2015-02-22 UTC
emmak joined the channel
# 00:03 GWG I just changed over to having Watch and Listen as two post kinds I support. I don't have Read. I'm wondering how Read and Bookmark compare.
j12t joined the channel
j12t joined the channel
# 00:30 aaronpk wordpress doesn't think these comments i am importing exist
# 00:33 aaronpk now wordpress says "Fatal error: Call to undefined function get_linkbacks()"
# 00:36 aaronpk i do not know how to proceed. that function should be defined in the plugin.
# 00:41 GWG aaronpk: Are you using the development or the repository version of Semantic Linkbacks?
# 00:41 GWG The function is in the development version.
# 00:42 aaronpk Whatever I installed through the Wordpress plugin directory
# 00:43 GWG aaronpk: I'll talk to pfefferle and see if he'll push it.
# 00:43 GWG The repository date is June, the item in question was accepted as a pull request in November
# 00:44 aaronpk oh huh, does that mean it's broken for anyone who installs it right now?
# 00:45 GWG Surprisingly, no one mentioned it in the last 90 days since the change occurred.
# 00:45 GWG Semantic Comments isn't quite ready for primetime, I guess.
# 00:45 aaronpk More importantly for me right now, can I install the repo version and it is fixed?
# 00:46 GWG !tell pfefferle Can you push the latest stable version of Semantic Linkbacks into the WordPress repository?
# 00:46 Loqi Ok, I'll tell them that when I see them next
# 00:47 GWG aaronpk: I switched my main site, as opposed to my test site, to not using development versions as much as possible because of this issue. I want to catch them
# 00:48 GWG aaronpk: Either way, you can disable Semantic Comments.
# 00:48 GWG It just does the Facepiles, if you want those
# 00:48 aaronpk You have the benefit of knowing the intricacies of these plugins. This is all very confusing to an outsider
# 00:49 aaronpk Right now all I need is to somehow verify I've imported the likes reposts and comments correctly, I don't care what they look like right now
# 00:49 GWG aaronpk: Agreed. I'm too close to the problem.
# 00:49 GWG Then leave Semantic Comments off.
# 00:52 GWG aaronpk: My latest philosophy is that anything I send to the WordPress.org repository should be good enough for someone other than me to use. Anything not there yet...
# 00:52 GWG Either I haven't gotten to or isn't quite at that stage.
j12t and snarfed joined the channel
# 01:24 GWG I'm trying to figure out if I can improve my display
# 01:25 GWG Right now it goes, "Verb Article Name/Thing by Author (Domain Name)
# 01:25 GWG Trying to decide if there is a better one.
# 01:27 Loqi acegiak: GWG left you a message 1 day ago: Remove mf2_s_posted_by(); from entry-header.php
# 01:27 Loqi acegiak: GWG left you a message 10 hours, 5 minutes ago: I need a beta tester
# 01:27 GWG acegiak: I was hoping you might stop by.
# 01:30 GWG acegiak: I did a rewrite I'm still testing on Github, that goes to the mf2_ key standard snarfed proposed.
# 01:30 GWG But I want to add event and RSVP if I ever get to it.
danlyke joined the channel
# 01:33 aaronpk GWG: are comments from twitter supposed to show as "___ mentioned this Article on twitter.com"?
# 01:33 GWG It should show the actual comment content if it is determined to be a reply.
# 01:34 GWG Semantic Linkbacks turns anything it detects as a reply from type webmention to type comment so it displays like one
# 01:36 acegiak GWG, snarfed: can someone explain the rationale behind that proposal? it seems like it's being harmfully specific
# 01:37 acegiak using the mf2 prefix suggests that response data can/should only be read/displayed in mf2 format which is not true
