it related to "Web Development"?

The eternal pendulum swinging between BEM and atomic css forever. The best is to start with atomic and then extract a few BEM use cases as duplication emerges. Then you've got the best of both worlds.

Roughly half of css is extremely specific to one place and not worth extracting. The remainder though is actually important to the design system and is worth deduplicating and formalizing into BEM / variants. The trouble is when you try to BEM everything and end up with thousands of classes that are only used once.

Forever watching people reinvent CSS classes for no reason. (I appreciate that this article actually does show doing it with classes, but it's still grating to me that these "frameworks" just keep reinventing style="...")

This speaker actually has some fairly good articles.

I asked them if there's a recording of their talk. They might reply here: https://front-end.social/@spartanatreyu@mastodon.social/117290599721612680

Or it may be uploaded on the conference's channel here: https://www.youtube.com/@stripecon/videos

We tried something like this at work, but users get confused compared to a more traditional tokenized filter UI, where filters are delineated better and each filter has an x next to it to remove just that filter. People aren't used to having multiple filters in one search field, unless it's just doing fulltext search across everything.

I know Github does this, but Github users are a small subset of the population.

What was analysed#

Using Airline Website Speed Benchmark data, I studied 111 of the world’s top airlines — pulling real-user Core Web Vitals (CWV) data from the Chrome UX Report, cross referenced with if they passed Google’s site speed ranking assessment, Skytrax service rankings, and the technology stack behind each site.

It’s altered or had a couple of passes, but this is pretty boilerplate in LLM research efforts these days.

The image is fine, the content is over the top detailed and misses the core issue which is the backend.

The points may be valid that there is room for improvement on the frontend, but the backend can’t be made more responsive without significant backend improvements, which is very risky because some of it literally skipped an entire generation of developers (meaning the ones who retired making this, didn’t hand off knowledge to the next generation).

You heard right on that conference, they were first and then they fell asleep at the wheel.

Transport and waypoint routing is one of the most challenging processes to get right in software.

Could you give an example of what makes this an AI article? If you're referring to the image, that is credited to the Illustrator Jeffrey Phillips.

As for airline backend infrastructure, I remember attending a great talk from Skyscanner, explaining how airlines were ahead of the curve getting computers but the problem is they haven't kept them modern. That however, is not an excuse for poor front-end performance that the article discusses.

AI article.

From my experience one of the core reasons they are slow is because the backends are still IBM, and every web query needs to be translated a number of times, often web <> java <> IBM/TPF, which is what bottlenecks each request.

I run my browser in a microVM so that 1. security issues shouldn't affect the rest of my system, and 2. if I need to restart, I can save the state to disk and resume it later.

Not sure how feasible this would be if you're on Wayland, but I use X11 so it's a simple matter of pointing the VM to my existing Xorg server port on the host.

Are you hosting on your own machine or server? If so, you can add some buttons that are hidden such that only bots can see them and ban anything that clicks them.

I'm not sure personally how you'd manage it with html+css with no client side scripting, but my theory is to just move your 'I'm a bot' indicator button way off screen with css, or render it invisible and super small so it's unlikely a human could ever find it.

If they are styled dive that listen to a click event with JavaScript, that could also prevent them from "clicking" it.

You have buttons on your website, which write data somewhere that matters, without any authentication shielding them from being clicked by unexpected bots/people? If no authentication, this is bound to happen bot or otherwise.

Good call they’re styled links, I will give that a try!