71–80 of 84 posts
Love this but please for the love of everything holy make a firefox extension. The bookmarklet works but you only see annotations if I click on the link. It's sad and clear chrome is the new IE and mozilla is again the underdog.
Earlier quoted context omitted.
Hypothesis has an open API - https://h.readthedocs.io/en/latest/api/
It's not the matter. If you use a saas, unless it's open source and you can self host, you are at the mercy of it. If the service is good enough and the team behind it seems in good faith, it can give you confidence for the switch. But promising to adopt a standard then not doing it doesn't inspire confidence. The system being closed source (no open API can change that), they need to build trust with some users. Part…
Earlier quoted context omitted.
Hypothesis has an open API - https://h.readthedocs.io/en/latest/api/
It's not the matter. If you use a saas, unless it's open source and you can self host, you are at the mercy of it. If the service is good enough and the team behind it seems in good faith, it can give you confidence for the switch. But promising to adopt a standard then not doing it doesn't inspire confidence. The system being closed source (no open API can change that), they need to build trust with some users. Part…
For those that don't, it was a locally running proxy service that would inject its own annotation UI into any site you wanted. Basically your own pocket comment/discussion board where you could discuss whatever URL you were on with other users of hoodwink.
https://github.com/whymirror/hoodwinkd
Also, wow, I can't believe that was 14 years ago. I miss _why.
I tried them a while ago. Really liked it but put it aside till they implement https://www.w3.org/annotation/ as promised. Just checked https://h.readthedocs.io/en/latest/api/ and the v2 API in development has nothing to do with the W3C standard. UPD: there is not a single issue open on the main repo wrt W3C standard implementation: https://github.com/hypothesis/h/issues?q=is%3Aissue+is%3Aope... . The post I took as…
Or is this something different from what you have in mind?
Earlier quoted context omitted.
The person I was responding to specifically mentioned "collaborative annotation" and startups. Non-collaborative annotation has different tradeoffs. It may be useful to you, but I can't imagine there's any money there for a startup. It avoids the anti-network effects that kill annotations, and it avoids publisher objections too, but it also doesn't have a very compelling value proposition for very many people. It's l…
Why is there no money in a personal annotation system? I'd think any large organization would be willing to pay a lot of money to use a tool that would make its members more productive and effective. In fact, Coda and Notion.so are very successful products and they're basically personal/private annotation systems (modulo storing everything in the cloud). I can imagine extending Coda and Notion.so with programmatic ca…
Love this but please for the love of everything holy make a firefox extension. The bookmarklet works but you only see annotations if I click on the link. It's sad and clear chrome is the new IE and mozilla is again the underdog.
It's so insulting they have an open issue and said they're working on it for more than 2 years when Firefox extensions framework is 99% compatible with Chrome extensions. Probably just have to repackage and publish, but apparently they can't be bothered.
The idea of collaborative annotation has been around since the beginning of the web. I tried to launch an annotation startup myself back in 2000, and there were several funded competitors back then. It never catches on. It would be interesting to analyze why people just don't use these systems when given the opportunity. Is this a solution in search of a problem?
I think generalized annotation has the following problem: Draw a graph of all web pages ordered by how many times they are visited. You get a power law distribution. On the left are things like CNN's front page, on the right things like a text file containing a cheesecake recipe someone posted in 1995 that is technically still online but nobody has visited in years. On the left side are pages that are visited too oft…
This is causing me to realize that the value of Wikipedia is not the fact that the pages are editable by the public; it's that there exists a process (as flawed and controversial as it is) for debating over the various contributions and edits that leads to a single, unified version of any given page. More than anything else, that consolidation process is Wikipedia. A successful consolidation process has to be present in order for any system of public contributions (annotations or otherwise) to produce a coherent result.
Earlier quoted context omitted.
It's not the matter. If you use a saas, unless it's open source and you can self host, you are at the mercy of it. If the service is good enough and the team behind it seems in good faith, it can give you confidence for the switch. But promising to adopt a standard then not doing it doesn't inspire confidence. The system being closed source (no open API can change that), they need to build trust with some users. Part…
Both the frontend and server seem to be open source. https://github.com/hypothesis/h https://github.com/hypothesis/client
If it's open source, and it's just a matter of following a standard, I don't think it's a reason for not giving it a try.
Earlier quoted context omitted.
I'm not entirely up-to-date on the annotation spec, but if all that's needed is to host the annotations somewhere, then Solid should already support that. It's just not really clear to me how a user should find out that there are annotations about a page available somewhere. (I work with Solid.)
Do you have views on the state and direction of Solid? I check up on it every few months, and it seems like nothing has really happened since the announcement years ago. The spec is mostly a stub , with no activity on Github. There still seems to be only the node-solid-server implementation, which also sees almost no work. Inrupt seems to have tried linking with/appropriating the project for the company brand for a w…
In terms of servers: there has indeed not been too much activity there recently, which relates to efforts having mostly focused on stabilising the spec (as far as I've seen, at least) rather than new things having been added.
It's undeniable that Inrupt and the project are very intertwined, given how many of the original project members are working for/with Inrupt, and it being founded by TimBL. I think both the community and Inrupt are still figuring out the best way to work together. And I think we've had the reverse problem as well to some extent, where the community might be making themselves a bit too dependent on Inrupt. I think the project can grow in two (complimentary) ways: top-down, with large organisations getting the project adopted at scale in big bangs, and bottom-up, where early adopters encourage additional adoption and it spreading from there. Inrupt is probably best positioned to support the former, but we haven't seen too much from the latter yet. That said, there are a number of community projects that aim to make a dent there, and the best way to keep up with those is through the This Month in Solid newsletter [3].
(Note: these are my personal views.)
[1] https://github.com/solid/process/blob/master/panels.md