Nice overview of WebGPU. Also does a decent job laying out the history of graphics APIs. With WebGPU bound to be more widespread and available outside of the browsers things will get very interesting.
Indeed, without deciding to put everything in the public domain, the face of the web would have been very different.
I must resist to redesign my blog I guess... In any case, this is a very nice style for content.
There are many more useful codes than are generally used. We shouldn't shy away from using them when it makes sense, it also means the client side must be ready for them. Very often client code makes wrong assumptions on the possible codes.
This is really a huge update. Brings in lots of features which were clearly missing.
Interesting tool... this is generally done with tools where you're captured into a GUI. Moving this to text and static generation opens the door to proper versioning etc.
Nice post explaining the common algorithms used for load balancing. Each having their own trade offs of course. Well done with tiny simulations.
Good reminder that links are the soul of the world wide web!
Looks like a nice reference about WebGPU. Unsurprisingly it covers some 3D basics as well.
This is a big milestone for 3D and computation on GPUs from the browser. I suspect it will have interesting security implications though, we'll see.
This ecosystem suffers from the same warts and doesn't seem to make any progress... lack of transparency, "we know better" mentality, tight coupling, lack of communication. This is especially problematic for something like a browser.
Nice balanced post on the pros and cons of GraphQL.
It's clearly way too reliable. This needs explicit hardening.
Very nice approach to avoid the font bloating on the web. I'm slightly concerned about the maintenance over time but at least it has proper fallbacks and the fonts used seem widespread enough (for now).
A bit of a rant so brace yourselves. Still, it's very much aligned with the current backslash against "everything must be an SPA" trend and makes very good points on how it happened. This indeed turned into a popularity contest based on false premises. Meanwhile... complexity increased dramatically on the web frontend side and the performances are bad for most users.
A bit of a rant, but even though React is well established at that point and here to stay (shake a tree and half a dozen React developers fall from it), it doesn't mean it can't be criticized. It does a good job at listing the main ergonomics problems React is suffering from. The funny part is towards the end, the envisioned solutions for another framework look eerily familiar to a Qt developer, it talks about signals and what looks like property changed notifications. :-)
Nice demonstration that web frontend can and should be organized like a regular GUI application (like a desktop application for instance). This will bring the same benefits in term of maintainability and modularity.
Looks like a nice editor to use in web frontends.
The article is a bit confused about what is really about SPAs in general or React in particular. Still it is clear they raise questions regarding accessibility and navigation. In turn, this requires a lot of careful implementation to make sure the user experience is acceptable.
This WebGPU framework is getting interesting. Definitely something to keep an eye on and evaluate for productive uses. Obviously requires WebGPU to be widely available before banking on it.