The result is that a social reader (tested with Monocle) will see the articles endpoint as a valid feed with no entries. I have modified my instance to mark up the article list items as h-entry, the timestamps with dt-published, the links themselves as u-url p-name, and (because my understanding is that this is desirable for feed discovery) the "Notes" and "Articles" links in the header with rel="alternate" type="text/mf2+html" and I'm pleased with the result, but poetry and invoke have me sufficiently baffled that I haven't been able to format, lint, and test.
Hey,
Good catch, I missed the microformats markup in this template, I just pushed a fix: https://git.sr.ht/~tsileo/microblog.pub/commit/404590206879bf52ccfae4efb251bdcc8d96203d
But I don't think the alternate meta tag is needed? I think most reader will assume there's mf2 markup on the HTML page?
Thanks!
I think you're correct that it is not technically needed; the h-feed documentation says to do that to aid in discovering the "feed or feeds" if your home page doesn't contain your primary feed, and the "feed autodiscovery" section had given me the impression that if the home page contained a feed but there were also links on the homepage that went to other feeds, they should be marked that way. On further inspection and reading, I think:
- I may just be straight-up doing it wrong. I think, to the extent it's supposed to be correct, it may be supposed to be for a
linkin thehead, not for ana, for one thing, and what I'd done was mark up the anchors in the header, not the links in the head.- I'm pretty sure the only use-case in which what I thought I was doing would matter would be someone using a reader or browser that actually detected, from looking at the home page, that there were multiple h-feeds available...and in that case I think I should be marking up leads to tags the same way.
I'm probably going to back that part of my changes out for now and ask around.