Reading List #43


Last Monday, we finally launched a client project which we were working on non-stop for the past few weeks, which was great but also left me a bit un-motivated for anything else the rest of the week. Looking forward to get back on track next week.

There were quite a few articles about various JS frameworks, especially React, and about Single Page Applications in general and how they can be hurtful to the end user experience. It looks like we’re finally getting back to using native Web features and progressive enhancement instead of the JS-laden status quo of today, which is a direction I very much appreciate. I love the idea of looking at frameworks as polyfills, which can be useful for a while, until the web and browsers evolve and render them useless. Kind of like jQuery was useful in the past, but isn’t necessary anymore for most of the things we were using it for, since browsers are much more capable today.

Frontend Development

๐Ÿ”ฅ On performance, mental health and the harm done by excessive frameworks

If you read only one article this week, please make it this one. It sums up so much of what is wrong with the current state of “Frontend Development”. Eric uses an experience registering with a mental health service here, which completely failed because of some bonkers framework decisions. There are many quote-worthy lines in there, but this one hit me square in the face:

What if I was suicidal?

Eric Bailey โ€“ Modern Health, frameworks, performance, and harm

Of course, this question is especially important, if someone is looking for and wants to register for mental health support. But I find it is also a great question to ask when designing anything. Imagine someone not able to access their bank account because your stack fails like this? Someone having to fill out a form for any important service, lets think food stamps, job-center application, taxes or what have you and then being denied access because of the shit-show going on in the back? And now, on top of that, imagine that person being suicidal at this very moment already. It is a very good question to keep in mind.

Eric Bailey โ€“ Modern Health, frameworks, performance, and harm

๐Ÿ›ด Speed for who?

Another brilliant article about how wrong it is that modern web frameworks prioritize the developers’ over the users’ experience. The top priority of all our work should always be the people using it. Period. Doing it any other way is doing it wrong, if you ask me. And also, as others mentioned in the comments, this “priority on DX” isn’t even making the development faster, which would have been the whole point of it. But before I take away too much, go read what Andy Bell has to say about it. I couldn’t agree more.

Andy Bell โ€“ Speed for who?

๐Ÿ”ง Disentangling Frameworks

Michelle Barker with a post about the considerations one should make, when choosing a framework. And why it’s almost always best to choose features provided by the native web platform instead of relying on framework X. She uses tailwind as an example here, but it applies to other framework choices as well. Another great read.

CSS In Real Life โ€“ Disentangling Frameworks

๐Ÿ–‹ Container Queries and Typography

This is an amazing example of how we can set relationships between font-size, line-height and measure with modern CSS and make lots and lots of the hacks we used before obsolete.

Robin Rendle โ€“ Container Queries and Typography


๐Ÿ‘ฉโ€๐Ÿ’ป How ChatGPT employed Kenyan Workers for Less Than $2/hour to classify content

Systems like ChatGPT are certainly impressive, but let’s not forget the human labour that is necessary to make them safe. OpenAI, the company behind ChatGPT, seems to have worked with a Kenyan firm (which also did outsourced content moderation for facebook in the past) with workers earning less than 2$ an hour to classify disturbing content.

Time Magazine โ€“ Exclusive: OpenAI Used Kenyan Workers on Less Than $2 Per Hour to Make ChatGPT Less Toxic


Made with โค๏ธ in Switzerland