#Loqiquestion is a post type for soliciting answer replies, which are then typically up/down voted by others and then displayed underneath the question post ordered by highest positive vote count rather than time ordered https://indieweb.org/poll
#@AJKandy↩️ Just like that? Nice. Then may I suggest adding a handy preformatted link to Tweet-vote each suggestion? As a user, I care about accomplishing the task of voting; I do not care about the back-end mechanism, whether it be webmentions, the electric telegraph, or carrier pigeons. (twitter.com/_/status/1227022992888086528)
#aaronpk> We don’t want to require you to have an account on this site because (frankly) we have no interest in your personal information. We also want you to own your vote, so we opted to use webmentions as our voting system.
#aaronpkYou can bookmark or like a page and send a webmention and that counts as a vote!
#@aaronpk↩️ This is a very cool use of webmention! Can I make a suggestion to add some minimal h-entry microformats so that when someone wants to show the page they liked it can show more than just the URL? Here's an example of what it looks like on my site right now. https://aaronparecki.com/likes (twitter.com/_/status/1227026004171870208)
#@AJKandy↩️ I didn't say that webmentions were bad per se. (Even if it feels like Sitemeter 3.0, but w/e) I'm saying that if you want to get feedback from people who aren't coders, make it easier for them to contribute. Lower the barriers. If you can tweet to vote, then add a button! (twitter.com/_/status/1227026579886206981)
#aaronpkoh no it's using webmention.io to receive webmentions haha!
#[jgmac1106]aarongustafon has his finger prints all over this
#[jgmac1106]its awesome, but how are they getting webmentions from facebook button?
#@AJKandy↩️ My point is, the instructions on how to participate, to a non-coder, felt complicated and/or intimidating. Webmentions is a mechanism that I, as a user, shouldn't really have to understand or even know about. Just tell me "Tweet this URL." That's all. (twitter.com/_/status/1227028002673815557)
#aaronpkThis looks like a good opportunity to update our docs on the pages they link to
#[jgmac1106]for stuff I know, I have to learn to implement, receive, and display...well actually they aren't even displaying just collecting in webmention.io
#[jgmac1106][aaronpk] just fork it and file a PR with the microformats
#aaronpkI dunno if I know how... it's using Eleventy and I don't know anything about that
#[jgmac1106]ohhh yeah I just looked everything is in markdown
#Loqi[Greg McVerry] I want a read AND write browser I should be able to connect a browser to my website, then go up to File, select new, choose my post type and hit publish. I should be able to react and publish to my site from pages I visit.
#aaronpkDid you submit it to their site via their markdown thingy?
#aaronpkI don't think they take ideas via webmention, just votes. Tho that'd be cool too
lalilulelo joined the channel
#[jgmac1106]They are going to get the webmention regardless if they accept ideas that way
#[jgmac1106]yeah...but it was just a sentence...I poked around the layouts, components, and pages to see it, and left with a reminder of why I like reading HTML
#[jgmac1106]that reminds me should bridgy represent a reply to another website in a Tweet? you do lose the reply context..I should have replied to the tweet as well
#Loqi[snarfed] hi! yes, that's right. bridgy currently only allows POSSEing a reply if it replies to a post on the same silo. this is intentional; discussion in #362. from that issue:
> people also often try to use bridgy publish to POSSE a reply to a tweet but ...
[KevinMarks] and jeremycherfas_ joined the channel
#[jgmac1106]I knew you would reply with the issue that probably ends in keep bridgy stupid..havent read it yet
#@dmitshurI’m aware that the server hosting my personal website is down right now. I expect the service to be restored in a few hours. If you’re looking to fetch modules from it, you can use the Go module mirror at https://proxy.golang.org in the meantime. Sorry for the inconvenience. (twitter.com/_/status/1227000562669015040)
#Loqisknebel has 19 karma in this channel over the last year (58 in all channels)
#jeremycherfasOn the question of e-content vs e-summary, at the moment I have a single div that displays the summary if summary!=content and the content if summary==content. That div has class e-content.
#jeremycherfasI'm going to try adding a div inside the logic with the correct class.
#jeremycherfasThere's not much I can do about the h-card being on every item in the feed without a major re-write of the twig template partials.
#jeremycherfas!tell sknebel Re "the author is only a child, not an author property" I wonder whether that is because my h-card contains `<data>` elements. I don't know whether that would make a difference to how it is parsed.
