I’ve wanted @mention/Webmention support on Hypothes.is for a long time. I had URL hacked my way into a solution a while back but never wrote about it.
I was reminded today that one can subscribe to an RSS/ATOM feed of annotations on their site (or any site for that matter) using the feed format https://hypothes.is/stream.rss?wildcard_uri=https://www.example.org/* and replacing the example.org URL with the desired one. Nota bene: the /* at the end makes the query a wildcard to find anything on your site. If you leave it off you’ll only get the annotations on your homepage.
To go a step further, one can also use this scheme to get a feed of @mentions of their Hypothes.is username too. If I’m not mistaken, based on some preliminary tests, this method should work for finding username both with and without the @ being included.
These are a few interesting tidbits for those who are using Hypothes.is not only for the social annotation functionality, but as a social media site or dovetailing it with their own websites and related workflows.
Backlinks in digital gardens, commonplace books, or wikis are just an abstract extension of the accounting concept of double-entry bookkeeping.
I’ve updated my footer so the copyright dates include 2021. I’ve also updated the Webmention button so that it now points at my standalone endpoint for those who may not see or want to use the input box on individual posts. Finally I modified the text that appears on both the standalone endpoint as well as the individual post boxes on each post so that the same text works to properly describe both cases.
I also spent some time trying to fix my fragmention/fragmentioner, but I’m not quite there yet.
In 1965 Ted Nelson imagined a system of interactive, extendable text where words would be freed from the constraints of paper documents. This hypertext would make documents linkable.
Twenty years later, Tim Berners Lee took inspiration from Nelson's vision, as well as other narratives like Vannevar Bush's Memex, to create the World Wide Web. Hypertext came to life.
I love the layout and the fantastic live UI examples on this page.
There are a few missing pieces for the primacy of some of these ideas. The broader concept of the commonplace book predated Nelson and Bush by centuries and surely informed much (if not all) of their thinking about these ideas. It’s assuredly the case that people already had the ideas either in their heads or written down and the links between them existed only in their minds or to some extent in indices as can be found in the literature—John Locke had a particularly popular index method that was widely circulated.
The other piece I find missing is a more historical and anthropological one which Western culture has wholly discounted until recently. There’s a pattern around the world of indigenous peoples in primarily oral cultures using mnemonic techniques going back at least 40,000 years. Many of these techniques were built into daily life in ways heretofore unimagined in modern Western Culture, but which are a more deeply layered version of transclusion imagined here. In some sense they transcluded almost all of their most important knowledge into their daily lives. The primary difference is that all the information was stored visually and associatively in the minds of people rather than on paper (through literacy) or via computers. The best work I’ve seen on the subject is Lynne Kelly’s Knowledge and Power in Prehistoric Societies: Orality, Memory and the Transmission of Culture which has its own profound thesis and is underpinned by a great deal of archaeologic and anthropologic primary research. Given its density I recommend her short lecture Modern Memory, Ancient Methods which does a reasonable job of scratching the surface of these ideas.
Another fantastic historical precursor of these ideas can be found in ancient Jewish writings like the Mishnah which is often presented as an original, more ancient text surrounded by annotated interpretations which are surrounded by other re-interpretations on the same page. Remi Kalir and Antero Garcia have a good discussion of this in their book Annotation (MIT Press, 2019).
It would create a more layered and nuanced form of hypertext – something we’re exploring in the Digital Gardening movement. We could build accumulative, conversational exchanges with people on the level of the word, sentence, and paragraph, not the entire document. Authors could fix typos, write revisions, and push version updates that propogate across the web the same way we do with software. ❧
The Webmention spec allows for resending notifications and thus subsequent re-parsing and updating of content. This could be a signal sent to any links to the content that it had been updated and allow any translcuded pages to update if they wished.
Annotated on February 09, 2021 at 02:38PM
In this idealised utopia we obviously want to place value on sharing and curation as well as original creation, which means giving a small fraction of the payment to the re-publisher as well.We should note monetisation of all this content is optional. Some websites would allow their content to be transcluded for free, while others might charge hefty fees for a few sentences. If all goes well, we’d expect the majority of content on the web to be either free or priced at reasonable micro-amounts. ❧
This also dramatically misses the idea of how copyright and intellectual property work in many countries with regard to fair use doctrine. For short quotes and excerpts almost anyone anywhere can do this for free already. It’s definitely nice and proper to credit the original, but as a society we already have norms for how to do this.
Annotated on February 09, 2021 at 02:46PM
Transclusion would make this whole scenario quite different. Let’s imagine this again… ❧
Many in the IndieWeb have already prototyped this using some open web standards. It’s embodied in the idea of media fragments and fragmentions, a portmanteau of the words fragment and Webmention.
Should transclusion work both ways, embedding content and letting the source know that I did so? ❧
If one is worried about link rot for transclusion, why not just have a blockquote of the original in excerpt form along with a reference link to the original. Then you’ve got a permanent copy of the original and the link can send a webmention to it as a means of notification?
If the original quoted page changes, it could potentially send a webmention (technically a salmention in function) to all the pages that had previously mentioned it to create updates.
Automatic transclusion can also be more problematic in terms of original useful data being used as a vector of spam, graffiti, or other abuses.
