Like others I’m bit worried about the shifting landscape for browser vendors, in terms of the incentives for funding changing. Maintaining and improving a browser engine is a task akin to that of a whole-ass OS. It requires extraordinary funds and I don’t think it’s controversial to say that both WebKit/Safari and Firefox are underfunded

Most of the funding for all three major browsers essentially comes from search businesses and much of the rest comes from sources with similar dynamics

Those dynamics are changing. Regulators have come to seeing much of the browser landscape as anti-competitive. Companies are largely in the process of demolishing web media as an industry in an effort to replace them with chatbots, LLM answers, and closed social media silos.

Now we have a situation where it’s much more cost-effective for tech cos to leverage a cozy relationship with the US administration to shut regulators down, both in the US and globally than it is to either change their platforms to be more consumer-friendly or keep funding open platforms

This means that all of the incentives are lined up to make drastic changes in the browser landscape much more likely over the next 4-5 years

The worst case scenario would be a repeat of the NN4/IE disinvestment duopoly era, a nightmare period after IE had effectively won but before the rise of Firefox where the only browser that wasn’t genuinely outright awful was IE Mac.

Except this time browsers are much more complex software

Obviously none of this needs to happen. Tech cos could demonstrate vision and increase funding as a long term hedge against the also increasing dysfunction of appstores, but that feels unlikely

So, the question on my mind: what’s the best strategy for us small-fry under these circumstances?

And the conclusion I keep coming to is that the sensible thing to do is to shift hard away from client-side JS except where it’s absolutely essential (e.g. for accessibility reasons). This is despite being a guy whose web dev career for the past decade or so has mostly centred on JS

CSS and HTML, even if you assume that they are equally complex as JS, have very different failure modes that make them more resilient. Too much JS is also a big problem in many other ways, such as costs and punishing constrained platforms such as mobile phones

The problem is that the industry does not agree and keeps ratcheting up the average JS payload and keeps hiring based only on skills in React and completely disregarding CSS and HTML

It’s clear what a sensible long-term web dev strategy looks like for most of the industry, but much of that same industry is hell-bent on doing the opposite: build bloated software based mostly on a fragile JS-oriented ecosystem, while simultaneously using LLMs as an excuse to both de-skill the industry and shrink the workforce down to a size that can, effectively, be drowned in a bathtub if they so much as dare to attempt collective bargaining