#dev 2025-01-16

2025-01-16 UTC
ttybitnik joined the channel
#
btrem
Can you point me to examples of that pattern? Because I've only ever seen <header> <main> <footer>. E.g. mdn's page on the main element https://developer.mozilla.org/en-US/docs/Web/HTML/Element/main has an example under "Try it" that has a <header> followed by <main>.
#
gRegor
btrem, reading up on that chat log, found your h-review: https://btrem.com/2021/11-yellow-squash-rating It doesn't appear to be linking to another URL, though, so no webmention would be sent. The `item` property has a URL that's the same page.
#
Loqi
[preview] [Brian Tremblay] — Rating/Notes
#
btrem
gRegor: Probably because I overwrote the html after giving up on wm.io.
#
btrem
If you read the chat, you'll see that aaronpk says that what I was trying to do was not working, because wm.io was "being too picky" OSLT.
#
btrem
I just went back to that page, which is orphaned, and looked at the source code. You'll see that there are empty elements and attributes. That's where there *were* links to and titles of a recipe post. But I gave up on it, and then overwrote that page. Which doesn't bother me that much because, as I said, it's orphaned. More or less by design.
#
btrem
I should delete those pages, though. Shame on me for not cleaning up. :(
#
[tantek]
will look btrem, surprised by that on MDN
#
gRegor
btrem, tested some and added clarification to the bug report: https://github.com/aaronpk/webmention.io/issues/176#issuecomment-2594196253
#
Loqi
[preview] [gRegorLove] Some more details on this with my testing just now: I first attempted to post an `h-review` with `in-reply-to` = https://btrem.com/2021/10-yellow-squash. When I sent the webmention, wm.io responded: ```json { "status": "no_link_found", "...
#
[tantek]
yeah I/we need to address https://github.com/microformats/h-entry/issues/32 and see if we can get consensus on a merge and I can make that happen
#
gRegor
Ooh, yeah that would be a fourth test I could run on wm.io. I expect it would work since it's an h-entry. The bug does seem to be that wm.io is only expecting `in-reply-to` inside an `h-entry`
#
gRegor
btrem, Unfortunately doesn't help you post reviews as replies right now, but it's some progress in narrowing down the bug :)
#
btrem
gRegor: I think that your test is what I was doing. If you see my initial comment, it says " h-review by p-item -> h-* -> u-url." But I can't be sure, since as I said, I gave up. As you may have noticed, the issue is more three years old. FAOD, that's not complaint that it didn't get fixed, but rather me explaining why my test page is now broken.
#
gRegor
Yeah, I read through the chat. That specific case does work though, with my third test today. They were both received: https://webmention.io/api/mentions.jf2?target=https://btrem.com/2021/10-yellow-squash Still an issue that it shows up as `type: entry` instead of `type: review`
#
btrem
Also, I'm not sure that merging h-review into h-entry will help, either. I thought the problem was in the link. But I never looked into the code. aaronpk frightened me away. (Again, see the old micformats indie log that I posted a short while ago.)
#
gRegor
Only the first one with an in-reply-to wasn't received successfully
#
gRegor
No, the link is OK. It's a very specific implementation bug in wm.io receiving the webmention if it has `h-review.in-reply-to`
#
btrem
Got it.
#
btrem
FWIW, it doesn't really matter to me. I now link my reviews to my recipes internally, so to speak, using my static site builder. So when I create a ferment, I compose an .md file that generates an h-recipe. And when I open it, I create a second .md file with a review, which is added to the h-recipe page. The review is not posted separately. It probably doesn't make sense to post the review separately anyways.
#
[tantek]
gRregor the workaround fro now is to use a multityped object, that is e.g. class="h-entry h-review" on the root
#
gRegor
Fair, glad it's working for you! I just got a bit fixated on trying the different options, haha
#
gRegor
Good point on multi. Done testing for today though :)
#
btrem
[tantek]: regarding markup patterns, I think the point of <main> (though it has no real semantics, and is only "informative") is to mark the main content of the page (hence the name). So it appears to be intended to /not/ have <header> or <footer> inside.
#
btrem
I feel like there should be a new axiom, sort of based on Godwin's Law: as the amount of time you spend with a software tool or language grows longer, the probability of you finding significant problems approaches 1."
#
btrem
That's me with 11ty, markdown, and yaml. :-D
#
gRegor
Similar, that quote in #events earlier today "Every CMS is great for the first week"
#
btrem
Exactly!
#
btrem
I guess not so much 11ty. But I really hate the esoteric (and probably seldom used) features of yaml. For some definition of "feature." ;-)
#
btrem
gRegor++ for the additional wm.io test.
