2018-11-11 UTC
strugee joined the channel
# 00:37 kisik21 any stable parsers for h-entries that are usable in production? aaronpk, I've heard you have one
# 00:37 aaronpk do you mean a microformats parser? there are parsers in a bunch of languages
# 00:38 kisik21 umm... I mean microformats -> json via web service
# 00:38 kisik21 API-like and curlable
# 00:38 kisik21 wanna show reply-contexts without JS, can consume JSON or CSV
# 00:40 Loqi aaronpk has 98 karma in this channel over the last year (307 in all channels)
# 00:57 [eddie] Well this isn’t good... the Share Sheet just isn’t showing up when I tap it for Indigenous 🤔
# 00:58 [eddie] Anyone else happen to have Indigenous for iOS installed and notice the Share Sheet won’t activate from various apps (like Safari)
# 01:14 [eddie] I guess I’ll just reinstall it (because it’s a dev build)
renem joined the channel
# 03:02 GWG gRegorLove: I'll follow aaronpk's yay
# 03:02 GWG No one has explained to me why people don't like h-feed
# 03:19 GWG I couldn't even get a GPS lock on my last flight
# 03:23 aaronpk that one is just a stylistic line. my actual flight traces don't look very nice
# 03:23 aaronpk I don't often get a GPS lock unless i'm at a window seat
# 03:30 GWG I meant, the whole itinerary feature.
# 04:27 aaronpk This is still about finding the author of a post within a feed tho right?
# 04:27 aaronpk I think "feed authorship" isn't the right term for that
# 04:30 aaronpk I think your step 2 is not quite what I do. I was trying to do more like: if there are 2+ h-* items, one of which is not an h-card, and if exactly one of them is an h-card, then the h-card is the author of the rest of them
# 05:07 aaronpk I haven't cared about finding the author of a feed yet
# 05:08 aaronpk different problem, likely requiring a different solution
# 05:12 aaronpk Right and if you look at Adactio's feeds you end up running through all those steps and not finding an author
# 05:15 aaronpk So there are some missing steps when looking at items within a feed. I think that was written primarily in the context of permalink pages.
# 05:16 gRegorLove That sounds like a publishing error. His permalinks don't have author properties either
# 05:16 aaronpk That was my first thought but then I realized there is enough information in his feeds to determine the author
# 05:19 aaronpk It's working now too! I finally see his picture on all his posts in Monocle
# 05:36 gRegorLove So my h-feed author thing was a separate issue. I think that's what threw me and had me focusing on feed authorship.
# 05:50 aaronpk Your intent was to describe the author of the feed right?
# 07:35 mblaney aaronpk I'm glad that the new authorship algorithm fixes the author issue for adactio.com, makes me want to add it too.
# 07:37 mblaney just a bit hesitant though, adactio has over 40 h-cards on his home page, so you're relying on ordering to give you the author? makes it a little bit precarious.
# 08:45 gRegorLove No, at the time I set up that homepage h-card my intent was just to do the u-author=/ thing on my posts. I didn't think about the fact my latest post is wrapped in h-feed on the homepage as well.
[kevinmarks] joined the channel
[jgmac1106] and eli_oat joined the channel
# 14:16 aaronpk mblaney: if you look at the tree view it's obvious which one is the right h-card. The trick is to use a microformats parser and not just use css selectors to find h-cards
[jgmac1106] and [eddie] joined the channel
# 14:36 aaronpk Ohh I just looked at his home page again... i didn't realize all the h-cards in his bedroll are also top level objects
[jgmac1106] joined the channel
# 15:51 GWG Why can't I find a podcast with a jsonfeed? It shouldn't be that hard.
# 15:52 aaronpk does any podcast software support jsonfeed right now?
# 15:53 GWG According to the specification, enclosures in RSS map to attachments in JSONFeed
# 15:54 aaronpk if anyone were going to be doing it I would expect micro monday
# 15:55 GWG NPR claims they do it, but don't link to it.
# 15:55 GWG I suppose, if I cannot find a good example to test against in 5 minutes, I should not worry too much
# 15:56 GWG Surprised [eddie] 's JSONFeed doesn't do attachments
# 16:01 aaronpk I guess I should add support for that to xray too
# 16:12 GWG I just added it to Parse This, my x-ray like library
[kevinmarks] and [eddie] joined the channel
# 17:03 [eddie] GWG, right now I built our JSONFeed to just focus on being perfect for Micro.blog, I do plan on revisiting all the potential properties of it and making sure it supports everything it can
jjuran and deathrow1 joined the channel
[jgmac1106] and eli_oat joined the channel
# 19:54 Loqi [arnabwahid] #1296 Remove Microformat Classes `.hentry`
# 20:14 Zegnat That PR is different from what the title suggested to me: it doesn’t remove the hentry class from the output HTML, but stops them from attaching CSS to it.
# 20:15 aaronpk Oh. That's a good step but not what I thought as well
# 20:17 GWG That's why I said it took too long
# 20:18 Zegnat TIL: actual hentry class isn’t part of the theme at all, but inserted by a WP core function `post_class()`
# 20:21 aaronpk at least now there is one fewer thing relying on .hentry in wordpress
# 20:31 gRegorLove Is post_class() hookable? Might be possible to remove .hentry that way
# 20:45 gRegorLove aaronpk, XRay doesn't appear to follow step 7 of authorship. Is that a known limitation?
# 20:52 aaronpk there's no microformats on the h-card page so it fails
# 21:01 GWG Re xray. How does x-ray decide how a page with a feed and h-card should be returned?
# 21:05 aaronpk it does some tricks to figure out whether it thinks the page is a feed or a post permalink
# 21:05 aaronpk but also it has a mode where you can tell it you're expecting a feed
# 21:06 GWG The mode is what I need to.implen
# 21:09 sknebel from what I remember it ignores h-cards as long as it can (which is why jgmac1106's about page always gets a random comment from it as Loqi's preview)
chrisaldrich and [tantek] joined the channel
# 23:04 mblaney aaronpk maybe we can change step 8 of authorship discovery then? "if the author has not yet been found, it will now be obvious." :-D