2018-04-29 UTC
jgmac1106 joined the channel
Gab1, renem, eli_oat, AngeloGladding, eli_oat1, eli_oat2, jgmac1106, sknebel, Zegnat, leg, snarfed, [unoabraham], [chrisaldrich], barpthewire, EmreSokullu, [pfefferle], iasai_, jeremycherfas, iasai, Mandrake and [kaushal_modi] joined the channel
# 11:41 [kaushal_modi] I'm not well-versed with DDOS. I read that wiki, but still not sure what a user can do to prevent that. Looks like the prevention measure need to be taken by the endpoint? Or is there something a user (WM source/destination URL) can do too? If so, can someone point to some examples? # 11:57 Zegnat There are some interesting discussions about it, [kaushal_modi] ... Afraid I do not have a lot of them at hand on my phone. # 11:58 Zegnat S lot of mitigation will probably end up at the receiver though, not the sender. # 11:58 Zegnat As a sender there is other stuff you can do, to show you aren't a spam mention. E.g. Vouch. Mandrake joined the channel
[unoabraham], [pfefferle], Mandrake and eli_oat joined the channel
# 13:28 GWG I've never figure out how I'd want to implement Vouch wiobyrne, jgmac1106 and barpthewire joined the channel
# 13:52 Zegnat Same here. I've also only received a single one with vouch, from gRegorLove # 13:54 sknebel vouch also doesn't necessarily help with the DDoS issue # 14:03 GWG People are always complaining about moderation getting in the way of the wonder of webmentions EmreSokullu, snarfed and [pfefferle] joined the channel
# 14:30 Zegnat People will also complain about unmoderated flow of content. Especially when an unspecified critical mass is reached # 14:31 Zegnat GWG, does askismet et al cover the webmentions WordPress pulls in? # 14:35 GWG It's gotten better since we changed the loading order, but not perfect wiobyrne joined the channel
# 14:46 Zegnat I am imagining a world where trackbacks and pingbacks do not have that, but webmentions do. Good selling point. KartikPrabhu, eli_oat and marcthiele joined the channel
# 15:26 Zegnat Integration with existing moderation and spam filter tools. # 15:26 Zegnat Though I am not sure if e.g. pingback does that or not AngeloGladding and EmreSokullu joined the channel
KartikPrabhu, EmreSoku_, eli_oat, snarfed, EmreSokullu and barpthewire joined the channel
# 16:33 GWG I am having a slow time of it designing a POSSE system that takes the same input for everything but outputs to the appropriate system..be it Bridgy Publish, or a native API # 16:37 GWG Bridgy Publish does this by using Microformats 2 to build a submission. I need to use the same, but where they are stored on my system # 16:39 GWG Looking at silo.pub and how it does this as well. EmreSokullu and eli_oat joined the channel
eli_oat1 and kaushalmodi joined the channel
# 18:19 kaushalmodi !tell Zegnat Thank you for your response on the DDoS issue. I have a static site. So I will have to just wait and watch and deal with the DDoS issue if it ever affects me. # 18:19 Loqi Ok, I'll tell them that when I see them next # 18:30 snarfed GWG: which native api(s) do you want to posse to that bridgy doesn't support? just curious # 18:33 GWG snarfed: Too many people are blaming me for Bridgy Publish not being able to interpret their Microformats because they didn't set them up. I just want to make sure I can support them transparently if I add them. # 18:35 GWG snarfed: I'm mostly trying to make it so it isn't a 'Bridgy Publish' plugin anymore. It's just a generic publish plugin # 18:36 snarfed sure, i understand. i read your post. :P i was just curious where else you want to add posse support for # 18:37 snarfed do you know how you'll handle bad mf2? use the underlying wp data directly instead? you'd then need to implement all the silo APIs yourself, right? # 18:37 GWG There is one I always say I'd like to do # 18:39 GWG snarfed: The other option is to figure out how to hook into other POSSE plugins if they have a filter to support this. # 18:41 GWG That's why the design is so important # 18:42 GWG snarfed: My predicition is that I won't get much interest in people integrating to it. # 18:43 GWG But, I didn't think I'd add weather. So who knows AngeloGladding, [pfefferle] and eli_oat joined the channel
# 19:19 dgold wouldn't it be easier to posse to pinboard at post-creation time, rather than using bridgy? # 19:21 snarfed dgold: implementing a whole silo API seems a lot harder than just sending a webmention to bridgy publish. or maybe i misunderstand? # 19:23 snarfed but i doubt simpler than including