> The reason Google and Facebook dominate the internet has little to do with technology and much to do with economics, namely scale economies and network effects.

And the network effects can only happen because there is a tight coupling between the data (contained and scrambled within the HTML page) and the presentation. Had data and presentation be decoupled, I could have combined network information from various providers and present it to me in one overview.

> I'll grant that CSS is a mess. What else?

Everything. The security model, the fact that the user has limited control over their identity (and it is not build into the protocols), the multi-threaded nightmare, no component standardization, very limited integration with low-level OS systems, the 'one-tab is one context' metaphor, the box-model, the misfit of page-to-interaction (and the history API), the non-existent voice interaction, the color-model, the DPI/pixels/ems transformations, the invisible local storage we barely manage, the resource utilisation, the lack of uniformity in the programming model, the peer-to-peer communication (WebRTC does not count, since it is not standardized and has quite some flaws), the lack of (easy) customization abilities (high-contrast websites), the limited math-rendering, the enourmous complexity... I could go on.

> You can create multiple URLs that point to the same object.

Yes, but they would not denote equivalence. I could use a redirect, but then one system needs to know the representation of the other. It would be fundamentally asymmetric.

Also, it would contradict the term Unique Resource Locator.

> no component standardization

This one is interesting because Alan Kay is so expressly against it ("browsers shouldn't have features"). He thinks every object should just render itself as a bitmap.

Besides, browsers do have standard components. That's what the semantic HTML tags like <button /> are. The specific views vary between browsers. Is that a problem? Regardless, almost no site uses the standard stylings that do exist. They use a de facto standard like Bootstrap, or they build their own.

I agree with some of your other points, but I don't understand your lament of "enormous complexity" while you're advocating for more features.

One thing about rendering to a bitmap is that it is more robust over time. Look how much code is in the browser to support older styles of html. Smart endpoints dumb pipes, prevents ossification. The browser tries to be far too smart for its own good and ends up making a mediocre job of everything while being hugely complex (as outlined by parent above)

I guess he meant: browsers shouldn’t need features. It should be a network effect. It’s good to make a distinction between design principles and their resulting effects. Most of what I described before were effects of bad design.

> And the network effects can only happen because there is a tight coupling between the data (contained and scrambled within the HTML page) and the presentation. Had data and presentation be decoupled, I could have combined network information from various providers and present it to me in one overview.

I'm with you in principle. I think such a modular architecture based on standard interfaces between objects is exactly what we need, but I don't see how it precludes e.g. the dominance of Google. Google dominates because they have all the search data. People don't want non-Google results in their views. Non-Google results are worse.

> Also, it would contradict the term Unique Resource Locator.

wikipedia defines URL as "Uniform Resource Locator", and i don't think i've ever heard "unique" in there

Yes, it was rather late yesterday. It is actually the dichotomy that a URL should be a unique resource identifier for a specific resource, but unfortunately it isn't. When we place a link, we want to be able to query on that specific URL (or URN, as it has become). We want to do late binding, since it is cheaper and more powerful.

what benefits do you see in URLs being unique? not necessarily disagreeing, just curious

- distribute resources at multiple locations

- reverse index to resources

- Easier to move resources

- Cryptographic signatures on id’s

> Had data and presentation be decoupled

"What if "data" is a really bad idea?" -- Alan Kay (https://news.ycombinator.com/item?id=11945722)

I just don't see how he has ever pushed for the direction you're advocating. If anything, it's the opposite.

As usual with Alan Kay, it's difficult to understand exactly what he's advocating for. On the one hand, he laments browsers having features, and thinks they should just execute code safely and provide a low level graphics interface, but on the other hand, he advocates for a universal polymorphic graphical object mesh (perhaps approximated by LivelyWeb) which would require a great deal more "features" (think Java) and is closer to what OP is describing.

I think he's made his way into this thread, so I invite him to clarify!

I read it as the browser should just be a VM. It's low level. Then the objects that run in the VM provide the features. This solves a lot of problems. The browser is slowly evolving to do this anyway with js, then wasm (reality is proving Kay right) but the underlying high levelness of it makes it a mess. A VM is more flexible and robust and at the same time simpler than a browser

The difference is that, in order to foster composition, the platform can't just be a VM. It must also provide the interfaces by which objects interact. Take Java. It's not just a VM. It's also the JDK/JCL.