2016-12-14 UTC
tantek, KevinMarks, KevinMarks_, cweiske, KartikPrabhu, j4y_funabashi and Sebsel joined the channel
# 13:08 Loqi ok, I added "http://ascii.textfiles.com/archives/1717" to the "See Also" section of /why
Sebsel, KartikPrabhu and singpolyma joined the channel
# 14:29 Loqi ok, I added "http://programminghistorian.org/lessons/sustainable-authorship-in-plain-text-using-pandoc-and-markdown" to the "See Also" section of /markdown
# 14:29 loqi.me edited /markdown (+105) "petermolnar added "http://programminghistorian.org/lessons/sustainable-authorship-in-plain-text-using-pandoc-and-markdown" to "See Also"" (
view diff )
KartikPrabhu and gRegorLove joined the channel
tantek joined the channel
# 17:21 Zegnat I would love to do a checkin for our HWC location
# 17:22 aaronpk ah yeah, i don't think it's settled enough for that yet
# 17:26 aaronpk i have a feeling 2017 is going to be the year we finally figure it out
# 17:26 tantek big question right now is whether it should be done with a new property on h-entry , or a new h-* top level object. leaning towards brainstorming the former
# 17:30 Zegnat I kinda need to figure out checkins, yes. Checkins and bookmarks are the 2 I am planning to fully handroll from 2017-01-01
# 17:31 Loqi A bookmark (or linkblog) is a post that is primarily comprised of a URL, often title text from that URL, sometimes optional text describing, tagging, or quoting from its contents https://indieweb.org/%F0%9F%94%96
# 17:31 tantek somehow I left it open for future use by review :)
# 17:34 aaronpk that's basically the reason we haven't settled on markup yet
# 17:34 aaronpk one aspect i haven't figured out yet is both showing the venue's website but also linking to my URL for the venue
# 17:35 aaronpk because i want my own venue URLs to be able to show all my checkins at a place
# 17:35 aaronpk i think that's the main thing i'm stuck on with checkins right now
# 17:36 Zegnat I kinda like the h-as-checkin, because it would allow me to use the exact same h-entry with location info as both a checkin or not a checkin
# 17:36 Zegnat So I could have a photo post with with location, just because, but I can also mark it as being a checkin
# 17:37 Zegnat Is there any material on the h-as-* syntax? Or is it scrapped?
# 17:37 Zegnat sebsel and I are mostly brainstorming at HWC, j4y_funabashi ;)
# 17:40 Zegnat Does *-as-* specifically mean ActivityStream? I was assuming it was an extra way to define post type. (Think aliases in SQL “AS”)
# 17:42 Zegnat From the brainstorming: I kinda feel like the entire post is my checkin. Not the hcard of my location.
# 17:42 Zegnat So to me it doesn’t make sense to have a specified p-checkin
# 17:42 tantek hasn't been archived to the wiki yet - on my list :/
# 17:46 j4y_funabashi cheers guys, I assumed it had already been discussed. it is certainly a tough one
# 17:49 j4y_funabashi Zegnat: I agree, checkins seem like a post with properties of their own, not just a property of an h-entry
# 17:49 Zegnat so there .h-entry.h-as-checking>.p-location makes sense to me
# 17:50 Zegnat I am not sure if a review makes sense on a checkin though, you can’t really review at checkin, only after.
# 17:52 Zegnat And while I brainstorm, sebsel is getting actual work done.
# 17:52 aaronpk if you're talking about a new top-level object, it would be just h-checkin, and wouldn't have multiple h-* classes
# 17:52 Zegnat Maybe. I personally do not have a usecase for posting venues as seperate things in my feed
# 17:52 aaronpk just keep in mind what the multiple h-* classes does to the parsed result of the page
# 17:54 j4y_funabashi is late to the checkin party and thinking I need to add a new itch for 2017
# 17:54 Zegnat Right. Hmm. So it would become either .h-entry.e-checkin or .h-checkin ...
# 17:57 tantek there's insufficient case for a whole new object
# 17:58 tantek because in all contexts (existing checkin apps, or checkin features on silos), checkins show up just like all other kinds of posts in people's streams
# 17:58 tantek so that implies they are just another h-entry
# 17:58 aaronpk right, they share most other properties of h-entry
# 17:58 tantek so the evidence of UX implies checkins are just posts
# 17:59 aaronpk contrasted with h-cards, which have very few common properties
# 17:59 aaronpk and you don't normally see h-cards in a list along with posts
# 17:59 tantek j4y_funabashi: that's where it's better to re-read previous IRC discussions
# 18:00 Zegnat yes, I want my posts to just be a checkin, or not be a checkin, I agree with that tantek
# 18:00 sknebel what's the microformat prefix for a boolean flag -is-checkin? :P
# 18:00 Zegnat But I feel like my entire post is part of the checkin. My content and/or photos included. So I don’t really like only adding a class for it on location
# 18:01 Zegnat sknebel, that’s what sebsel raised here at the table as well, feels like a boolean
# 18:01 aaronpk not sure what you mean 'adding a class for it on location'
# 18:01 aaronpk it's more like "because this h-entry has a p-checkin property, it should be presented as a checkin in the UI"
# 18:02 Zegnat sure, but would p-checkin become synonymous with p-location? Else I will just be doing .p-checking.p-location. Or maybe even .h-entry.e-checkin
# 18:03 aaronpk no, the p-checkin would *be* the location. you don't need p-location also
# 18:03 aaronpk class="p-checkin h-card" which ends up with a property called "checkin" that is an h-card of the venue you checked in to
# 18:04 Zegnat I don’t mind a *-checkin property. I think that’s where we need to go. But what would you expect inside the p-checkin when you request it from your parser
# 18:04 Zegnat Alright, so I would have h-entry with .p-location.h-card, and then when I want to mark them as checkins I swap p-location for p-checkin