a single HTML <a> tag in a post (since most of our CMSes auto-send wms). right? # 19:23 dgold post: url, description(title). its not "a whole silo" # 19:23 snarfed also academic since bridgy doesn't support pinboard # 19:30 snarfed hey GWG is the wordpress indieauth plugin's auth code verification response JSON or form-encoded? # 19:31 GWG I believe the spec says it is supposed to be # 19:32 snarfed yup it does, but servers may still return form-encoded depending on the Accept header. # 19:33 GWG I don't think I implemented it, why? # 19:34 wagle anyone use lychee? (I've woken up and am poking at my photo album again) any gotchas? # 19:37 snarfed wagle: so what? "freshness" is a bad indicator of code/project quality imho # 19:37 GWG I don't think I have ever used App Engine # 19:38 snarfed GWG: no need, just type your site's url into the indieauth text box on that page # 19:38 wagle snarfed: I've love that, but the damn kids gotta go around breaking everything else, including the dependencies # 19:39 wagle snarfed: wasnt a deal breaker, just a "hmmm" # 19:40 snarfed forever, or until the package repo disappears, which then would be a much bigger problem :P # 19:40 GWG snarfed: Logged me in but didn't redirect me back. # 19:40 snarfed huh, ok, then it didn't test my new code. is that a bug in the wp plugin? # 19:41 dgold wagle: looks like they've created a new ownership structure as of this month # 19:41 wagle dgold: yeah, adding to the hmm factor, but most of the time stamps are 2-4 years old # 19:42 wagle since the hmm factor was "high", thought i'd ask around a little to see if there was any scuttlebutt # 19:42 wagle dgold: need it to work for 50 years.. well, 5? # 19:43 wagle photo photos appear to still rule for longevity # 19:43 wagle paper photos appear to still rule for longevity # 19:44 snarfed nah. paper degrades, fires, theft, loss, only one copy, etc. i'd trust LOCKSS + regular data migrations over physical. # 19:44 wagle but I'm t the wrong end of the country to show anyoone the paper edition # 19:45 snarfed hah yes plus that, way less convenient and useful # 19:45 wagle snarfed: yeah, but I will be dead in 50 years # 19:45 GWG snarfed: Let me turn on debugging # 19:46 wagle dunno how to get relatives to migreate every 3-5 years # 19:46 snarfed oh that's a totally different question. thought you were just talking about your own photos # 19:47 snarfed but again digital is arguably better there. you can keep and maintain copies of your relatives' *digital* photos, but not their paper photos # 19:48 snarfed also the JPEG format has lasted many decades, so for just photo files themselves, no real migrations needed, just backups # 19:49 wagle havent gotten a lot of interest.. main problem now is that I have the paper copies, and they shouldn't go poof when I die.. so gotta get the two sides to saw which goes to which side of the families # 19:50 wagle yeah, but my energy level has been sucking badly # 19:51 GWG snarfed: I seem to have broken my debug code. # 19:52 GWG I haven't used it manually to watch a transaction in a while # 19:54 Loqi snarfed has 60 karma in this channel (382 overall) jgmac1106 joined the channel
# 19:58 GWG snarfed: It looks like it was the improved error handling. It is only logging errors. This wasn't reported as an error, so it didn't log it. I just moved it to log everything. # 19:58 wagle scan first requires that II annotate each pair of fronts and backs on the hundreds of old photos then implement a way to find the paper again.. any suggestions? or am I a pioneer? # 19:59 snarfed tons of services for this, and people who have posted about doing it # 20:00 snarfed maybe less about bidirectional indexing (ie back to paper), but definitely tons on scanning at least # 20:00 wagle I'm pretty confused, brain no longer works right # 20:01 GWG snarfed: I think it is a URLencode issue # 20:02 GWG snarfed: The reason it isn't redirecting. # 20:03 GWG snarfed: It is trying to redirect to.. https:\/\/tiny.n9n.us\/wp-login.php?redirect_to=https%3A%2F%2Foauth-dropins.appspot.com%2Findieauth%2Foauth_callback&reauth=1&action=indieauth&response_type=id&client_id=https:\/\/oauth-dropins.appspot.com\/&state=%7B%22endpoint%22%3A%22https%3A%2F%2Ftiny.n9n.us%2Fwp-json%2Findieauth%2F1.0%2Fauth%22%2C%22me%22%3A%22https%3A%2F%2Ftiny.n9n.us%22%7D&me=https:\/\/tiny.n9n.us&_wpnonce=4935d40ed5" # 20:04 GWG snarfed: You are embedding information in the state? # 20:06 wagle now gotta recall why I thought piwigo was