# 01:38 snarfed i don't care what the prefix is; we just needed something for interop, so we picked one
# 01:38 snarfed it's definitely not entrenched yet; we can easily change it
# 01:38 GWG I just want a standard. Using the microformat properties does make sense over arbitrary terms
# 01:39 GWG The part I'm onboard with is _card, _cite, _in-reply-to etc.
# 01:39 acegiak additionally I don't like the enforcement of strict categories as a rul
# 01:40 GWG aaronpk: Linkbacks? It parses looking for in-reply-to.
# 01:40 acegiak at the moment in indie web kinds? because the reply kind is selected
# 01:40 GWG aaronpk: Comment Type - comment, as opposed to webmention.
# 01:41 aaronpk are "likes" and "reposts" comments or webmentions?
# 01:42 acegiak for the two categories of data set out in the proposed spec you've put the url of the author in the _cite object
# 01:42 acegiak whereas that should also probably be in the _card object
# 01:42 GWG acegiak: Yes, and if the h-card is stored locally, it would be in _card. But if it is not stored locally, I'm putting it in _cite.
# 01:42 snarfed yeah. the one thing i'm not convinced about is storing arrays in post meta
# 01:43 snarfed following the _card and _cite examples on the wiki, a normal post would store everything in mf2_entry[], not in top-level fields. which seems odd
# 01:43 GWG snarfed: Top level properties by themselves, nested ones would have to be an array.
# 01:44 snarfed GWG: right, i understand the proposal, i'm just not convinced it's the right idea yet
# 01:44 GWG I'm considering the post itself to be h-entry.
# 01:45 GWG That is why I'm working on a new version, and doing testing. I want to stabilize it.
# 01:46 acegiak is there a reason we can't do multidimensional storage?
# 01:48 GWG Not specifically, other than the general consensus to avoid more than one level
# 01:49 GWG This is why I'm debating it. I've pushed it to my development version, but I can keep changing it before the next stable version
# 01:49 snarfed and that whatever we do isn't response-specific, since the micropub plugin will write it too
# 01:49 acegiak well having two different array objects is more than one leve?
Unifex joined the channel
# 01:50 snarfed you all do more wp work than me. i've seen more flat fields than arrays in post meta, but if you all think arrays are more common, i can switch micropub to that
# 01:51 GWG I'm in favor of arrays, but not multidimensional ones
# 01:51 acegiak snarfed: micropub should call actions hooks that indieweb post kinds and other plugins should then hook into
# 01:52 GWG acegiak: The 'raw' data, so to speak
# 01:53 acegiak GWG: I don't mind having a single array for storing response data like we have now
# 01:54 acegiak but I don't understand the reason for having two separate ones and why that's different to a multidimensional array?
# 01:56 GWG acegiak: I was thinking of it because I'd like to someday switch to pulling in author data from the source.
# 01:56 GWG But you never know if that will happen.
# 01:56 acegiak actually, if we did muldimensional we could do response[0]['author']['name'] we could have multireplies
# 01:57 GWG acegiak: Threaded responses is a possibility
# 01:58 acegiak GWG: nono, like, if you want to respond to two posts with one post
# 01:59 GWG That would require a bit more finesse in the Post UI
# 02:00 acegiak GWG: yeah but that's a situation where the object storage might actually get in the way
# 02:00 acegiak whereas I don't see how you can't pull author data directly from the source with the single array system
# 02:01 acegiak I'm basically arguing that greater flexibility is better for understructure like this
# 02:03 GWG So, you are suggesting key[0], key[1] for multi-response?
# 02:04 acegiak theoretically it should actually be key[0]['author'][0]['name'] in case posts have multiple authors
# 02:05 snarfed micropub itself actually falls down at that right now, only syndication and category can be multi-valued
# 02:05 acegiak if we're not doing this stuff then we may as well be using flat meta fields
# 02:05 snarfed (+1 to the use case, at least. i'm agnostic of storage details)
# 02:05 GWG I'm in favor of multiple values per key.
# 02:06 GWG I would suggest that if there is only one, it be stored as a single level array
# 02:07 GWG Then check to see if it is a multidimensional array, and if so, change behavior.
# 02:07 acegiak than just knowing it's going to be an array and just grab the first one?
# 02:08 acegiak at the moment I think we should store in a multivalue array and then, while we're just dealking with single replies we can just pull the first entry from the array
