#tantek.comdeleted /credentialing "content was abstract concept with no current relevance to the IndieWeb, no active IndieWeb examples or brainstorming towards active implementation"
#GWGIf I have it redirect back to the same URL that shows the authorize page, with an extra parameter that bypasses the screen and instead, just redirects to the code, is that an issue as someone could try to force that?
#skippyi may be in the minority here, but I find the loqi content in here very disruptive to the conversation. :( i dont want to completely ignore loqi, because then i might miss out on !tell messages left for me. :(
#skippybut all the wiki edits are less-than-useful.
#aaronpkWe could talk about moving all wiki edits to meta. Right now just the event edits are in meta
#skippyi'm the new guy here, so i dont want to try to overly impose my own preferences. I just find it hard to follow a thread, or read a back scroll, while trying to manually visually filter out all the loqi content spread between the humans chatting.
#skippyi dont idle in meta, so moving it all over there would be fine with me; until sucj time as I have a meta conversation to have and need to chat there ;)
#ZegnatYou mean a meta chat like the discussion about what wiki edits should show in which channel? ;)
#kaushalmodiI agree with skippy. Ironically I just made few edits. Would be nice if they ended up in meta.
#ZegnatI wonder if we should add a note to the wiki interface that checking “minor” makes an edit not show in chat.
#Zegnat(sknebel should Kaja do its autolinking etc as a minor edit?)
#skippykaushalmodi: why did you select hugo for your site generator? curious to learn what other people's motivations are for the tools they select.
#sknebelZegnat: could do, since I think it can be trusted now - early on it was intentionally not, to catch mistakes
#sknebel(if your client has good filtering features, you probably could exclude the wikiedits only if you find them really annoying)
#skippyyes, I could learn how to filter just the wiki edits via irssi. i can do that, if the consesus is that most people find real value in the edit announcements here.
#skippylike i said, I dont want to force my preferences. just expressing them right now, to see if i'm the outlier.
#ZegnatNo, I think it is a good discussion to bring up skippy! Especially if we start to more strictly refer conversations from #indieweb to here, it is important not to break up the flow.
#ZegnatThat is one of the reasons we limited Loqi output in #indieweb in the first place :)
leg joined the channel
#aaronpkspeaking of which, this conversation should take place in -meta :)
[miklb] joined the channel
#skippywhen POSSEing notes to twitter, do folks generally capture the Twitter URL after it's been posted, for inclusion on their original copy?
#aaronpkI have little icons that appear next to the timestamp of my post with all the syndication links
#skippythat leads to another question I've had about all the indieweb markup, generally.
#skippyall this markup is to make things machine-readable and parsable for various uses. I'm curious what the use case is for machine-consuming all the syndication links, etc...
#aaronpkhelps with deduping posts when they are received at the other end
#aaronpkif you look at some replies on my posts, they will also have syndication link icons
#aaronpkso if I get a webmention from an indie URL and it says it has a twitter URL syndication, then when I get the bridgy webmention for that I can match it up and avoid showing the reply twice
#skippykaushalmodi: why did you select hugo for your site generator? curious to learn what other people's motivations are for the tools they select.
#GWGaaronpk, I understand that the token endpoint shouldn't accept codes without scope, which would prevent response type id from issuing a token. But how does the authorization endpoint do anything?
#aaronpk"If the authorization code was issued with no scope, the token endpoint MUST NOT issue an access token, as empty scopes are invalid per Section 3.3 of OAuth 2.0 [RFC6749]."
KartikPrabhu, tantek and kaushalmodi joined the channel
#kaushalmodiskippy: About why I chose Hugo.. I had got burnt by trying to get Octropress installed, and then stay away from Ruby stuff ever since. Then about a year back, I came across Hugo on one of the HN posts, and the "one static binary" really appealed me. Additionally installing the Go toolchain was simply extracting an archive and putting that in your PATH. With Hugo, I simply upload the static built binary of its Dev version and build my site usin
#kaushalmodi.. build my site using that. The benefits of awesome speed, community support, awesomeness of the Go templating which resembles sort of to Lisp got realised later, and I am glad I picked up this tool.
[kevinmarks] joined the channel
#[kevinmarks]The speed is a big thing, but you usually only see that if you're importing an already big site.
#skippyimporting all my tweets into my site has raised my build time from < 1 second to ~4 seconds on my macbook; and ~20 seconds on my server.
#tantekwow ok that may explain that. individual file overhead tends to be an I/O (thus perf) bottleneck
#skippy20 seconds to generate 15K pages is pretty darned fast, so I can't complain too much. :) Either spend resources generating each request on demand, or spend resources generating the whole site with each occasional addition of new content. :shrug:
[kevinmarks] joined the channel
#petermolnar20 seconds for 15k pages is close to insane
#skippyhavent decided if i actuall WANT to display all my tweets that way or not.
#snarfedskippy: many of us keep tweets (etc) on our sites but de-emphasize them by just not showing them on the front page
#tantekit's not either or, there's a third option which is lazy regeneration, regenerating pages only upon first request, and then caching that static file
#tanteksince clearly it's easy enough to make that error
#tantekhad a thought re: SWAT0 and IWS, how about trying to get micro.blog's indieweb support to the point where it can place at least one of the A,B,C roles by IWS?