the better alternative to lychee # 20:08 sknebel I seem to remember not liking the very JS-heavy Lychee UI, but that might not be current anymore, or something you see differently # 20:10 wagle just remembered that I have to prototype the annotation system too see what I need to annotate, etc # 20:11 sknebel wagle: oh, and Lychee didn't have support for multiple users I think? # 20:12 GWG It is not redirecting. Are you asking for form-encoded support? # 20:12 wagle sknebel: hmm.. i need that for letting relatives say "oh! thats grand grandma betsy!" # 20:15 snarfed GWG: for the code verification request i send? i send Accept: */*, then detect and handle both form-encoded and json responses # 20:17 GWG snarfed: It doesn't seem to get to the verification step. # 20:17 snarfed if you're not redirecting to me, then yes, it wouldn't # 20:17 GWG Something is different about the redirect # 20:17 GWG I tested every IndieAuth client I could find. # 20:19 Loqi Zegnat: kaushalmodi left you a message 1 hour, 59 minutes ago: Thank you for your response on the DDoS issue. I have a static site. So I will have to just wait and watch and deal with the DDoS issue if it ever affects me. # 20:20 Zegnat I will try when I get back out of my slumber phase ;) # 20:21 pstuifzand I could also authenticate with the oauth_dropins website, but not without adding a application/json content type first. It seems it was missing from the verification step. jgmac1106 joined the channel
# 20:30 GWG !tell snarfed I found your problem. # 20:30 Loqi Ok, I'll tell them that when I see them next # 20:31 GWG !tell snarfed You are missing response_type=code . Without it, you are in the authentication, not authorization flow, which does not return a code. # 20:31 Loqi Ok, I'll tell them that when I see them next # 20:38 GWG sknebel: But is he trying to authorize or authenticate? Now I'm confused. # 20:38 GWG I may be expecting something different. snarfed joined the channel
# 20:41 GWG snarfed: I think I shouldn't try to troubleshoot while I'm thinking about something else. # 20:41 GWG But, it still looks to be a urlencoding issue. snarfed and [snarfed] joined the channel
# 20:47 [snarfed] thanks for debugging GWG! not high priority, feel free to punt # 20:47 Loqi [snarfed]: GWG left you a message 16 minutes ago: I found your problem. # 20:47 Loqi [snarfed]: GWG left you a message 15 minutes ago: You are missing response_type=code . Without it, you are in the authentication, not authorization flow, which does not return a code. snarfed joined the channel
# 20:47 GWG [snarfed]: I think I did find a problem I want to fix with the workflow you are trying to use snarfed joined the channel
# 20:49 [snarfed] ok! also if you're going to test your plugin, feel free to test against bridgy instead of o-d since bridgy has actual users :P # 20:52 GWG [snarfed]: It's stripping after the state. # 20:52 GWG Something about the encoding is an issue for my code # 21:05 GWG [snarfed]: Okay. Should be fixed now # 21:05 GWG I urlencoded the state before passing it in. # 21:05 GWG [snarfed]: How would I test against Bridgy? jgmac1106, EmreSokullu, renem, leg and chrisaldrich joined the channel
# 22:06 Loqi Countdown set by tantek on 2018-04-21 at 3:04pm PDT jgmac1106, [miklb], srushe, wagle and snarfed joined the channel
[kaushal_modi] Hello all.
Loqi The Vouch protocol is an anti-spam extension to Webmention. Webmention with Vouch depends on understanding Webmention https://indieweb.org/Vouch
klezlab.it edited /Hacker_News (-81) "/* POSSE to Hacker News */ there's no way to follow someone on the site" (view diff)
snarfed happy to accept PRs though! https://bridgy.readthedocs.io/en/master/#adding-a-new-silo
GWG snarfed: It is trying to redirect to.. https:\/\/tiny.n9n.us\/wp-login.php?redirect_to=https%3A%2F%2Foauth-dropins.appspot.com%2Findieauth%2Foauth_callback&reauth=1&action=indieauth&response_type=id&client_id=https:\/\/oauth-dropins.appspot.com\/&state=%7B%22endpoint%22%3A%22https%3A%2F%2Ftiny.n9n.us%2Fwp-json%2Findieauth%2F1.0%2Fauth%22%2C%22me%22%3A%22https%3A%2F%2Ftiny.n9n.us%22%7D&me=https:\/\/tiny.n9n.us&_wpnonce=4935d40ed5"
Zegnat snarfed, I am half awake. I was able (I think) to login on the oauth-dropins.appspot.com. At least it says “ Authenticated https://vanderven.se/martijn/ on IndieAuth. ”
pstuifzand I could also authenticate with the oauth_dropins website, but not without adding a application/json content type first. It seems it was missing from the verification step.