# 02:08 GWG So, we go back to the properties.
# 02:11 GWG I'm in favor of using cite as a key, to be consistent with microformats.
# 02:13 GWG Which would mean a nested h-card.
# 02:13 GWG ['cite']['card']['name'] for the Author Name
# 02:14 GWG ['cite'][0]['card']['name'] for the Author Name
# 02:14 GWG Oh, yes. Multi-author for single citation
# 02:15 GWG So, should the mf2_ prefix be kept?
# 02:15 GWG I don't think there is a better one.
# 02:16 acegiak but mf2_ lets everyone know what the subkeys will be
# 02:16 GWG It basically means we're using the mf2 property names.
# 02:20 GWG I've written the migration function. This time, it does it a bit differently. It will take a while to write the display functionality. I started on it, but you are both right about multiple authors/replies.
# 02:21 GWG I just don't know how I'll do the data entry.
# 02:21 acegiak GWG: assume only one for now as far as data entry
# 02:21 GWG The only way to continually add post UI metaboxes is javascript.
# 02:21 GWG acegiak: That is what I'm going to do.
# 02:23 aaronpk i'm having a heck of a time with wordpress' duplicate comment detector
almereyda_ joined the channel
# 02:24 GWG aaronpk: Not sure. Never researched that.
benwerd joined the channel
# 02:24 aaronpk it makes the script terminate immediately and outputs text
# 02:24 aaronpk which is not ideal for trying to return a JSON response
KartikPrabhu joined the channel
# 02:44 aaronpk oh god lol my local dev wordpress is sending webmentions oops
tantek joined the channel
# 02:46 aaronpk and the domain is "case.wp" so at least all the webmentions i'm sending are invalid
# 02:46 GWG acegiak: Github version using new data structure.
# 02:49 aaronpk i'm adding them all to the media library attached to the post
# 02:49 aaronpk but if I re-run the script I don't want to add it again
# 02:49 aaronpk (i already have it checking for duplicate posts and comments)
# 02:50 tantek did anyone answer or catch cphyc's question about markdown for Known events?
# 02:50 tantek it's an interesting question, and I don't have an answer either. There's no standard microsyntax / picoformats for events AFAIK. Though it would be cool to research one.
# 02:52 Loqi Ok, I'll tell them that when I see them next
# 02:52 tantek !tell jonnybarnes re: wikipedia links, feel free to update any wikipedia links in the indiewebcamp wiki to their rel=canonical locations.
# 02:53 Loqi Ok, I'll tell them that when I see them next
# 02:54 tantek !tell GWG were you able to get Pierre-O to document his actual use-case that he was trying to solve? Or was he stuck on plumbing-centric (OPML) thinking?
# 02:54 Loqi Ok, I'll tell them that when I see them next
# 02:54 GWG tantek I'm right here. And yes. I was.
# 02:54 Loqi GWG: tantek left you a message 22 seconds ago: were you able to get Pierre-O to document his actual use-case that he was trying to solve? Or was he stuck on plumbing-centric (OPML) thinking?
# 02:55 Loqi hardcodedpathsforfilediscovery has -1 karma
# 02:55 tantek GWG, where on the wiki did you get Pierre-O to document his actual use-case he was trying to solve?
snarfed joined the channel
# 02:56 GWG tantek: I did not get that far, to be honest
# 02:57 GWG I was answering yes to the second question
# 02:57 GWG I was trying to get him to document a use case.
# 02:57 GWG But I hadn't gotten him off plumbing before he left.
# 02:59 GWG I think the wiki entry Pierre-O did not establish anything useful.
# 03:00 tantek GWG, thanks for trying to get him to uplevel his thinking
# 03:01 tantek it defined a piece of jargon and plumbing, so that's good at least
# 03:01 GWG I was about to this morning, but then my internet connection went down and I completely forgot until you mentioned it now
# 03:02 tantek.com edited /OPML (+293) "stub, dfn, xml based, defacto standard, subsections for braisntorming, root follow file proposal, see also" (
view diff )
# 03:03 GWG Not sure. He mentioned it as well.
# 03:04 gRegor` "Short-blogging with no text limit. Curation and web survey. Cool automatic text layout. Forums under each post. Advanced and automatic thematisation." from seenthis.net. Their own site isn't very helpful up front, haha
