#microformats 2022-11-30

2022-11-30 UTC
#
tantek
edited /Special:Log/patrol () "marked revision 70586 of [[h-recipe]] patrolled"
(view diff)
#
[jacky]
all hail h-app
#
[tantek]
it's h-app ening
#
[schmarty]
bahahahaha
#
Loqi
[schmarty]: lol
#
gRegor
hahaha
gRegor, [jeremycherfas], angelo, gRegorLove_, ur5us, alex2, tiim, barnaby, [manton], ash[m] and btrem joined the channel
#
btrem
@aaronpk @tantek re e-logo, I'm already using that. Don't know if anyone is making use of it, but you can see it here: https://greenbankplants.com/
#
Loqi
Greenbank Plant Nursery
#
btrem
Besides e-logo, I'm using an h-card class name on the `text` elements in the svg: `<text class="p-name fn">Greenbank Plant Nursery</text>` (simplified for clarity).
#
[tantek]
btrem have you tried it with any microformats parsers to see what is extracted?
#
[tantek]
could make for an interesting test case!
#
Loqi
[Brian Tremblay] Adding Microformats to SVG
#
btrem
As I matter of fact, I have. :)
#
[tantek]
that is very cool btrem++
#
Loqi
btrem has 2 karma over the last year
#
btrem
Note that my proof that the text element is getting parsed by search engines doesn't necessarily mean they're parsing the *microformats* therein. It might be paying attention to the text element because its ancestor is `<h1>`.
#
btrem
I don't know of any way to prove that a microformat is being used in some special way by a search engine.
#
[tantek]
that's the problem with (most / big / generic) search engines, you can't really prove anything about their behavior
#
btrem
Right.
#
btrem
Both ddg and google include the business' address, marked up as h-adr. That would tend to make me think they *are* parsing the h-card. But no way to know for sure.
#
[tantek]
right, they could be parsing mf2, or they could be running some energy intensive ML-based text scraper
#
btrem
What is ML?
#
btrem
(Yes, I know you can't do that here. So I'm actually asking you!)
#
[tantek]
machine learning
#
btrem
Doesn't seem all that different than e-content.
#
Loqi
[aaronpk] i have no problem with embedded svgs on a website, but when you are talking about transporting them around via microformats that's a bit scarier
#
btrem
Speaking of microformats and svg: I just learned that microdata attributes are not permitted on svg elements, whereas class is. So if one wants to add machine readable bits of data, microformats is the only way to do that. With the caveat that, again, there's no way to know what e.g. search engines do with any of this stuff. Maybe they process microdata in spite of the spec.
ur5us joined the channel
#
btrem
Is the microformats/microformats.io repo dormant?
#
Loqi
[microformats] microformats.io: Simple explainer site about Microformats.
#
aaronpk
you can put all sorts of interesting things in an SVG, things that I would be worried about republishing in a reader
#
aaronpk
HTML sanitization is normal, there's a short list of allowed tags that I support in the reader
#
[tantek]
aaronpk, at least SVG has well-defined restricted profiles that you can subset to
#
aaronpk
but if I'm going to be rendering SVGs, i would need to do some sort of SVG sanitization too
#
[tantek]
HTML sanitization is still non-standard
#
aaronpk
like, javascript is allowed in SVGs, that sounds like a great XSS vector
jamietanna joined the channel
#
tantek
created /svg (+145) "stub with btrem's post & example!"
(view diff)
#
[tantek]
btrem ^ feel free to expand!
#
gRegor
btrem, we still use microformats.io, just may not have been updated recently. PRs welcome. There's separate repos for subdomains on there too, like php.microformats.io
jamietanna joined the channel
#
sknebel
gRegor I assume btrem asks because of their open
#
sknebel
PRs :D
#
gRegor
reviewing :)
#
btrem
Yes, that's why I was asking.
jamietanna joined the channel
#
btrem
I've been working on a complete rewrite of the micformats.io page, removing the bootstrap css stuff, which seems unnecessary for such a simple one page site. And then simplifying the markup. That would make the source code a bit more of a good example of coding with mf.
#
btrem
But I haven't pushed that yet. I was waiting for the minor changes to be merged.
#
@brianwisti
↩️ Not that I found specifically for pytest, but there are numerous microformats2 validators and at least one cross-format handler on PyPI: https://pypi.org/project/granary/
(twitter.com/_/status/1598070100263067650)
#
gRegor
btrem, merged all but https://github.com/microformats/microformats.io/pull/54, small note and need to resolve branch conflicts, otherwise good
#
gRegor
btrem++
#
Loqi
btrem has 3 karma over the last year
#
Loqi
[btrem] #54 Fixes alt attributes
#
gRegor
I subscribed to notifications, too. I had it only on 'mentions'
[mattl] and tiim joined the channel; tiim left the channel
#
btrem
edited /svg (+411) "Adds simple example of microformats in svg."
(view diff)
#
btrem
Well, I copied/modified the code from another page, and apparently used a deprecated tag `<source>`. Annoyingly, the wiki software doesn't tell me what the /replacement/ is. (grumble grumble)
#
btrem
gRegor Ah, well I just fixed the PR, but didn't mention you. So you probably don't know. (Well, you do now.) :)
#
gRegor
btrem, the replacement is <syntaxhighlight lang="html">
#
btrem
Thanks gRegor++.
btrem joined the channel; btrem left the channel
#
btrem
edited /svg (+18) "Fix: replaces deprecated <source> tag with <syntaxhighlight>"
(view diff)
#
btrem
edited /h-card (+80) "Fix: replaces deprecated <source> tag with <syntaxhighlight>"
(view diff)
#
btrem
edited /microformats2-parsing (+54) "Fix: replaces deprecated <source> tag with <syntaxhighlight>"
(view diff)
#
btrem
edited /microformats2-parsing (-11) "Fix: corrects syntaxhightlight lang attributes"
(view diff)
#
btrem
edited /h-recipe (+36) "Fix: replaces deprecated <source> tag with <syntaxhighlight>"
(view diff)
#
btrem
edited /h-recipe (-13) "Fix: corrects syntaxhightlight lang attributes"
(view diff)
#
btrem
edited /existing-rel-values (+72) "Fix: replaces deprecated <source> tag with <syntaxhighlight>"
(view diff)
#
btrem
edited /svg (-1) "Fix: corrects syntaxhightlight lang attribute"
(view diff)