#LoqiOpenID was a protocol for using a web address as an identity to sign-in to websites; it is losing support, is effectively dead (versions 1 & 2 are both deprecated, sites are dropping support), and has been replaced on the IndieWeb with web-sign-in and IndieAuth https://indieweb.org/OpenID
jackjamieson, swentel and [eddie] joined the channel
#[eddie]I’m thinking about how my nicknames cache is built. Currently it’s just a huge yaml file on my server which is not easily editable. So I’m thinking of transitioning to having single file h-cards for each person named by their primary url
#[eddie]So it seems like the proper way to add contacts to that would be to receive Micropub posts that are of type h-card and when that is received to assume it is a contact
#[eddie]Am I missing any other uses of h-card via Micropub?
#[eddie]So url can be an array and then the primary one would be also in uid
#[eddie]The idea is I could then add to Indigenous’ share sheet that if I’m on a page with an h-card it would have an option to “Save Card” or “Save h-card”
#[eddie]So if I went to aaronparecki.com, I click the share sheet and then with a tap of a button it’ll parse your h-card and embed it in a Micropub request and send it to my Micropub server
#[eddie]I’ll probably want to show an edit form to be able to easily add things like category or missing info
#aaronpkmy Micropub endpoint already knows how to create h-cards so I look forward to trying that out :-)
snarfed joined the channel
#GWGIs there a web client that does h-card micropub?
snarfed joined the channel
#[eddie]Awesome aaronpk. I’m thinking since most websites don’t know how to create h-cards, I’m thinking I might hide the feature behind the micropub post-types query
#[eddie]Because an h-card of a random person to a Micropub that doesn’t deal with h-cards differently than h-entries could end strangely lol
#LoqiA nicknames cache is a way indieweb sites store information about people to improve the user experience of the site owner referring, mention, and/or linking to those people https://indieweb.org/nickname_cache
#LoqiCSRF or Cross-Site Request Forgery is an attack that OAuth and IndieAuth clients (relying parties) need to be aware of, wherein victims unknowingly follow a link to a relying party callback URL prepared by an attacker https://indieweb.org/CSRF
#AngeloGladdingSSRF is an attack that IndieAuth servers and Webmention receivers need to be aware of, wherein attackers abuse URL parsers to various nefarious ends.
#ZegnatThough I did some improvements during IWC NYC that still haven't been merged to master.
#ZegnatI for one appreciated that note in the spec, aaronpk! Good reminder for implementers
#AngeloGladdingyeah aaronpk i remember reading that note
#AngeloGladdingmy problem the other day w/ webmention testing was a bug in my own URL parser
#AngeloGladdingi've always known there were *issues* so i've tried to enforce simplified URLs
#ZegnatHmm. Looking at that safecurl code they disable cert verification and then mess with host headers. That's not the way to go. Unless you have to support old PHP and curl they should do what I show and pass the resolved IP for the domain to curl and let it handle those details itself.
#AngeloGladdingbut it seems this "New Era of SSRF" as the accompanied video demonstrates requires a second look at a single line of warning in a spec
#ZegnatI'll have to check that video tomorrow and see if there is anything else I could add.
#AngeloGladdingthe guy has a hard taiwan accent and it's 50 minutes long but it's basically all relevant to IndieWeb as we sling URLs all day
#AngeloGladdingbut there's actually many vectors and unless you're on PHP 7.1+ (i think) you're vulnerable
#AngeloGladding""curl doesn't verify that the URL is 100% syntactically correct. It is instead documented to work with URLs and sort of assumes that you pass it correct input"
#AngeloGladdingthat was what the curl team had to say after they patched his first vulnerability
#AngeloGladdingit's PHP 7.0.12 or earlier for the bug described on slide #69
#AngeloGladdingcrafting abusive URLs almost akin to XSS
#AngeloGladdingbad comparison w/ XSS, forget that -- the solutions provided in the spec are valid but i didn't fully grasp the scope of the problem when i read it at first