# 03:04 gRegor` Buzzword generator site?
KartikPrabhu joined the channel
# 03:07 gRegor` It looks like a cross between a tumblr and a forum.
# 03:07 tantek gRegor`: feel free to take a stab at fixing the page as you see fit!
# 03:07 gRegor` Busy working on my site :)
# 03:09 GWG I think I will contemplate my next move...after I shovel some snow.
# 03:19 aaronpk i still have to figure out how to not import duplicate photos
# 03:31 aaronpk i think i'm going to set the image title to the original URL and then check if there is a post with that title already
# 03:31 aaronpk not ideal, but that's apparently the best i can do given the limitations of media_sideload_image
benwerd joined the channel
# 03:51 GWG Right now my display goes, "Verb Article Name/Thing by Author (Domain Name)
# 03:51 GWG Trying to figure out if something better is an option
# 03:53 aaronpk huh i appear to have missed something with getting these to show up as replies
# 03:53 GWG Example: Liked Ain't No Sunshine by Bill Withers (last.fm)
# 03:56 GWG aaronpk: What is the problem exactly?
# 03:58 aaronpk setting "comment_type" to "comment" vs "webmention" seems to have no effect on the display, it always shows "___ mentioned this article on..."
# 03:58 aaronpk i believe i'm setting the data internally correctly
# 03:58 GWG Did you set the semantic_linkbacks_type to reply?
# 03:59 aaronpk but now it says "___ replied to this article on ___"
# 04:04 aaronpk okay my wp_comment table has comment_type=comment, the wp_commentmeta table has a property semantic_linkbacks_type=reply for that comment
# 04:12 GWG Remove the comment type entirely?
tantek, addal and almereyda joined the channel
# 04:58 snarfed aaronpk, your migration is flooding bridgy a bit :P
# 04:58 snarfed load wise i don't care, but the wm source urls are wrong, so they're failing
# 04:58 GWG snarfed: But it is interesting to watch aaronpk try to cope with WordPress.
# 05:00 kylewm why would migrating to wordpress send stuff to bridgy?
# 05:01 aaronpk i'm creating posts in wordpress which sends webmentions
# 05:01 aaronpk is there a way to disable that? oh guess i could turn off the webmention plugin during import
# 05:02 snarfed except these are the response/interaction urls *from* the silos, ie the ones bridgy sends back
# 05:02 snarfed so kylewm is right, i'm not sure why it's happening
# 05:02 snarfed maybe the wm plugin is triggering on comments too
# 05:03 kylewm you mean like source=case.wp...&target=brid-gy.appspot.com...?
# 05:03 GWG snarfed: It should trigger on comments.
# 05:04 snarfed kylewm: actually the opposite, target is a wp post, source is e.g. brid.gy/repost/twitter/caseorganic/430437892742078465/430439869505372160
# 05:04 snarfed the twitter ones specifically are rate limited and failing
# 05:05 aaronpk noted, and will disable when I do the real import
# 05:05 GWG aaronpk: Your importing has revealed a lot of useful things though.
# 05:05 aaronpk otoh this would be an effective way of notifying all the things she's linked to about the new URLs
# 05:06 aaronpk aw man so I broke bridgy for everyone until the rate limit is lifted?
# 05:11 aaronpk right but that's the same account bridgy uses for everyone else's webmentions too
# 05:12 snarfed these aren't outgoing wms, they're fetches of source urls in received wms
# 05:23 acegiak huh, neat. tweeted at local news outlet asking if they knew why the poice were sending a small army down around my hometown and they responded saying they'll try and work it out
# 05:26 GWG Not the tweeting, the small army.
# 05:27 acegiak probably a raid of some kind or they'd issue a warning
# 05:27 acegiak but the question is, how can I get the news outlet to accept indieweb comments on their articles
# 05:33 GWG acegiak: News outlets are very nervous about commenting in general.
snarfed joined the channel
# 06:15 GWG Trying to figure out what is next.
# 06:28 acegiak GWG: so should I start backing up my db to try switching to this new structure
# 06:31 GWG acegiak: The new structure is in the Github repo. The display works pretty much the same, but the display routines are rough and need more refinement.
