I'm more pessimistic. I'm wondering: Has there actually been a case where the "company founded around that hacked-up JS implementation" for a non-trivial feature did not happen to be one of the browser vendors themselves?
I used to lurk on the WHATWG mailing list and while the amounts of improvements they did to the web cannot be overstated, their descision-making process always seemed to be opaque to put it mildly: Features brought up by users where often debated back and forth without any clear goal until the spec editor basically said "we'll add it" or "we won't add it" with no clear reasons given. Other features "hit the ground running" that were apparently decided elsewhere and were only presented in the mailing list to discuss implementation details.
(Note: Massive generalization of the project history though, including the part where I and other key project members worked at the browser vendors at various times. Having insiders at the browser vendors does massively help move a standard forward.)
Annotations? Definitely. At least one member of the Hypothesis team came from Genius (formerly Rap Genius) which is the largest annotation service on the web.
For [1] to work, you need to either run the bookmarklet or Chrome extension. This isn't the norm though; most annotated pages can be viewed directly from the genius link, and you can even link directly to a specific annotation, for example [3].
[1]: https://www.bloomberg.com/graphics/2015-paul-ford-what-is-co...
[2]: https://genius.com/summary/www.bloomberg.com%2Fgraphics%2F20...
[3]: http://genius.it/7163911/www.theverge.com/2015/6/24/8836087/...
And if Genius is hosting the page that's annotated, that's one thing. If they're simply proxying it on demand and adding annotations, then I don't see how it violates copyright unless every proxy server violates copyright.
This is a key distinction that separates proxies from caches and archives.
They tried dirty SEO tactics that Google ultimately punished them for, and their staff have made off-color remarks in the past. There's a lot more if you search for it.
http://www.recode.net/2014/5/26/11627246/rap-genius-co-found...
Things at Genius have been pretty controversy-free since 2014.
I spent time trying to figure out if hypothes.is even uses the protocol and data model in their own software. I'm not confident enough in my analysis to say they absolutely aren't, but I'm also not convinced they are really using it. They would probably counter that it wasn't even a standard yet, but I'm not even sure how close their current model is to this.
They did have some of their employees in the W3C working group. Several of the others seemed to be in the "library" field.
When I use the hypothes.is extension, I see queries in my browser's developer console that do not look anything at all like they are querying annotation collections as defined in the spec.
The data returned from these queries is close to the spec but includes many fields that aren't in it. These extra fields seem to be conveying important information for their application.
Edit: I ran one of their annotations through the W3C tests. Five failed tests. Including one of the most important fields:
{
"name": "1:2 Implements **Annotation _id_ key** which has a **single value** that is a **string of format uri** - [model 3.1](https://www.w3.org/TR/annotation-model/#annotations)",
"status": "FAIL",
"message": "assert_true: ERROR: Annotation is missing id key or its value is not a single string of format uri.; expected true got false"
},
Of course, you run the risk that you end up with de-facto standards and everyone having to implement one implementation's extensions (oh, hi WebKit-prefixed properties!).
Correct. But since it has all this "JSON-LD" stuff baked in, you need to also carefully define and publish the extra fields.
By the way, the Genius team is backed by Marc Andreessen. While Genius and Hypothesis aren't formally related, the two annotation products are very similar just for different audiences. Tim Berners-Lee is even featured in the video at the end of the post.
This is kind of a crazy comment to have to make when the two biggest creators of the web browser support annotations. People who've worked on browsers absolutely have a vested interest in the web annotation standard.
https://hypothes.is/blog/a-letter-to-marc-andreessen-and-rap...
Historically, web standards where a committee gets together and decides how a feature is going to look without the buy-in of users or browser vendors have a very poor track record of adoption. The way actually-successful web features get standardized is that users start clamoring for it, which leads someone to build a hacked-up JS implementation of it, which leads to a company founded around that hacked-up JS implementation, which leads to competition, which leads to browser vendors building it into the browser, which leads to an open standard.
Trying to skip steps doesn't seem to work. If you build the feature without users who want it, nobody will use it. If you build the company without the prototype, you won't get a working implementation. If you build it into the browser when there's a dominant monopoly company, people will continue to use the company rather than the browser's version (this is the story of Google vs. IE+Bing & Facebook vs. RSS & semantic web). If you standardize it before it's been adopted by multiple browsers, people will ignore the standard (this is the story of RDF, the semantic web, and countless other W3C features that have fallen into the dustbin of history).
And if any one of those parties are not at the table when the standard is written, they'll ignore the standard anyway.