What's the semantic way to say in HTML "this work was cited in that academic paper"?
— Terence Eden (@edent) June 28, 2020
I know how I cite something in SchemaOrg (see screenshot) - but I want to do the reverse. Like a pingback or webmention.
Is this possible / sensible / useful? pic.twitter.com/MBmdRd970Q
Pingbacks are essentially dead and in personal experience some of the few sites that still support them are in academia, but they’re relatively rare and have horrible UI in the best of times. Webmention is a much better evolutionary extension of the pingback idea and have been rapidly growing since before the spec was standardized by the W3C.
I’ve sketched out how individual academics could use their own websites and publish pre-prints and syndicate them to pre-print servers and even to their final publications while still leveraging Webmentions to allow their journal articles, books, other works, to accept and receive webmentions from other web publications as well as social media platforms that reference them.
I think the Microformats process is probably the best standardized way of doing this with classes and basic HTML and there is a robust offering of parsers that work in a variety of programming languages to help get this going. To my mind the pre-existing h-cite is probably the best route to use along with the well-distributed and oft-used <cite> tag with authorship details easily fitting into the h-card structure.
As an example, if Zeynep were to cite Tessie, then she could write up her citation in basic HTML with a few microformats and include a link to the original paper (with a rel=”canonical” or copies on pre-print servers or other journal repositories with a rel=”alternate” markup). On publishing a standard Webmention would be sent and verified and Tessie could have the option of displaying the citation on her website in something like a “Citation” section. The Post Type Discovery algorithm is reasonably sophisticated enough that I think a “citation” like this could be included in the parsing so as to help automate the way that these are found and displayed while still providing some flexibility to both ends of the transaction.
Ideally all participants would also support sending salmentions so that the online version of the “officially” published paper, say in Nature, that receives citations would forward any mentions back to the canonical version or the pre-print versions.
Since most of the basic citation data is semantic enough in mark up the receiver with parsing should be able to designate any of the thousands of journal citation formats that they like to display any particular flavor on the receiving website, which may be it’s own interesting sub-problem.
Of course those wishing to use schema.org or JSON-LD could include additional markup for those as well as parsing if they liked.
Perhaps I ought to write a longer journal article with a full outline and diagrams to formalize it and catch some of the potential edge cases.
The IndieWeb wiki also has some initial documentation at https://indieweb.org/citation
Thanks for this write-up Chris. It has me itching about salmentions and having webmentions passed along. I really want to have a link to my ReadWriteRespond posts on my Collect site to aid with linking. However, I was concerned about the implication in regards to Webmentions. Reading this post I am thinking that I would just need to put a syndication link on ReadWriteRespond and the webmention will flow back. Now to think about how I implement this.
One of the things I’ve seen hiding around is how Brid.gy does mentions, and I’m guessing it’s for platforms that handle salmentions. In replies to multiple people it includes blank anchor tags with links to the URLs of the people and links in a particular post. This way anything that receives a reply/reaction should also have the link in it for upstream commentary.
I’ll try something like this here with a link to one of your RWR posts, and see if you notice the mention? If you reply back, you’ll likely need to include a similar “empty” link to maintain the chain. Naturally, if it works manually, then it could be something that we could puzzle out automating. Perhaps David Shanske has thought about it, experimented, or has ideas? I know that Matthias Pfefferle has done some work with the Salmon protocol which is loosely related.
I’ve done things like this manually in the past, but only in special cases. It would be nice if it were automated though. The tougher part may be getting bigger publishers to support/implement it.
I can also see some interesting use for it in the journalism space so authors can own a copy of their articles published in other outlets.
(Update, since we’re talking about salmention, I’ve manually upstreamed this response to the original tweet, though Aaron’s response is missing from that thread).
Syndicated copies:
Chris, is this like what I do with the addition of microformats, such as class=”u-in-reply-to”, to links I want to appear as comments on my own site?
Something like that.
Since the comment section of WordPress sites with Webmention set up will send them automatically to any link in the reply, you just need links to all the upstream things to be automatically added to your replies. It would be nice if the plugin would do this automatically for any links on the page as well, but for now you’d have to do it manually.