[jdmichal]

I'll repost a comment by mcphage on why this doesn't work, in general. TL;DR is that common ontologies sound great but don't work in reality, because it leaves no room for value-added services from individual providers. (Recipes would probably be OK, because there's just not a lot of value-add you can really provide. Or, alternatively, I'm just completely ignorant about how value could be added.)

https://news.ycombinator.com/item?id=13214134

"""

I'm not sure this goal is very practical, even in the toy example you used (being able to swap data sources for weather forecasts).

If you can use a common vocabulary to access multiple APIs, that requires that all APIs implement the same feature set. Which means getting the API sources to agree on the features to implement, and how to describe them, and stop them from adding any features on that the others don't have. But of course, they'll all be motivated to add their own features, to distinguish themselves from their competition.

And once a API consumer is using a feature that other API producers don't support, then the consumer is locked into that producer, and the whole shared vocabulary is for naught. And of course the API consumers will be looking for additional features, because those translate into features that they can offer to their customers.

Basically, this requires API producers to work together to hobble their ability to meet their customers' needs, all to make it easier for their customers to drop them for a competing endpoint. So it looks like a net negative for everybody.

"""

[kelnos]

Eh, that's not entirely true.

To still allow for competition, you define a base feature set and representation, and then you allow vendor extensions. You need some sort of standards body that can promote vendor extensions to standardized, supported things. And clients can choose to support whatever (or no) vendor extensions that they want to.

However, I agree with you that it's not very practical, but for different reasons: 1) competitors don't necessarily like to cooperate to that level and 2) it will slow down progress a lot, which is a decently good reason for #1. And 3), which I think is the big one:

Companies doing this stuff really don't want standards if they're the first-mover, because standards necessarily enable competition. If I'm an anti-competitive producer (or even just a producer that doesn't mind competition, but wants to maintain a head start long enough to secure a market position), I don't want to start off with a standard: I want to do my own thing, and get people to adopt it, and then I can lock them in, at least temporarily. If someone comes along later and clones my format, that's fine, but they have to do work to figure it out, and I still own the format, so I'm naturally ahead.

[jdmichal]

> To still allow for competition, you define a base feature set and representation, and then you allow vendor extensions. You need some sort of standards body that can promote vendor extensions to standardized, supported things. And clients can choose to support whatever (or no) vendor extensions that they want to.

Right. The problem is in the first step. The moment a consumer likes a vendor extension and begins relying on it, they are locked in until the standards body gets around to standardizing it. So all this cooperation to pick a standard and maintain it, and consumers still end up locked in because they like certain extensions more than others. And software providers for consumers still have to write individualized support for all the providers to in order to manage all their extensions.

So all these cycles went to building a standard, and where's the actual win? We still have handlers customized to individual providers. We still have consumers choosing to rely on singular providers.

[kelnos]

That's just the tradeoff you make. The lock-in is only temporary until the new feature is standardized. If users like the non-standard feature enough to use it and want it in the standard, then it's a good thing. Otherwise you end up with stagnant crap and no innovation.

[kerkeslager]

You left out users when you said "everybody".

Yes, this model makes it so content creators actually have to create content or their customers will drop them for a better content creator. That sounds great for users.

I'm not sure why I should care that a few user-hostile rent-seeking entities won't have complete control of the internet anymore.

The API extensions causing vendor lock-in complaint is fairly bogus. Features would be driven by the content renderers, not the content creators. It's that very abdication of power that browsers have given to content creators that the system I'm proposing would avoid.