#social 2018-07-14

2018-07-14 UTC
timbl joined the channel
#
saranix widowclip++
#
Loqi widowclip has 1 karma
#
widowclip Neat! Also have you seen https://brutaldon.online/ ? It might be a better client if you find Mastodon client cluttered.
#
saranix wow...how many *dons are there now? :-P
#
widowclip A sizable herd to be sure haha.
#
technomancy <3 brutaldon
#
technomancy so much easier to operate with a keyboard than the default masto client
#
cwebber2 hi
#
cwebber2 nightpool[m]: well, my move is finally wrapped up :)
#
cwebber2 so we should be able to meet next weds
widowclip joined the channel
#
aaronpk Good cause I'm gonna be on a plane!
#
saranix hmm. It should be possible to submit an ActivityPub implementation report via ActivityPub....
#
aaronpk Relevant to our signature discussion earlier... https://toot.cat/@jamey/100370521534196878
#
Loqi [Jamey Sharp] cryptography question Show more After I receive an HTTP response secured with TLS, I'd like to be able to hand that response to somebody el...
DenSchub joined the channel
#
aaronpk is looking forward to that Microformats update getting rolled out to more servers soon. That'll fix the "show more" thing here
xmpp-social and KjetilK_ joined the channel
#
puckipedia hah, HTTP signatures and JWTs implemented
timbl and JanKusanagi joined the channel
#
fr33domlover puckipedia, what are JWTs?
#
fr33domlover ah json web tokens
jankusanagi_ joined the channel
#
aaronpk puckipedia: what’d you do with JWTs?
#
puckipedia aaronpk: I'm using them (or well, limited) as local auth tokens
#
aaronpk ah
#
puckipedia like, I wrote a tiny RS256-only jwt implementation
#
puckipedia and in kroeg, all tokens are signed by the key of the owner
#
puckipedia technically, you could have an account without storing the private key on the server
cwebber2 joined the channel
#
puckipedia oh no I just realised
#
puckipedia {"type": ["Accept", "Reject"], "object": "example/follow_request"}
#
saranix I see no problem with that. I would process them in the order given.
#
puckipedia except. there's no order. it's a set of types
#
saranix json defines order I think
#
saranix type[0] then type[1]
#
puckipedia actaully now I do wonder
#
puckipedia yeah, it's actually ordered??
#
saranix why is that surprising? It wouldn't be much of an array if it was random.
#
puckipedia json-ld arrays, usually, are actually sets
#
puckipedia as in, e.g. the to/cc/bcc/bto properties in AS2 are sets
#
puckipedia so to: [Public, ~puckipedia] == to: [~puckipedia, Public]
#
saranix are you sure about that? Just because they result in the same behavior doesn't mean they are ==
#
puckipedia "Since graphs do not describe ordering for links between nodes, arrays in JSON-LD do not provide an ordering of the contained elements by default."
#
saranix still... I think that just means it's OK to pull the set from SQL and let it be any order, not that they are ===
#
saranix in a boolean sense
#
puckipedia is there a practical difference
#
saranix only if you are pedantic and trying to cause a problem where there is none? ;-)
#
cwebber2 hello hello
#
puckipedia hiya
#
cwebber2 puckipedia: it's ordered in json, but not in the rdf graph
#
puckipedia I think kroeg-rs is getting close to finished
#
cwebber2 admittedly, I do treat the order as significant
#
cwebber2 even though you "shouldn't" from an rdf standpoint
#
cwebber2 for type, that is
#
puckipedia amusignly, I don't
widowclip joined the channel
#
widowclip puckipedia: Will Kroeg-Rs be open source? Is it already? I was thinking of starting an AP project in rust it would be lovely to have a resource like that, and I would certainly be happy to contribute if I can.
#
puckipedia it's mostly open source right now
#
puckipedia https://git.puckipedia.com/kroeg
#
widowclip Thanks for the link!
#
puckipedia requires a non-recent nightly, I run it on 2018-06-23
#
widowclip How come?
#
puckipedia the more recent await/async merges
#
widowclip Ahh