#jeremycherfasOK, wait. I have put the h-card on every page. That works much better. Thanks for the inspiration.
#jeremycherfasI think when we discussed it a few months ago I got the impression it didn't matter because parsers ignore things they do not understand.
#ZegnatI am not seeing any author information though, at least not on the tag feed page.
#ZegnatIndeed. Just means I need to know to ignore it when looking at the parser output of your site ;)
#ZegnatNo, looks like for the discovery algorithm to wotk, the h-entry needs to have an author property that is a link to your homepage. Then the h-card can be used given it too links to your homepage with its url property.
sscarfe, gxt, [xavierroy] and [Michael_Beckwit joined the channel
#jeremycherfas!tell Zegnat I've used the empty link approach `<a class="u-author" href="/"></a>` to point to the h-card on my home page, and as far as I can tell that now works on tag pages, post pages and the home page. But is an empty link of that sort bad practice in any way?
#Zegnatjeremycherfas: I would say it is somewhat bad practice because browser may still let you keyboard-navigate to it and it may be read by screenreaders.
#LoqiZegnat: jeremycherfas left you a message 1 minute ago: I've used the empty link approach `<a class="u-author" href="/"></a>` to point to the h-card on my home page, and as far as I can tell that now works on tag pages, post pages and the home page. But is an empty link of that sort bad practice in any way?
#jeremycherfasI know I've done this dance before, which was why I ended up with `<data>` elements in an h-card. But I have p-name in that card. Could I have p-author? I don't think that would help.
#ZegnatThat way the h-entry will have an author property linking to your homepage
#jeremycherfasWhat, precisely, would I want inside the articles? That's the bit I'm not getting. There's no need to display me as author because I am the only author. Gregorlove used to have his h-card invisible, but no longer does.
#jeremycherfasWhen I just had the h-card inside the articles, it was not being parsed as an h-card. So do I need to add something to it to ensure that it is seen as an h-card?
geoffo, KartikPrabhu and [jgmac1106] joined the channel
#[jgmac1106]Tumblr experimental feture that automatically replaces your words with emoji. In the screen shot you see the word OK and key becoming an emoji
#Zegnatjeremycherfas: if you have that HTML as part of your h-entry, all you needed was a u-author class on the <div class="h-card">
#jeremycherfasI thought u- had to be on something with a url, rather than on something that might have a url nested in it.
#ZegnatSo whether you use u- or some other mf2 prefix is actually up to you. It defines the fallback for applications to use in case they do not understand the nested h-card
#Zegnatif you use p-author it will take the p-name from h-card as the fallback. I just like to use the URL as fallback.
#LoqiZegnat has 15 karma in this channel over the last year (56 in all channels)
[grantcodes] joined the channel
#sknebelYou could also do the "empty link" with a days element if you wanted btw
#Loqisknebel: jeremycherfas left you a message 4 hours, 13 minutes ago: Re "the author is only a child, not an author property" I wonder whether that is because my h-card contains `<data>` elements. I don't know whether that would make a difference to how it is parsed.
#sknebelI think that tell has been answered - it wasn't marked as the author + with p-author or u- author)
#[snarfed]oh [jgmac1106] you're trying to use a fragment to instruct granary (and maybe other tools) to use just a specific part of a page?
#[jgmac1106]or if there is another way to say just convert this specific h-feed on the page let me know
#[snarfed]that's unusual. granary doesn't support that, nor does bridgy. i don't know that it's an established technique, or that mf2 parsing etc have specific rules on handling fragments specifically
#[snarfed]i definitely get the use case, it's just not a standard technique that tools widely support yet afaik
#Loqi[OliverJAsh] #681 listen: support hash/fragment permalinks to entries inside feeds
#[jgmac1106]that issue is a sepeate issue with Bridgy, I wish it would recognize fragements for pages with short notes so I could do one file rather than having a permalink for each entry
#[snarfed]right. i gave them as examples that this broader idea, directing tools to do things with specific elements of pages, has been requested
#[jgmac1106]I was mistaken though and thought I could use granary with fragments to specify an h-feed
#[snarfed]i'm not aware of many (any?) indieweb tools that actually do it right now though
#[jgmac1106]none, it was my mistake, now goes to look if he messed up all the rss feeds on his homepage.....ohhh wait only one h-feed on my homepage, all other granary links point to feeds in other plaxes
#sknebelAlthough I'd need to check which parsers actually have releases supporting that yet
#LoqiSebastiaan and I were on a philosophy bender, apparently. One of the concepts to
come out of it is the head cache. When you have filed something in the back of
your mind for future use.
