#microformats 2020-12-12

2020-12-12 UTC
#
[tantek] except for all the false positives of the stuff inside which are not ingredients
#
[tantek] from experience we know that such false positives noise is FAR worse a problem / inconvenience than per-element class attribute markup
#
btrem On a separate note, I've been looking into something to replace <object> in an include-pattern. Per our conversation yesterday, I took a look at new features in HTML5. I've only spent a day, but I have not found anything yet. This isn't suprising. It would be odd for a markup language like HTML to provide a mechanism for including part of a
#
btrem document in that same document for the simple reason that it's already in that document.
#
btrem I'm still working on this, though. Trying to come up with something that would work for a h-menu.
[chrisaldrich] and [tantek] joined the channel; btrem left the channel
#
[tantek] there's been lots of work on similar things actually, like web components, and HTML templates
#
[tantek] and yes to start with, better to come up with markup techniques that don't need use of the include-pattern
milkii, [KevinMarks], [tw2113_Slack_], KartikPrabhu, rawtext, jacky, [chrisaldrich] and strugee joined the channel
#
jeremycherfas One reason I stopped using recipe microformats was the hurt of manually adding the class to each ingredient. That and an apparent lack of consumption.
[schmarty] joined the channel
#
[schmarty] The idea of putting mf2 semantics on a list container (ul, ol) instead of each child element is interesting to me! I don't recall seeing anything like that in microformats parsing, before. Not sure where to start in evaluating that...
[KevinMarks] joined the channel
#
[KevinMarks] Well xoxo did that
#
[KevinMarks] Also parses dl dt dd as a dictionary/object
#
[KevinMarks] That was before we had a class based approach
#
[KevinMarks] http://microformats.org/wiki/xoxo
[jgmac1106] and [schmarty] joined the channel
#
[schmarty] Feels like a very different approach from mf2 as it is now.
#
@ZornGerhard ↩️ Sehr gut, in den Beispielen für RDFa wird verwendet: generische Elemente wie div oder Span als Grundbaustein und spezielle Attribute dann für die eigentliche Daten des Microformats (twitter.com/_/status/1337805689020952579)
milkii joined the channel
#
KartikPrabhu it could work for certain cases, like h-feed on a ul or ol could interpret all the list items as h-entry automatically
#
KartikPrabhu or even h-feed can treat all children <article>s as h-entry
#
KartikPrabhu sort of piggy-backing off of HTML semnatics
#
[schmarty] I'm skeptical of assuming list items of a parent h-feed are h-entry (h-feed can and is used for collections of h-card, h-review, etc). I think there would need to be a way to specify what the child items should be treated as.
#
[schmarty] To speak to jeremycherfas' example above it would be handy to specify "each item in this list is a p-ingredient".
#
[schmarty] Something like <ul class="l-p-ingredient"> comes to mind but it'd be a whole new thing
[tantek] joined the channel
#
[tantek] [jeremycherfas] the lack of consuming code is perhaps the primary reason to not "advance" any specific microformat or feature proposal. It's a good test as to whether or not there is any sustainable ecosystem relevance to any particular idea. And if there's no ecosystem relevance then it's a waste of time to bother with extra markup
#
[schmarty] indieweb.org/recipe seems to suggest that the only known recipe consumers actually consume mf1 hRecipe rather than mf2 h-recipe.
vesper11, [jgmac1106], gRegorLove and [chrisaldrich] joined the channel