#tantekpetermolnar: re: https://news.ycombinator.com/item?id=16764602 in particular this is one way fail to ever produce a spec: "... I suspect we're better off thinking about all possible features, because this allows us to define a protocol that generalizes better."
#Loqi[amelius] Emphasis should be on the requirements.
What do we want the system to do?
And how can we control our privacy?
Let's answer these questions first.
#tantek"thinking about all possible features" = you will never ship
#tantek"what's a minimum feature set for a social network"? Twitter 2006
#LoqiA person tag (AKA people tag) is a person mention that is also a tag on a post that refers to a specific person by URL rather than just a word or phrase, and is done as an explicit tagging action by the user, beyond just mentioning a person via hyperlink / h-card / or @-name, autocompleted or not https://indieweb.org/person-tagging
[jgmac1106] joined the channel
#tantek2 min timebox expired on yet another handwavy Hacker News discussion of decentralization.
[snarfed], chrisaldrich, snarfed and barpthewire joined the channel
#@spigot↩️ This is the output and format from http://Brid.gy. Not so good. The link is there but what the heck is the rest?
I’m working on getting webmentions working on the site, this is a bit of a setback. (twitter.com/_/status/981957032537305088)
#GWGWhat is the code for a redirect_uri not being on the same host or whitelisted?
#LoqiIt looks like we don't have a page for "code for a redirect_uri not being on the same host or whitelisted" yet. Would you like to create it? (Or just say "code for a redirect_uri not being on the same host or whitelisted is ____", a sentence describing the term)
#aaronpkmissing parameters are invalid_request, invalid stuff in the parameters are invalid_grant
#[jjdelc]Hey guys, I've been working on this MP client for the last couple of months https://grumble.isgeek.net/ I've been using it with my (also homebrew) mp server
#LoqiCORS is an acronym for "cross-origin resource sharing," a mechanism for allowing browsers to make JavaScript requests to fetch resources from other domains https://indieweb.org/CORS
#[jjdelc]The client does need CORS to be enabled on the MP endpoint though
#[jjdelc]but if it's hanging on "Initializing" that's a different issue
#[jjdelc]I made a change, looks like I didn't have `vue.js` registered with UpUp for it to be available as offline resource
#[jjdelc]btw, the Micropub spec says nothing about CORS on the micropub endpoint, for my client to be full client side, I needed to enable CORS on my backend
#ZegnatI’d say the Micropub spec expects a client to have the ability to access resources over HTTP. JavaScript in browsers doesn’t do that. Not really a spec problem.
#ZegnatThough can you catch the CORS error? Maybe display an error to the user that they need CORS enabled on their endpoint for this to work?
#ZegnatHmm. And probably not only the endpoint either, but also on the page where you run endpoint discovery. As well as on the IndieAuth endpoints. Yikes.
#[jjdelc]I will not discover that until after successful login
#skippyi know nothing about JS. just sharing what Firefox dev tool console is showing me.
[gerwitz] joined the channel
#[gerwitz]Does anyone have a favorite means for deciding a Micropub post is “photo” (e.g. from OwnYourGram) instead of “note that contains a photo” (e.g. from Micro.blog clients)?
#grantcodesYeah source maps won't cause issues (still good to fix though)
#grantcodesThe issue I'm having is with my own token endpoint returning an error for some reason
#reliable.servesarcasm.comcreated /User:Reliable.servesarcasm.com (+528) "Created page with "<h3>Hi!</h3> <p> I'm Dylan Harris. I've had a website up for a couple of decades. It's all my own cat food. Don't go visiting it, it's running on an Intel chip so old it's im..."" (view diff)
#LoqiPost Type Discovery specifies an algorithm for determining the type of a post by what properties it has and potentially what value(s) they have, which helps avoid the need for explicit post types that are being abandoned by modern post creation UIs https://indieweb.org/post-type-discovery
#grantcodesThe other issue you'll run into with a js client is you can't read headers, some people may have their endpoint links in headers.
#KartikPrabhuha! yeah just saw that on my reader too
#skippyaaronpk: you advised earlier to capture the twitter URL of POSSEd content, and link to it from the original with u-syndication. Curious about workflow: do you post the original content to twitter first, then capture that tweet url, then include that in your content that you post to your site? or do you post to your site, submit to twitter, then edit the original?
#KartikPrabhuI suspect the later. which is also what I do
#aaronpkmine is all automated, so I first post to my website with a micropub client, my site stores the post, then it POSSEs to whatever places were selected, and when it does that it writes the URL back to the post when it's done
#skippygotcha. thanks. for a database-powered site, I see the utlity in that. for my current efforts to use a static site generator, I'm leaning toward the other model, to minimize local writes / regenerations.
#aaronpkmine isn't database-powered, it's just generated on the fly
#[kaushal_modi]skippy: I do the same (publish the site twice): 1. Publish site 2. Post to twitter, get the tweet id 3. Add id to the post front-matter and republish
#bearskippy: I also do the same, I maintain all webmention, likes and other items as secondary files containing the info. If those secondary files are touched, I rebuild the articles associated with it
#KartikPrabhualso don't think file vs database is that different in this case, is it?
#bearI don't think so - it's more of: where do you store the metadata for things and can it be updated with changes propogated to your generated output
tantek and [gerwitz] joined the channel
#[gerwitz]@zegnat and [aaronpk]: thanks, I needed a refresher on some obviousness, there!