#aaronpkschmarty: at least it'd be relatively easy to export your subscriptions from Aperture manually but OPML export is probably a good idea for it anyway, especially if anyone is running a shared instance of it
#schmartyaaronpk: i think i will make a db backup right quick now that it's become my canonical reader :}
#schmartyaaronpk: also, i have a handful of feeds that are totally feeds but that aperture Just Won't Handle. i am happy to dig into that on my own if you have any pointers on where to look for parser debugging info. i'd also be happy to make a GH issue if that's preferred.
KartikPrabhu and billbennettnz joined the channel
#aaronpkschmarty: feel free to open github issues for them on XRay. XRay is handling all the feed parsing going on. So you should be able to demonstrate the bug by linking to the XRay parsed url for the feed.
#aaronpkMake sure to use "expect=feed" in the query string too cause that triggers feed parsing mode.
#Zegnatjeremycherfas, moving here from #chat, if you just make a folder /micropub on your domain, isn’t that accessible? I would expect Grav to only capture requests when no files exist. You shouldn’t have to use Grav’s request routing to run a seperate PHP script
#Zegnat(Unless of coure you specifically want to make it a Grav plugin)
#jeremycherfasThat's the conclusion I have just come to. I had a nice chat on the Grav slack with a woman who is quite indieweb minded and she advised exactly that. She has modified voxpelli's micropub to work with Grav.
KartikPrabhu, [kevinmarks], calumryan, iasai_, yatil, [pfefferle] and swentel joined the channel
#sknebelaaronpk: does Aperture only work if I use tokens.indieauth.com or do you have some trickery on first login to remember the token endpoint? (currently wondering if/how I can solve that in my bridge)
#aaronpksknebel: it stores the users token endpoint in the database
tantek joined the channel
#aaronpkthat's also the reason each user has a unique Microsub endpoint rather than having everyone share the same one in Aperture, so that I know where to send the token to to verify it
#sknebelaaronpk: aaahhh. I didn't connect those two
#swentelaaronpk, I've added an jf2 feed option to the drupal plugin, but I wonder how useful it is right now .. e.g. aperture doesn't discover it for instance, or am I missing something ?
sebsel and [sebsel] joined the channel
#aaronpkswentel: I only use jf2 as an internal format right now rather than something I expect to be published on the web. I could maybe see that changing but I don't know how much I expect that to actually happen
#swentelI also wonder a bit how to add discovery for my microformat pages on /timeline for instance (I used that url in aperture), but rel="feed" isn't discovered either. Maybe I should just add an rss/subscribe page with all the options :)
#Zegnat2 personal sites that publish jf2 there. I thought aaronpk wasn’t doing it anymore, but I guess it was pure parsed microformats2 json and not jf2 that didn’t make it into the new p3k
#tantekZegnat: jf2 is the API format from webmention.io
#Loqifriending (AKA add friend or the verb to friend) is a general term applied to the various silo user interactions with sites regarding reciprocal friendship https://indieweb.org/friending
#jeremycherfaszegnat Is it usable locally now? I obviously need authorization and token to continue, and gimme-a-token needs me to open a hole with ngrok, but the details for that would change every time. I'm wondering whether I can use your auth and token to do everything locally.
#tantekgreat to see blog posts with thoughtful comments again
#jeremycherfas!tell aaronpk I'm trying to get skippy's micropub working for Grav. If I authroise Quill through ngrok, do the access and authroization tokens persist through different sessions with ngrok, or is the ngrok URL crucial to the validity of the tokens?
#jeremycherfasON the other hand, if I want to test with Quill, I would still have to sign in each time I open a new session with ngrok, or else install Quill locally, which looks a little difficult.
#jeremycherfasI'm doing all this in MAMP, so I can create a tokens.db with phpmyadmin, but then not sure how to use the schema.sql
#jeremycherfasMAMP seems to have a SQLiteManager folder but it is empty. Let me go searching.
#ZegnatMAMP should have the SQLite plugin for PHP ready to go. Do you have sqlite installed on your machine? Then you can use the command line to create the DB.
#LoqiMAMP is a web server software package for MacOS that has Apache, MySQL, and PHP in it that you can run on your Mac laptop to test your indieweb site locally https://indieweb.org/MAMP
#ZegnatOh, I don’t know if it has a visual DB manager
#jeremycherfasI created it on the command line, and now I cannot see how to access it. I guess I better install sqlite
#ZegnatI need to write a better readme on that part. But it is hard to automate the creation for people, as it is essential that you place the db somewhere secure.
#ZegnatIf either of those work, you already have sqlite installed, and you should be able to just do the cli examples given in the Mintoken README and not have to think about it ever again
#jeremycherfasSo I'll download the browser and see what happens.
#ZegnatThe browser is still really nice for opening up the SQLite database files and looking at what is happening for sure!
#jeremycherfasOK. I'll try the CLI stuff. But yes, I did want to see what was going on!
#ZegnatYou’ll be able to see the tokens issued, and when they got revoked, stuff like that in the db file.
#aaronpkZegnat: out of curiosity, why didn't you store the tokens in files?
#ZegnatGuess I personally didn’t want to bother with one file for every token ever issued. That seemed like a great case for a db. And I like SQLite.
#GWGSo, this is a dev question...how can a site tell consumers that it doesn't want that site to display its profile image or other data? Even if it sends webmentions?
#aaronpkI mean it's definitely a good use of a DB, but now you have to walk people through setting up the sqlite file :)
tantek joined the channel
#ZegnatAnd otherwise I have to walk them through setting up a config file. But I see what you mean, that is often more straight forward.
#ZegnatI guess part of me feels that if you want the security, you’ll have to be OK with the initial hurdle of setting up the db? :P
#ZegnatI was going to include an sqlite db ready to go, but I thought the chance of people leaving it in the same (publicly accessible) directory as endpoint.php was too big
#jeremycherfasOmnibear says Authetication successful but seems to have hun on Retrieving access token
#GWGsnarfed, Bridgy sends a webmention with a source of bridgy.appspot.com or something like that, but the page has a u-url of the true source, correct?
#Loqi[pfefferle] Perhaps we should also rethink bird.gy. Why should I, as a site owner, register at bird.gy to get likes to my tweets. Why not build a service where twitter users can register to send pings to sites, they like or tweet? Something like flattr tried som...
#tantekGWG, a long long time ago, a bunch of folks thought they could effectively theroetically make privacy preferences something machine understandable that everyone would comply to. They were wrong. They created something that was overly complex and ignored. AKA P3P https://en.wikipedia.org/wiki/P3P
#GWGI am just reading the situation as people wanting a choice. I want to give them one
#GWGSo, I want to add a 'tuning knob' to turn down the sensitivity on data collection
#tantekGWG sounds better to dcoument the use-cases first
#jeremycherfasThose sqlite examples do not have localhost or anything like that.
#schmartyonce wrote a search engine that crawled for P3P data and re-ranked search results according to how close they matched the user's privacy policy settings
#schmartythen all of the major search engines updated their API terms of service to forbid re-ranking search results. so. goodbye Privacy Finder.
#[jeremycherfas]!tell Zegnat I will do that, but many of the errors were the result of my ignorance. My feeling is that copying the CLI instruction should come first, as that is the easiest option for people like me.
#wolkenbruch.netcreated /gopost (+221) "Created page with "gopost is a tool to make posts to several social networks at once and also generates a web html snippet to embed in your (static) website. This is work in progress, currently..."" (view diff)
#tantek.comedited /Falcon () "(-781) implemented and shipped as of 2018-05-02: reacji in-stream reply-context and general reply-context in-stream! move use-cases to appropriate pages" (view diff)