# 06:33 GWG Now that it is multi-dimensional, I need to rewrite them again.
# 06:39 GWG But, either way, I have to rethink what I did before.
# 06:41 GWG I hope so. But I keep trying to make things better
# 06:41 acegiak yeah but the new structure is based on flexibility
# 06:41 GWG I don't want to redo the data structure, but I keep trying to improve the presentation.
# 06:41 GWG I still need to figure out a good way to hide/add meta boxes.
# 06:42 GWG I keep trying to avoid learning javascript.
# 06:43 acegiak ESPECIALLY because hiding metaboxes would shoot me in the foot without multikind posting
# 06:43 GWG I was thinking of the sort of javascript to add extra boxes for additional authors and URLs.
KevinMarks_ joined the channel
# 06:46 GWG The only other thing I was considering was showing/hiding a box for check-in data.
# 06:46 KevinMarks_ I had a random thought about using css for h-feed reading and disclosure. Style all elements hidden, then show h-entry p-title, on hover show p-summary
# 06:46 GWG KevinMarks_: I love using CSS to avoid Javascript.
# 06:47 KevinMarks_ Not sure if css has enough options to make this work without js entirely though
# 06:47 GWG KevinMarks_: There's something called the checkbox hack that works pretty well, but only for adjacent elements.
# 06:48 KevinMarks_ There is probably a devious way to use :visited to track link clicks that are local ids
# 07:00 KevinMarks_ I like the idea of a "feed view" that is just a stylesheet that hides everything not in h-feed/h-entry
# 07:02 KevinMarks_ Then I do an unmung.com variant that just knows which elements to add h-entry etc to on a site (eg ello)
KevinMarks__, wolftune, KevinMarks_, frzn, Acidnerd, eschnou, catsup, KevinMarks, sanduhrs, elima_, loic_m, cweiske, iandevlin and interactivist joined the channel
danfowle1 and sanduhrs joined the channel
# 15:16 GWG aaronpk: We're all pulling for you
snarfed joined the channel
# 15:16 aaronpk do you know of a way to disable the duplicate comment check?
# 15:17 aaronpk two different people named "Alex" liked one of amber's photos
# 15:19 GWG remove_filter('commentdata','comment_duplicate_trigger'); - Try ths
# 15:19 aaronpk hm looks like it de-dupes also on email address, which I am not setting obviously
almereyda joined the channel
# 15:20 aaronpk what would happen if I set the author URL as the email address?
reidab, ShaneHudson, GWG, adonaldson and AMF joined the channel
# 15:26 aaronpk is resorting to commenting out the duplicate check in the core code
elf-pavlik joined the channel
# 15:32 aaronpk is there a place I can file an issue against wordpress to ask that they add a way to disable that check?
LauraJ, KevinMarks and frzn joined the channel
GWG joined the channel
# 15:51 GWG Could it be the flood control, I noticed this in pfefferle's code.
# 15:51 GWG Thought he might have had the same problem
# 15:52 aaronpk i disabled the flood control when I found that code
ben_thatmust_ and ben_thatmust joined the channel
# 15:58 aaronpk i'm getting around this by adding a passive-aggressive disclaimer on my plugin screen describing how to edit the core wordpress files
tantek and danlyke joined the channel
# 16:01 GWG aaronpk: There is a 3 year old open ticket on this issue
Deledrius joined the channel
wolftune joined the channel
# 16:11 aaronpk okay last step is to make sure all these redirects work
# 16:12 aaronpk btw the plugin generates .htaccess rules for redirects from the old URLs to the new wordpress URLs
# 16:28 aaronpk my suspicion is most people who run wordpress run it on apache
# 16:29 aaronpk btw this plugin is by no means perfect, but it's a good start and I hope someone else tries it out at some point and sends a PR for stuff
LauraJ and almereyda joined the channel
# 16:33 rascul easy way to thrash your disk and slow down your site :)
# 16:35 aaronpk yuou can copypaste the rules into httpd.conf if you really want, same syntax
# 16:38 rascul troubleshooting can get messy quick when you have multiple levels of htaccess
KevinMarks, upper-- and kylewm joined the channel
wolftune and KevinMarks joined the channel
# 17:32 KevinMarks So with the WordPress mf2 template and that importer, how close can you get to replicating a wp blog completely?