# 18:05 Zegnat That gets pretty close to the aforementioned boolean/flag
# 18:07 sknebel arguably shouldn't the h-card be p-location and p-checkin?
# 18:08 Zegnat not if we decide they are synonymous, sknebel, that would be applied
# 18:08 tantek sknebel, booleans rarely work well over time in formats
# 18:09 sknebel it meets the meaning of p-location as well, and one less case to handle for when my application doesn't care about the difference
# 18:09 tantek you typically end up wanting to extend any booleans that you end up using
# 18:09 tantek Zegnat, the " feel like my entire post is part of the checkin" feels are true for all the other kinds of posts too
# 18:09 Zegnat I would be tempted to do p-checkin on h-adr or h-geo though, not just h-card. But I don’t see that being a problem
# 18:10 Sebsel well, you can have a .p-checkin.h-card for, say, a park, and then a .p-location.h-geo of your exact location
# 18:10 tantek the "feel of the entire post" does not make any sense as a methodology for driving format design
# 18:10 Loqi sebsel has 1 karma in this channel (7 overall)
# 18:10 aaronpk foursquare records the exact lat/lng of the checkin along with the venue
# 18:10 aaronpk and they use it to make little point clouds that represent the venue
# 18:11 Zegnat h-card can contain geo, but if I am in the woods, I may not want to create a h-card for a specific tree, yet still checkin. So I would reserve using h-cards for actual venues
# 18:12 Zegnat tantek, while true, I would like the checkin property content you get back from a mf2 parser to make sense. And I am not sure I like that property being just the address.
# 18:12 Zegnat But that gets us in the debate of wether u-photo should be within e-content or not ;)
# 18:12 aaronpk it's not uncommon for a property to have multiple values
# 18:12 tantek it makes less sense for it to have "all the info"
# 18:12 tantek that's what the h-entry is for, it contains "all the info"
# 18:12 sknebel and if we need an h-photo to encode details about the photo and not the post ;)
# 18:13 tantek sknebel: you're welcome to document use-cases and real world examples of that if you find them
# 18:13 Zegnat sknebel: I am not sure, that’s why the “feel” ;) I think I might go with the p-checkin as special alias for p-location and see how it feels
# 18:13 aaronpk right now in h-event, p-location can be an h-geo, h-card, or h-adr
# 18:13 tantek zegnat, this is part of the previous IRC location discussion
# 18:13 tantek p-location is the location of where the post is/was made
# 18:14 Zegnat Yes, these discussions tend to circle, that’s probably why we don’t really have nailed it down yet, tantek
# 18:14 tantek which is different than what the post may be *about*
# 18:15 tantek e.g. I come home from vacation but post a photo from vacation
# 18:15 tantek the p-location is home, the location-tag is of whatever the photo was of
# 18:16 tantek Zegnat, no they circled for a while, then we figured it out
# 18:16 Zegnat if p-location is *always* my current location, that would mostly negate checkin completely. I guess I missed the result of the previous discussion!
# 18:17 tantek yes, p-location is the location from which the post was published, like the location on a tweet
# 18:17 gRegorLove Someone's going to drop a link to today's logs on /checkin right, so next time we have one more reference point? :)
# 18:19 Zegnat That would mean though that a checkin-after-the-fact (much like an rsvp-after-the-fact) would contain separate p-checkin and p-location data, tantek?
# 18:19 Loqi ok, I added "https://chat.indieweb.org/dev/2016-12-14" to the "See Also" section of /checkin
# 18:20 Loqi postsquare (or postsq) is a portmanteau of "post" (meaning after) and "Foursquare" (the checkin service) meaning a checkin post made after the fact, that is, after the person posting it was actually at the location, or perhaps after they arrived at the location https://indieweb.org/postsquare
# 18:21 Zegnat So I guess I should use .p-location.p-checkin.h-[card|adr|geo] if I am posting from my checkin location
# 18:22 Zegnat Or we have to define p-location as implied to be p-checkin unless a separate one is specified.
# 18:24 Zegnat yes, sorry, p-checkin implies p-location unless otherwise specified
# 18:29 Zegnat Hmm, I need to remove h-feed from the permalink pages
# 18:33 Sebsel So, we had a HWC and all we did was discussing and implementing a way to say we where there :)
# 18:38 Zegnat aaronpk, I will have to fix that when I get home, I don’t currently have access to my backend code from where we are sitting
prtksxna joined the channel
# 19:42 Zegnat tantek: yes, we posed for a shot together right before leaving
KevinMarks and KartikPrabhu joined the channel
singpolyma, KevinMarks_ and KevinMarks joined the channel
KevinMarks joined the channel