As an example, I can “transclude” a portion of your page onto my own website as a reply context for my comment and syndicate a copy to Hypothes.is. If you’ve got Webmentions on your site, you’ll get a notification.
communities disappearing from self-hosted forums and even Livejournal to places like Tumblr and Twitter, and to a lesser extent Reddit, was a move from spaces we controlled to spaces designed to control us
of course, this happened for good reasons - accessibility first and foremost. it allowed many new communities to form, too. if we want to have control, we need to ensure access too.
phpBB is terrible but it does what it wants to do. same with Discourse, though some design decisions are... odd. mastodon, on the other hand, does _not_ do what it tries to do - build communities around microblogging
what we need is to build federated forum software with two-way syndication - accessible from the Fediverse, from Twitter, from Tumblr, even from IM platforms.
for example - these posts syndicate to Twitter, but comments don't syndicate back, which means I have to maintain a real presence there.
If you like, you could use Brid.gy to get comments and reactions back from Twitter with Webmention support for your site. I’ve outlined some of it for how I’m doing it on WordPress, but the idea is very adaptable for any website out there, and there’s a growing list of pre-existing code one could leverage.
(Hint: this also works for other common social platforms which Bridgy supports. As examples, I’ve got two-way communication set up between my site and Github and Mastodon just to name a few, so I don’t need to actively visit those sites on a regular basis. I pipe most of the content into a social reader like Monocle or Indigenous and reply directly from there.)
Webmention can be used as some of the community glue for things you’ve mentioned in your thread as well. As an example, I can post on my website and syndicate that content to IndieWeb.xyz (using Webmention) where others can discover it (perhaps by category) and interact with it using their own websites. If they have Webmention support as well we can have a site to site conversation that could potentially all be mirrored on IndieWeb.xyz which acts as a conversation and discovery hub.
This ecosystem is slowly growing and flourishing, but we still need work on making it all easier and more accessible as well as helping to guard against potential abuses and bad actors to make things safer for bigger public communities at scale. (I notice you’ve got a great site, that touches on and covers some of these topics like security and identity.)
I built a tool to analyze incoming webmentions. This new side project generates monthly reports to see how and where content is mentioned.
This may be the first version of someone specifically doing analytics of Webmention on their own website. Very cool!
Replied toa post by Mike Rockwell (mike.rockwell.mx)
Is there actually a benefit to showing the Webmention field on your site? Does it actually get used? It feels like all of the Webmentions are automated through the protocol/API, not manually by copy and pasting a link.
I’ve turned it off for now. We’ll see.
There are some sites that have receiving implemented but not sending, and it was primarily meant for that. (This is probably most often people who are using webmention.io as their proxy endpoint by registering and adding a line of code to their header. Something I do with both a TiddlyWiki and MediaWiki installs that don’t have custom software/plugins yet.)
On some sites I follow, I use those boxes about once or twice a month. I use it a bit more frequently on my own site to manually send myself webmentions from other sites that don’t send them, but which I come across either randomly or via refbacks.
@ambrwlsn90 - Thanks for writing that post. A question - how does Webmentions deal with spammers trying to add links to your posts? That was historically the big problem we wound up with both pingbacks and before that trackbacks.
Dan, since you’re in the WordPress space, there are several pieces in place there. Akismet and other anti-spam tools can still be used to filter webmentions just like any other comment/response on your site.
If you moderate your responses on your site, the webmention plugin has an “approve & always allow” function as well as domain allow-listing for people you know and trust.
It also bears saying: there’s also nothing that says you have to display webmentions on your site either, you can use them simply as notifications on your back end.
In my experience, I’ve also seen people strip active links, scripts, etc. out of their received webmentions as a security precaution. I believe that the WordPress suite of IndieWeb plugins does this by default.
If you need/want to go further, you could work on implementing the Vouch extension of Webmention. Any additional ideas or brainstorming you’ve got to help mitigate these sorts of harms is most welcome.
For the record, for Webmention to work as a protocol, it requires a link to your site to actually appear on a public web page–something neither trackback/pingback required and made them even easier/cheaper to game.
As I mentioned last year, I was awarded a Software Sustainability Institute Fellowship to pursue the project of setting up a Cultural Heritage/GLAM data science network. Obviously, the global pandemic has forced a re-think of many plans and this is no exception, so I’m coming back to reflect on it and make sure I’m clear about the core goals so that everything else still moves in the right direction.
So much of the tone of this piece has not only an IndieWeb feel, but also sounds like it could benefit from some of the organizational structure that I’ve seen the IndieWeb employ, particularly small scale events and collaborations I’ve seen at Homebrew Website Clubs. This may help to
“Organise less-formal learning and sharing events to complement the more formal training already available within organisations and the wider sector, including “show and tell” sessions, panel discussions, code cafés, masterclasses, guest speakers, reading/study groups, co-working sessions, …”
How do we enable more positive serendipity & discovery via our websites, between both existing friends & family, and new positive interactions. And how do we raise barriers to spam, harassment, and other unpleasant social media interactions.
It's not just "fixing comments", it's much more than that.
Respectful responses include even simple interactions like likes, bookmarks, and reposts, as well as indirect comments like quote tweets.
A broader user-centric conversation about the problem that Vouch is designed to solve one or some parts thereof.
Photo by John Cameron on Unsplash
Hammersmith, London, UK |
A comment on the fragile state of society today, by artist Tim Fishlock
I have recently implemented #webmentions on this site and they seem to be working rather well but what are they exactly? The splendid @alistapart explains a little bit more.