singpolyma.netedited /GNU_social (-41) "Move old history information down the page a bit. The relevant information (that the project used to be called StatusNet) is part of the summary at the top still." (view diff)
singpolyma.netedited /GNU_social (+219) "/* Open Questions */ Describe how to put profile info into an Atom feed for GNU Social consumption" (view diff)
tantekNote the "/Social Web" - that should be "/ Social Web", and there's a return (cr) in the original that somehow Bridgy Publish removed, and thus errantly appended "Social" to the end of that IG URL
petermolnarI've found a solution for my photo vs article tell-apart issue from yesterday: I do set a article character limit to separate short/long, but in case it fall to photo identification, compare the text difference percentage between the post content and the photo description, coming from IPTC data; if it's within a certain range ( 'cause I do copy the IPTC data as post content by hand when posting a photo ), I can assume it's a photo post
nt0, j12t, loic_m, elima and indie-visitor joined the channel
ben_thatmustbemeyeah, so on android at least, when launching a browser window inside an app, its totally insulated, so no saved passwords, you have to log in manually to whatever site you are using via typing in info on phone keyboard
aaronpkwhen it comes down to it, it turns out computers don't need to know/model the full depth of types of objects in order for the information to still be useful to people reading it
kylewmyeah you're right I think. it's useful for me to know if something is an event, but i don't much care if an h-card is a person/location/corporation/etc.
[shaners]I’m adding Pages to Homesteading now. Pages almost identical to Articles. Just not in the timestream. Duh. We all know and agree on that. (I think.)
aaronpkbasically i now have a "post" object, which can have either text/plain or text/html content type, and can have a bunch of properties like in-reply-to. i can also set a custom URL for any post
kylewmactually I'm kind of surprised that you (aaronpk) are the big proponent of implicit types, it seems like you have the most to gain from explicit with all the different metrics and stuff you track
[shaners]aaronpk: In two years, if I’ve rebuilt, am rebuilding, or want to rebuild HS to remove explicit post types, I will buy you dinner. If not, you’ll buy me dinner. Deal?
Loqifollow is a common button in silo UIs (like Twitter) that adds updates from that profile (typically a person) to the stream shown in an integrated reader, and sometimes creates a follow post either in the follower's stream ("… followed …" or "… is following …") thus visible to their followers, and/or in the notifications of the user being followed ("… followed you") https://indiewebcamp.com/follow
voxpellialthough he has a different indie-config setup than most, where he caches the result of the indie-config in his site's localhost or something rather than loading it live
aaronpkalso ironic is the fact that you can't upload files with JSON posts, so if you are going to post a photo or video, you need to use form encoding anyway
aaronpksingpolyma: kevinmarks: yea there are any number of ways you *can* get files into JSON post requests, but none are standard the way multipart encoding is
aaronpkwhoa when did this happen "Beware. This specification is no longer in active maintenance and the HTML Working Group does not intend to maintain it further."
aaronpk!tell tantek do you know what happened to HTML JSON forms? http://www.w3.org/TR/html-json-forms/ "Beware. This specification is no longer in active maintenance and the HTML Working Group does not intend to maintain it further."
snarfedfirst, retrying responses is much improved. it retries *all* URLs, works harder to find candidate URLs (including syndication URLs), and you can page through old responses so you can retry them
snarfedsecond, for finding your user page: along with remembering your accounts when you log into them, it now also lets you type your username, full name, or domain instead of user id in the URL (for silos where it needed that before)
gRegorLove"Symantec performed another audit and, on October 12th, announced that they had found an additional 164 certificates over 76 domains and 2,458 certificates issued for domains that were never registered."