Kicks Condor

LEECHING AND LINKING IN THE HYPERTEXT KINGDOM

I FIRST DISCOVERED
THE 【TECHS-MECHS】WHO
ARE A CLAN OF SOUTH
OF THE BORDER GUNDAM
BREAKING DOWN
IMMIGRATION FENCES
WITH THEIR
IMPRESSIVE MANOS
MECANICAS

PLUNDER THE ARCHIVES

This page is also on dat.

Reply: Simple Specs

Manton Reece

I’m skeptical of such complicated specs ever being widely adopted. It would take a very popular real-world app to push it forward (like Mastodon did for ActivityPub, which I also think could be much simpler).

Well, we’ve been through this before: XML-RPC vs. REST+JSON. And no one remembers SOAP—which had tremendous backing, but was just so riddled with tags, you would just get lost in the stack.

ActivityPub is just unbelievable—so many layers of JSON-LD, Webfinger, Salmon.

Webmentions seem so much simpler—even though they’re not exactly comparable. However, Microformats are very difficult. They just are messy in practice—don’t you think? I kind of wish we could go with a combination of Webmentions and JSON Feed. But it’s too late to go inventing another spec. 😉

  1. Well the good news is because Webmention is data agnostic, while I agree microformats can be tricky, if a better solution pops up in the real world, we can easily shift to adopt it. As you said, we shouldnt go out and “try to invent a new spec” but if something makes sense and people try it out and it works, then webmention can always adjust. The webmention spec would never need to change, websites would just have to replace mf2 parsers with whatever the new parsing strategy is.
  2. @kicks Agreed, especially about the layers of JSON-LD that (I think) burden these new formats. Remembering SOAP, I searched my blog archives and found this post from 2003. I might have written it differently today, but pretty much still holds up 15 years later.

  3. Reply: Webmentions Agnostic

    Eddie Hinkle

    Well the good news is because Webmention is data agnostic, while I agree microformats can be tricky, if a better solution pops up in the real world, we can easily shift to adopt it. As you said, we shouldnt go out and “try to invent a new spec” but if something makes sense and people try it out and it works, then webmention can always adjust. The webmention spec would never need to change, websites would just have to replace mf2 parsers with whatever the new parsing strategy is.

    In a way, Webmention doesn’t even seem like a spec—for crying out loud, it is just source ➡️ target ➡️ endpoint! So, really, microformats are the substance of Webmentions. I think it would be interesting to see a variation of Webmention which did JSON, perhaps using Activity Streams or something. I almost wonder if this is what Parecki is doing by offering three different forms of JSON on all of his posts.

    It’ll be hard to beat Microformats, though, in that having a single definitive HTML source is amazing—compared to having all of these other source documents floating around: feeds, ActivityPub outboxes and such. So I’m content to stay put. I already get plenty of great stuff out of Webmentions. Like this ping from you, Eddie!

  4. @kicks I think part of the complexity comes from a desire to solve all the problems. I drafted a related post last week after looking over Solid, but it's a little negative... Need to re-read and decide whether to post it.

  5. @manton this thought deserves to be expanded.

    Micro.blog is a great case study in standard adoption driven by solving one small problem at a time.

  6. Haha, yeah you’re right. Source and Target aren’t much of a spec. I think some of the stuff that makes it cross over the line as a spec is “how do you discover someone’s webmention endpoint?”, “how do you verify a webmention’s authenticity?”, etc.

This post accepts webmentions. Do you have the URL to your post?

You may also leave an anonymous comment. All comments are moderated.