The problem, just like with a computer cache, is that ...
[datashaman] joined the channel
#ZegnatLoqi does pick up the note content, but not the author
#Zegnatjeremycherfas: author looks good on that tag feed page now! I am asuming you use the same code everywhere you render articles? Then author is fixed!
#[daveymoloney]Hi everyone. I'm an Indigenous on Android user, love it. Had been having some authorization/token issues for the past few weeks but that appears to have been resolved today. I noticed when i regained access to my account on Indigenous today that none of my syndication links are being displayed in the app when i want to post. Checked on my site and all syndication links are there, it just looks like Indigenous can't see them right now.
#[daveymoloney]Have tried refreshing Endpoints within Accounts option but that hasn't worked. Any advice much appreciated.
gRegorLove joined the channel
#jamietanna[m]!tell snarfed I should be on track to have the Bridgy PR sorted tonight (UK time) - if you're happy with it, do you expect it being live today? Or after we merge there's still a few things to do? Just thinking about documentation changes and whether I let them hold back the core code, or if it doesn't really matter
#jamietanna[m][daveymoloney] I'd recommend killing the app fully and trying again, in case it needs to restart. Also what are you using for your Micropub server? Have you tried using the Indigenous `Send access token in body` setting, in case your hosting provider is blocking `Authorization` header? I'd say its also worth raising an issue on the repo so we can dig into it a bit more
#jeremycherfasThat's kind of interesting. And a bit sad.
nickodd joined the channel
#[daveymoloney]@jamietanna[m] Thanks Jamie. I killed the app and restarted a few times but that didn't work so i removed my account and tried to log back in again but am repeatedly getting authentication error status 401 now. Aperture is the microsub server I'm using. Had been having issues for the past while authenticating to other microsub clients (Monocle and Together) but managed to login again to all successfully today so thoguht it was resolved.
#[daveymoloney]I'm with Reclaim Hosting. Will raise an issue on the repo if this persists
#[jgmac1106]JeremyCherfas yeah better to give a h/t to those who own their platform rather than calling out those who don't
#[snarfed]jamietanna hey, great! i'm happy to merge without docs, but yeah, we need them (and delete) before we can deploy to prod
#Loqi[snarfed]: jamietanna[m] left you a message 26 minutes ago: I should be on track to have the Bridgy PR sorted tonight (UK time) - if you're happy with it, do you expect it being live today? Or after we merge there's still a few things to do? Just thinking about documentation changes and whether I let them hold back the core code, or if it doesn't really matter
#jamietanna[m]Okey dokey - I'll get them all in together then 🙃
#jamietanna[m]Snarfed and once that's done, is it fairly straightforward to get it live? Hoping to get a blog post ready for when it's live, and I realise you're busy too so knowing rough timelines would be appreciated!
#[snarfed]are you expecting to do the multiplexing thing to support multiple UI flows with the single redirect url? ie signup, interactive publish, and delete?
#[snarfed]there are existing examples in the codebase, but still, that will take some effort
[LewisCowles] joined the channel
#[LewisCowles][jeremycherfas] owning your own platform without machine-consumable data is still a step further than paying Wix or hosting on facebook, twitter etc
#[snarfed]facebook and twitter are different than wix. wix is a web host that lets you use your own domain, which is key for indieweb.
#[jgmac1106]snarfed here is another reason I want granary to recognize fragments, trying to remove all my video content from YouTube and allow people to subscribe on a per show basis: https://jgregorymcverry.com/myvideos
#[jgmac1106]"Your Wix site and all of its content is hosted exclusively on Wix's servers, and cannot be exported elsewhere."
#[LewisCowles]I’ve got it in a really good place with Heroku
#[jgmac1106]....oooh excited...since I will be doing a ton of installs....
#[LewisCowles]but sadly it still needs more work. Is there a dev or contributor willing to work with me to get tickets actioned, before I lose context?