benwerd joined the channel
# 17:34 GWG I know a few things still missing
# 17:34 KevinMarks Also can you import into known from wp, then import back and not lose any content?
almereyda joined the channel
# 17:34 aaronpk this importer made a complete copy of caseorganic.com by reading *only* the HTML
# 17:35 KevinMarks It's a good case to see where we have gaps in mf2 representation
# 17:36 GWG KevinMarks: I need to work on a few things for my starter theme and hope people make themes based on it. I have one so far, and acegiak is using a fork of it as well I think.
# 17:37 aaronpk GWG: caseorganic is going to need an mf2 compatible theme because the default wordpress one is not
# 17:38 KevinMarks Right, that was the next thought - if you ran it on the wp site would it make another good copy
# 17:38 aaronpk as long as the theme marks up comments/likes/reposts properly yes
eschnou, kerosene and michielbdejong joined the channel
# 17:48 aaronpk "Cannot redeclare mf2_plugin_notice()" apparently i can't use that function name in the mf2 importer
# 17:50 GWG We are both using the same prefix
# 17:52 aaronpk "Fatal error: Call to undefined function get_linkbacks_number()"
# 17:55 GWG aaronpk: Back to the Semantic Linkbacks problem.
# 17:56 GWG I intend to push pfefferle on that. If not, I'll have to add the function to the theme for backward compatibility.
# 17:56 GWG aaronpk: I pushed it upstream because I thought it should be.
wolftune and KartikPrabhu joined the channel
# 18:11 GWG I feel inspired to marshall the forces of the WordPress Outreach Club.
snarfed joined the channel
yakker, snarfed, LauraJ and KartikPrabhu joined the channel
# 18:49 snarfed even with a token that *has* read_stream, 2.x /me/home only returns public posts, and not even many of those
# 18:50 snarfed i really don't want to go back to reading fb inside fb again
wolftune joined the channel
# 18:54 snarfed actually no i'm wrong. the one post i get from /me/home isn't public. it just doesn't give me any others
KevinMarks_ and gRegor` joined the channel
# 19:07 kylewm thinking about just deleting my facebook so I can ignore its API issues :P
# 19:08 GWG kylewm: Hiding from the world isn't a solution
# 19:10 rascul i want to delete my facebook account, but there's a few people i must still use it to communicate with :(
# 19:12 GWG I still have AIM running using BItlbee because people refuse to leave it
eschnou joined the channel
# 19:14 kylewm to be clear, i meant that tongue-in-cheek... i would be happier not to have to deal with the FB API anymore
jacus, j12t and snarfed joined the channel
# 19:28 KevinMarks Is it worth writing up these fb api closures to explain to journalists?
# 19:35 GWG The XMPP one was an issue no one seemed to pick up on
eschnou, benwerd_, benward______, tjgillies, CaptainCalliope_, uranther, Garbee, bigbluehat, amblin, GenioDiabolico and elf-pavlik joined the channel
lmorchard joined the channel
# 19:57 Loqi jonnybarnes: tantek left you a message on 2/21 at 6:53pm: re: wikipedia links, feel free to update any wikipedia links in the indiewebcamp wiki to their rel=canonical locations.
JonathanNeal, halorgium, GWG, Deledrius, bret, loic_m, hadleybeeman, wilfredh, voxpelli, prtksxna, sanduhrs, friedcell and marclaporte joined the channel
# 21:56 GWG KartikPrabhu: What are you thinking?
almereyda joined the channel
# 22:04 GWG Ah, Snapchat...I have a choice between putting my data on a silo or a silo that deletes it forever immediately.
wolftune joined the channel
awolf joined the channel
lukebrooker, sammachin and benwerd joined the channel
snarfed, addal, frzn and KartikPrabhu joined the channel