This blog, or rather the (b)log-In system, is a bit of an outlier in that posts do not have explicit URLs. There's not a page you can visit to view a single, targeted post on its own.
While this is not an issue for normal matters, it does seem to cause problems when sending webmentions to certain endpoints. These endpoints appear to ignore the URL fragment that denotes the post, #p3 for example, and just reference the first h-entry on the page, therefore always pulling from the first post of the day.
The W3C spec has no mention of URL fragments and the IndieWeb Wiki merely states that:
"The development of marginalia has demonstrated that Webmention can be used to comment on specific parts of the content of a post."
but support is not yet defined, just that thought should be given to:
- How and when a webmention sender should use a fragment in the source and or target URLs
- How should a webmention receiver handle a fragment in the source and or target URLs
The 'common sense' approach that a lot of endpoints seem to take is to reference the first h-entry after the point on the page denoted by the fragment. This behaviour may not actually tally with the above but could be one option when processing the source URL.
My choices, therefore, are as follows:
- leave things as they are, ignoring those endpoints that don't handle URL fragments, and wait for the spec (or an extension) to properly define fragment support, or
- implement a pseudo-page system as I have for replies to comments
@colinwalker is there any reason you don't want a separate page for a post?
The blog operates almost as a stream of consciousness, designed to flow down the page so it doesn't really have posts per se, just sections within the day.