8 ms·

I tried them a while ago. Really liked it but put it aside till they implement https://www.w3.org/annotation/ https://www.w3.org/annotation/ as promised. Just c

I tried them a while ago. Really liked it but put it aside till they implement https://www.w3.org/annotation/ https://www.w3.org/annotation/ as promised. Just checked https://h.readthedocs.io/en/latest/api/ 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%3Aopen+sort%3Aupdated-desc+w3c https://github.com/hypothesis/h/issues?q=is%3Aissue+is%3Aope.... The post I took as a promise: https://web.hypothes.is/blog/annotation-is-now-a-web-standard/ https://web.hypothes.is/blog/annotation-is-now-a-web-standar...

smarx007 6y ago
Slightly on a tangent, but https://github.com/hypothesis/h/issues/777 https://github.com/hypothesis/h/issues/777 could be a good target for https://solidproject.org/ https://solidproject.org/ to address. The Web Annotation Vocabulary is defined in RDF, so there should be zero overhead.
- Vinnl 6y ago
  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.)
  - jchook 6y ago
    This seems pretty critical. I imagine the concept of annotations would benefit from "communities" of annotators who have a common interest. Otherwise popular pages may get annotation overload and have a poor news-to-noise ratio.
  - the_duke 6y ago
    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 while, which also felt weird.
    - Vinnl 6y ago
      There is quite a bit of activity happening around the spec, I believe (though I'm not following it too closely), but lots is spread around a number of different repositories [1], and it's not immediately clear which are active. (People are discussing how to improve that [2].) 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 https://github.com/solid/process/blob/master/panels.md [2] https://github.com/solid/process/issues/202 https://github.com/solid/process/issues/202 [3] https://solidproject.org/newsletter https://solidproject.org/newsletter
woah 6y ago
If your use of a piece of software is contingent on something like that, it sounds like you didn’t really have a need for it in the first place
- BiteCode_dev 6y ago
  Rather, OP is probably already using one and don't see the point of migrating to yet another closed system.
  - tedmiston 6y ago
    Hypothesis has an open API - https://h.readthedocs.io/en/latest/api/ https://h.readthedocs.io/en/latest/api/
    - BiteCode_dev 6y ago
      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. Particularly the tech saavy ones that have been burnt in the past and hold data that is precious to them.
      - the_duke 6y ago
        Both the frontend and server seem to be open source. https://github.com/hypothesis/h https://github.com/hypothesis/h https://github.com/hypothesis/client https://github.com/hypothesis/client
        BiteCode_dev 6y ago
        Then I stand corrected. 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.
aeontech 6y ago
A sibling comment points out that it seems to be implemented here: https://github.com/hypothesis/h/blob/master/h/presenters/annotation_jsonld.py https://github.com/hypothesis/h/blob/master/h/presenters/ann... Or is this something different from what you have in mind?

smarx007 6y ago

Slightly on a tangent, but https://github.com/hypothesis/h/issues/777 https://github.com/hypothesis/h/issues/777 could be a good target for https://solidproject.org/ https://solidproject.org/ to address. The Web Annotation Vocabulary is defined in RDF, so there should be zero overhead.

Vinnl 6y ago
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.)
- jchook 6y ago
  This seems pretty critical. I imagine the concept of annotations would benefit from "communities" of annotators who have a common interest. Otherwise popular pages may get annotation overload and have a poor news-to-noise ratio.
- the_duke 6y ago
  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 while, which also felt weird.
  - Vinnl 6y ago
    There is quite a bit of activity happening around the spec, I believe (though I'm not following it too closely), but lots is spread around a number of different repositories [1], and it's not immediately clear which are active. (People are discussing how to improve that [2].) 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 https://github.com/solid/process/blob/master/panels.md [2] https://github.com/solid/process/issues/202 https://github.com/solid/process/issues/202 [3] https://solidproject.org/newsletter https://solidproject.org/newsletter

woah 6y ago

If your use of a piece of software is contingent on something like that, it sounds like you didn’t really have a need for it in the first place

BiteCode_dev 6y ago
Rather, OP is probably already using one and don't see the point of migrating to yet another closed system.
- tedmiston 6y ago
  Hypothesis has an open API - https://h.readthedocs.io/en/latest/api/ https://h.readthedocs.io/en/latest/api/
  - BiteCode_dev 6y ago
    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. Particularly the tech saavy ones that have been burnt in the past and hold data that is precious to them.
    - the_duke 6y ago
      Both the frontend and server seem to be open source. https://github.com/hypothesis/h https://github.com/hypothesis/h https://github.com/hypothesis/client https://github.com/hypothesis/client
      - BiteCode_dev 6y ago
        Then I stand corrected. 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.

aeontech 6y ago

A sibling comment points out that it seems to be implemented here: https://github.com/hypothesis/h/blob/master/h/presenters/annotation_jsonld.py https://github.com/hypothesis/h/blob/master/h/presenters/ann... Or is this something different from what you have in mind?