#tantekit creates more easily hotswappable solutions
#rrixRight now this thing just ingests web browser history, and stores textual snapshot of the page
#tantekwhereas the "put everything in a giant knowledge graph!" approach tends to cause you to get stuck with whatever that one thing is, or constantly maintaining it, or being very scared when you migrate
#rrixThe data model is stupid enough that dumping JSON blobs of each item in the store and doing ... something with them makes me feel mostly safe with it, but that's a good thing to keep remembering
#rrixcase in point: the two unmerged instances of my blog
#tantekHackpad is an interactive collaborative document editing silo that was acquired by [[Dropbox]] and is shutting down on 2017-07-19 (email to users). Existing users were encouraged in a 2017-04-21 email to try using [[Dropbox Paper]] instead, with the options of importing Hackpad pads into Paper, or exporting your pads into a single zip folder, or do nothing and let pads be automatically migrated to Paper on 2017-07-19, or opt out of
#Loqi[strk] #48 Fails to find link attributes in HTML with uppercase HEAD tags
friedcell joined the channel
#[pfefferle]strk sure, but I think we discussed it before... I would prefer to not edit the php lib for WordPress, because the changes might get lost if I had to update the lib again ?
#GWGpfefferle, I think it is a combination of how Known marks it up and how Semantic markup interprets it. That is why I want to work on the handler.
#[pfefferle]colinwalker GWG I think that is the point... there are a loot of "correct" ways to markup a like and that makes it hard to handle all...
#[colinwalker]I've not even got that 'like of' in my comments.
#[colinwalker]Scratch that - thought it was for something else
#[pfefferle]the like-of is the recommendet property to send likes, but the plugin also supports rel="like" and class="u-like"
#GWGpfefferle, when I did the mf2 code in Post Kinds, I kept throwing we examples at it to test. I may do that again
#GWGBuild a simple UI for the mf2 parser in Semantic Linkbacks to display what it returns from a given URL
#[pfefferle]I wrote the initial plugin against voxpellis webmention tester
#[colinwalker]I'm sending class="u-like-of" and that is recognised correctly.
#[pfefferle]colinwalker it is not only the attribute, but also where you put it... known seems to add it outside of the h-cite/h-entry... if you send a post, it is inside the h-entry
#[pfefferle]colinwalker so the parser adds it to different levels of the JSON tree
#[pfefferle]colinwalker and that makes it so hard to check the mention type
strk[m], barpthewire, friedcell, nitot, loicm_ and tantek joined the channel
#strk[pfefferle]: ah, now I understand - but you could test it and report upstream if you like it or not, as I think the upstream maintainer (openid-php) is having prolems testing that path
#strkit's easy to try: just use uppercase tags in your openid page, and it will fail (an uppercase <HEAD> tag is enough to break it)
KevinMarks, friedcell, hs0ucy, tantek, Leeky, loicm_, nitot, singpolyma, mlncn, ebel, leg, [kevinmarks] and gRegorLove joined the channel
#miklbthis seems like a slippery slope "but known is very wide spread, so we should support it asap..." especially if the question is whether known is properly marking up the webmentions
#miklbthen it is a bug in how the WP plugin interprets the mf2 and stores it.
[jeremycherfas] joined the channel
#[jeremycherfas]One of the indieweb plugins is playing havoc with the presentation of my site, making visible things that I had made invisible with CSS. Not sure yet which one.
#[jeremycherfas]I don’t have that many posts, maybe 200, but they involve my podcasts and I need to think hard before I change too much. On the other hand, may be better to do that thinking before I get too much further with indiewebifying.
#[colinwalker]aaronpk Funny you should mention your PR on known as I got a bookmark from jeremycherfas that is also only seen as mention - presumably why
#[jeremycherfas]Ah. Anything I can do about that? I shared that one via Quill.
mlncn, plindner, M-mxuribe, Salt[m], M-tom, martymcguire[m], pniedzielski[m], TheGillies, M-podviaznikov, jcgregorio[m], rdesfo[m], afrogeek[m], SpEcHiDe, mindB, M-hotzeplotz, strk[m], garlox[m], jaduncan[m], j-hernandez[m], crasch[m], dgold[m], Guest86807[m] and friedcell joined the channel
#[jeremycherfas]Looking in detail, my child theme is such a mess, I should certainly consider going back to square one and either constructing a “proper” child theme or else switch now to one of the indieweb-enabled themes.