#
Loqi
gRegor has 35 karma in this channel over the last year (126 in all channels)
#
aaronpk
really wish i had that domain lol
#
[tantek]
want me to hard code it into the auto-linker?
ming__ and yewscion joined the channel
grufwub, sp1ff, strugee, jimw, AcesAndEights, thegr8whoopdini[, ludovicchabant, eatyourglory, karjala, gRegor, gRegorLove_, [KevinMarks], glacier, [Murray], doesnm, nemonical and Guest6 joined the channel
#
jeremycherfas
Do any of the more hardware-savvy people here have suggestions on how to turn the three 1TB SSDs hanging off a USB hub on my Mac into something more efficient? I had hoped there might be some kind of enclosure, rather than a hub, but there doesn't seem to be.
#
jeremycherfas
Or maybe I should just bite the bullet and spring for two new 2TB SATA drives and a little RAID enclosure?
GuestZero_, gRegor, gRegorLove_, btrem, jeremycherfas, ttybitnik, thegreekgeek, sebbu2, [fluffy], yewscion and grufwub joined the channel
#
aaronpk
[tantek]: is there a way to get a link to a specific version of an MDN doc like this? https://developer.mozilla.org/en-US/docs/Web/HTTP/Cookies
#
aaronpk
IETF doesn't like linking to living standards, I need a stable reference for it
#
carrvo[d]
Looks like the page has a link to the GitHub source, which could be turned into a specific version. Not really what you are looking for 😦 but could be a workaround.
#
[Joe_Crawford]
which has in the path the commit hash which gets you that. Not pretty, and not *precisely* the rendering of the page.
#
[Joe_Crawford]
But does have the content.
#
aaronpk
hm, yeah not ideal but might work
#
[Joe_Crawford]
https://perma.cc is directed at judicial and academic purposes. In a way standards-making is .... both academic and judicial.
gRegorLove_ joined the channel
#
carrvo[d]
Joe_Crawford++ that is a really interesting service!
#
Loqi
Joe_Crawford has 17 karma in this channel over the last year (119 in all channels)
mondodaemon joined the channel
#
aaronpk
[tantek]: the reference is actually specifically to the cookie prefixes here https://developer.mozilla.org/en-US/docs/Web/HTTP/Cookies#cookie_prefixes
#
aaronpk
which is not in RFC6265
max1 joined the channel
#
[tantek]
hmm, this looks like a more precise technical reference for that: https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Set-Cookie#cookie_prefixes
#
[tantek]
which probably should be incorporated into an update of RFC6265
#
aaronpk
true, let me see if maybe it already has...
#
aaronpk
omnomnom
#
aaronpk
[tantek]++ for helping track down references
#
Loqi
[tantek] has 29 karma in this channel over the last year (146 in all channels)
#
[tantek]
that's what I'm here for, pedantic standards details 🙂
#
btrem
Re the perma.cc page, I quote: "Use your Perma Link as you would the original cited URL. Your citation is safe from link rot — it will never change or break." So they promise. But obviously if the perma.cc site goes away, so do all of your citations. Seems like using that service is putting all your eggs in one basket, iow, maybe risky. :-o
#
gRegor
Plus side, it's apparently by Harvard Law Library and other law libraries, so hopefully a bit more stable than your average service
#
gRegor
Redundancy is always good though. perma.cc + wayback? :)
#
gRegor
what is perma.cc
#
Loqi
perma.cc is a service developed by the Harvard Law School Library for archiving web citation links https://indieweb.org/perma.cc
#
btrem
I didn't look into it, but I figured it was probably more stable than average, as you said. Still, it seems weird to say, "web links can disappear. So instead use web links." IYSWIM.
#
gRegor
Nice to see you at the FrESH meetup today btw
#
btrem
File under "naming things is hard": I'm trying to come up with good names for blocks outside of the main content of my pages. I thought maybe I could borrow vocabulary from book publishing, the there's too much overlap with web vocabulary where the overlapping terms mean different things.
#
btrem
Like I could use "preface" and "postface", except that "post" has a specific meaning in static site generators. So what happens six months from now when I see div.postface? "Is that the front of the post? What is that?"
bterry joined the channel
#
carrvo[d]
Why not reference with the original link AND the perma.cc link? Then your eggs are in multiple mirror baskets.
#
btrem
There's "front matter" and "back matter", but that's even worse for me. Front matter is the structured data at the top of my .md files. I'd hate myself six months from now for using that term. Especially since I'm currently experimenting with putting text in my front matter like acknowledgements that will appear after the <main>, thus in a "back matter" block!
#
btrem
carrvo[d]: might be a good idea, though publishing patterns don't really accommodate that ATM.
#
greg.just.greg
Would something like "headnote" and "footnote" work, or does that just create the "head" version of the "post" problem?
#
carrvo[d]
btrem, searching for a better name is good, but I wouldn't discount "postface". "Post-" is actually a prefix for many common words (postgame, postpone, postmortem, postcard). Balanced with the context that "post" is used (web, post office, et cetera).
ttybitnik joined the channel