#[jgmac1106]ask in #knownchat or hit [mapkyca] directly on Twitter, he is best on Sundays though UTC-1 time
#[jgmac1106]maybe even see if the Collective would support the work, like I said I would have no problem paying anyone to get us closer to one button install....@benatwork is fiduciary responsibility
#[LewisCowles]over 200k lines deleted to get S3 working lol
#[jgmac1106]I hired [mapkyca] to help with OAuth and SSO with Known, and build a course builder plugin...A heroku based S3 install be awesome, I could maybe help fund if you need help. My goal is to be able to charge seat license and give 50% to collective
#[snarfed]to be fair that was just the AWS SDK that was oddly vendored directly into the repo
#[jgmac1106]are you talking the s3 data storage plugin or something in idno core?
#[LewisCowles]Yeah I know. Most of the other files got deleted too
#[snarfed]loc change wasn't really that meaningful 😁
#[LewisCowles]pretty meaningful as it doesn’t work without 😛
#[jgmac1106]So glad Erin convinced Ben to move away from the name Idno.......
#GWG[snarfed]: Where did Bridgy pick up link rel=feed as noted in the docs? Do you remember the genesis?
#[tantek]!tell aaronpk plz add any wiki pages you want "updated" or "refreshed" to /to-do like even with the < < operator and add the link of the page that links to them for the context of *why* they might need an update (for what new audience). and then !tell tantek /to-do#See_Also here 🙂
#[jeremycherfas]That’s for micro.blog, which is doing a 30-day challenge with a prompt per day. There’s a hashtag but I cannot remember it and do not use it myself because Known has problems with hashtags in its feeds.
#[tantek]Now checking [benatwork]’s site similarly
#[tantek]offest=560 works for him however 1000 doesn't
#[tantek]odd, different error, seeing photo stream pages with some photos missing
#[tantek]ok tracked his down. https://werd.io/content/photos/?offset=420 is the last page with all photos working. on the "older" page from that with offset=440, the last photo is not showing (and it shows alt text instead)
#Loqi[[snarfed]] hey aaronpk any chance you still have the query you ran for me back in 2017 to collect all domains that wm.io had received a mention for? would you mind running it again? not urgent!
#jamietanna[m]Hey snarfed, I'm having some difficulty with delete - looking at the network tab in Firefox even when clicking Disable it seems to go through the authZ flow with Meetup 🤔
#[snarfed]right, that's intentional, it's how we check that you actually own the account
#[snarfed]maybe_add_or_delete_source() may do the hard work for you of multiplexing signup and delete. interactive publish is maybe the one you'll have to do real work for. i assume you'll put the operation requested by the user into state?
#@6loss↩️ Aha, I should look into that - replying to posts is a tricky one currently, although a couple of us are playing with webmentions. I use an app which connects to the tt-rss server, but haven't seen anything microsubby in there. I imagine there's some overlap... (twitter.com/_/status/1227344108546613248)
#jamietanna[m]Hmm it looks like I've broken it some more. I may have to park it until tomorrow - if you're able to have a look and see if you can spot anything snarfed gimme a shout - but no worries if not
#sknebeltakev[m]: also, feel free to ask for help in #microformats if things are unclear! (the tests also might have the occasional bug, so if something doesn't seem to make sense it might not be your fault...). And once you have something, it's a good idea to put up a page like php.microformats.io so people can try it out easily
#GWGI think I need to rewrite that as an algorithm, but I fear I am on the wrong track
[Aaron_Klemm] joined the channel
#[Aaron_Klemm]Do modern IndieWeb tools require a domain name, specifically, or can it all be done on URLs? (e.g. OpenID used to work with a url, didn't require a domain). Looking at IndieAuth prompted this thought.
#aaronpkmost things don't assume just a domain, IndieAuth specifically has a bunch of language describing these kinds of variations
#Loqiaaronpk: [tantek] left you a message 2 hours, 9 minutes ago: plz add any wiki pages you want "updated" or "refreshed" to /to-do like even with the < < operator and add the link of the page that links to them for the context of *why* they might need an update (for what new audience). and then !tell tantek /to-do#See_Also here 🙂
#[tantek]it's a bit of #ownyourdata policy embedded in tools yes. if you don't own the domain then you don't actually own the data. thus requiring a domain works to encourage a more user-agency reinforcing policy
#aaronpksome tools do assume one user per domain but that's a decision they make for their own reasons