#[colinwalker]jeremycherfas it made me realise mine needed a tweak so it's not all bad ? But seriously, I hear you. The theme I'm using is a highly modified version that I've pulled apart and added to so there's probably all sorts in there that shouldn't be.
#[jeremycherfas]I really like your theme, but not for this particular site of mine. So, I think I need to bite one bullet or the other. Probably cleaning up the child theme, for now.
jmelesky, Pierre-O, pfefferle, strugee`, KevinMarks_, wolftune and [chrisaldrich] joined the channel
#[chrisaldrich]jeremycherfas did you ever install the Post Kinds plugin on that site? iirc i think that it will/can persist in some of its outputs even if it's disabled after the fact.
#[chrisaldrich]If this is the case, you may have set your default for posts as an aside which could be injected into all the posts and potentially still be causing your issues after deactivation.
#[chrisaldrich]Or rather your default Kind was set to note which could have auto set the post format to aside. I think there was a small window recently when the code may have done something like this.
#tantekgRegorLove: good point re: name badges - maybe an update with the new logo?
snarfed and nitot joined the channel
#aaronpki believe i have a script that generated a PDF with the names. i'll dig that up
#gRegorLoveLooks like we'll have a small turnout for Bellingham, but that's ok. First year, plus it's on weekdays. Looking forward to more focused conversations and more time to work on stuff.
funwhilelost and [jeremycherfas] joined the channel
#[jeremycherfas]chrisaldrich I honestly cannot remember. I think it was another WP site where I activated Post Kinds. I’m going to try and see how hard it would be to duplicate the content of my current static home page in SemPress or similar.
KartikPrabhu joined the channel
#[jeremycherfas]My current child’s parent looks like it has been abandoned anyway — nothing for more than 2 years — so the investment in a new child is probably long overdue.
#gRegorLoveI did mis-read "current child's parent" as in humans and was very confused by the terminology for a moment.
#[jeremycherfas]Actually, Independent Publisher theme looks rather interesting. With a bit of tweaking for the presentation of post kinds, it could satisfy my needs. But I’m going to keep looking too.
#[colinwalker]Oh, the benefit of coming in _after_ that exchange had run its course
[mko] joined the channel
#[mko]Has anyone ever tried to POSSE or export their Activity Log from Facebook into something usable for archival purposes (i.e. ActivityStreams JSON)?
#[mko]Couldn't find any good references on the wiki
#tantekmko I think you can get AS1 from FB using granary?
#Loqigranary is a library and REST API that frees you from social network snowflake API and exposes the sweet social data foodstuff inside as HTML and JSON with microformats2, ActivityStreams, Atom, XML, and more https://indieweb.org/granary
#[mko]Oh. I meant the actual `allactivity` page contents.
#[mko]There's a substantial amount of "hidden" social activity that Facebook doesn't expose in a way that Granary can currently get to it (that I can see anywys)
#[mko]If you look, the HTML is basically garbage, so it's not really set up for parsing into ActivityStreams or Microformats or anything else structured off the bat, hence why I was wondering if anyone had tried to parse it out yet.
#gRegorLoveThat'd be cool if there's a way to consume it. Not aware of one offhand.
#[mko]Yeah. It got me thinking because it holds the keys to some of the most interesting content, most notably the timeliness of Likes and interactions with other people that do not generate posts on the user's own page.
#tantekmko, that strikes me as the kind of thing snarfed would want even just for himself
#[mko]Oh, I know that there's no current way to do it except maybe some sort of "hack" that has a basic knowledge of the HTML structures of the page and knows how to extract particular interactions from it.
#[mko]Just was hoping someone had already done so.
#tantekalso sounds like an unmung opportunity (KevinMarks), since unmung is all about unmunging whatever random feed format thingie you have into something readable (and parseable!) in HTML :)
#[mko]Yeah. It's easy enough to "archive" it (along with all the related content). The drawbacks are that it's time consuming, and it's not really usable for anything meaningful.
#[chrisaldrich]I wasn't as productive as some who've been working on 100days projects, but it was an interesting exercise to look back at the past year a bit.
#[chrisaldrich]Special thanks to everyone here for all the help over those years!