#aaronpkswentel: I'm not sure why there aren't any releases tagged. I don't actually remember writing that tho lol
#swentelaaronpk, heh, well I'm writing a websub component for the drupal indieweb module, and I'm too lazy to write the discover part. That little library seems to do it well :) (actually, haven't tested it yet)
#swentelI could just copy/paste parts of it of course
#@mattiasgeniar↩️ that `htop` header was indeed very distinguishable from all other blogs ... i hope this new layout is too, gotta get used to it :)
for the comments: i keep hearing good things about webmention, seems easier to integrate (twitter.com/_/status/1192388163475591169)
#swentelaaronpk, trying out websub.rocks for testing. subscribe and subscribe are fine. But I fail to see how I can get a notification in - trying out test 300, but when I see the blog I don't know what todo there :)
#swentelI guess switchboard.io also does async validation right? websock.rocks does it immediately when trying to subscribe, kind of annoying in a way I guess (but makes sense too)
#aaronpkHm i don't remember the details of the tests anymore, it's been a while
jeremych_ joined the channel; fLsh42Discord[m] left the channel
#@ramsey↩️ First I’ve heard of it. Is there more information about it? I can’t find much on their homepage. Is it an open source project? Federated? Using webmentions, micropub, activity streams? (twitter.com/_/status/1192439468231417856)
#petermolnaroh, the nail was for the twitter rejecting dev accounts
#[manton][snarfed] I was thinking about that yesterday with that news of Twitter employees looking up private info for Saudi Arabia. That's easier when you have many accounts concentrated in one place.
#aaronpkat least the initial comments look like they're using OAuth as a template
#[cleverdevil]Hey I’d love to be wrong. Seven years at DreamHost made me pretty jaded about WordPress haha.
#GWGaaronpk: That's why I volunteered to comment and asked you to monitor the repo
[chrisaldrich], eli_oat, asymptotically, ola, [asuh], leg and [LewisCowles] joined the channel
#[LewisCowles]Does anyone know about iframe communications changes between firefox 60 and 68?
#[LewisCowles]I have a friend who is currently undergoing chemo and he’s suffering a bit with something he’s doing between iframes. I’d really like to help him
#[LewisCowles]> Some time between Firefox60 and Firefox68, code I had written over many years while teaching to communicate with IFrames has stopped working. The changes are supposed to make it more difficult for hackers to exploit them. Details as to what they have done are sparse, not helped by the fact they seemed to have screwed their own relevant wiki pages in the process. I’m currently working through my scripts to get them working again, but it
#[LewisCowles]a slow process. The iframe has two parameters, ‘allow’ and ‘sandbox’ which I suspect hold the key. Most tutorials on the subject are also screwed in the same way my code is, so not much help there either.
KartikPrabhu, leg and eli_oat joined the channel
#[LewisCowles]I’m thinking he either has default content-security policy issues or another missing http header
#[LewisCowles]I know I added this to a ruby app at work. TBH I try to avoid frames of all kinds
#[manton]IndieAuth question if [aaronpk] or anyone has thoughts on this... Micro.blog as an auth provider currently assumes you have a hosted blog on Micro.blog, so if you delegate an external blog to Micro.blog's auth URLs, things don't work when Micro.blog is figuring out what to put in the "me" URL response. But I'm think if an external URL is verified on your account, I could support it. Does that make sense or are there any security concerns I
#[jgmac1106]I don't have the slightest, I was using the RSS feed
#[jgmac1106]as long as the post kind RSS feeds are proper for note, photo, article, reply everything should work
#[jgmac1106]...well reply I don't know how that works
#[jgmac1106]I just wish the POSSed notes showed up on my micro.blog site and not just in the feeds
#GWGIt works well enough, but I don't get back the links for rel syndication
#gRegorLove[manton] I think as long as the domain doesn't change between what's entered and what's the auth endpoint returns. e.g. if I enter micro.blog/gregorlove as my profile, it would be go against spec for the auth endpoint to return gregorlove.com
#gRegorLovebut if I entered gregorlove.com -> mb auth endpoint returns gregorlove.com, should be ok
#[snarfed]i was mostly joking. the scraping is the easy (ish) part. the endless arms race of evading their bot detection is the hard part.
#mblaneydo you have more details on that? if scraping was more distributed would that be less detectable?
#[snarfed]oh god, lots, all over the web. there's a whole sketchy cottage industry here. a brief taste in https://github.com/snarfed/bridgy/issues/854#issuecomment-548844715 . rotating IPs, spoofing User-Agent, running some bits of JS to generate reasonable-looking browser fingerprints...it really is an arms race
#Loqi[snarfed] agreed on all counts. i share your frustration!
i actually wrote that crawler you mentioned, a bit ago. www.facebook.com uses JS, so crawling that is hard, but m.facebook.com (with a logged in session cookie) is pure HTML and eminently scrapeable,...
#[snarfed]indieweb as a hobby, i like. evading big competent teams' bot detection as a hobby, no thanks.