<jbenet>
i think it makes sense to turn our "keystore" into a "keychain". what we need is a global merkeldag of keys. (basically what a CA is supposed to be, but with really really nice API. maybe we can convince keybase to do it).
<jbenet>
(keys, users, and relationships between users + keys (access, control, etc))
<cryptix>
jbenet: sounds nice - just have to be very clear about the 'publish' step - some people might want to establish their relationships in private
<Blame>
im actually not that keen for a merkle-dag for mapping identify->key (is that what you are talking about?)
<cryptix>
i guessed so
<Blame>
it creates an oversight problem. (you can only prove that things are in the merkledag. You cannot prove something is not unless you can check the entire merkledag)
G-Ray has quit [Ping timeout: 256 seconds]
<Blame>
And as it is used over time, unless it is refactored, it devolves into a block-chain style structure.
<cryptix>
btw Blame thx for your input on the tor stuff id really like to compile that info and see what needs to be done to create 'client only' nodes
<Blame>
it is great for static records, but bad for a structure that needs updating
<Blame>
cryptix: NP. I'd love to help out with that. "client only" is not that hard. You just have to make sure that nodes doing try and get peer/routing information for lookups, reads, and writes
<jbenet>
Blame think about it like a merkle web of keys and capabilities
<Blame>
'merkle web'?
<jbenet>
Sorry merkle dag
patcon has joined #ipfs
<Blame>
np. 'merkle web' is just an assumption shattering idea.
<Blame>
so, cryptographically backed web of trust?
<jbenet>
I know which is why I prefer merkle dag but people have a harder time with "dag" than "web"
<jbenet>
Btw awesome stuff with stack stream -- there was a conversation today on Twitter about unison and lambda calculus on a merkle dag
<Blame>
im not sure how those things fit together.
<cryptix>
wow yea - i'll clean the kitchen while my mind goes up in flames again
<jbenet>
Wanted to mention stack stream but didn't know whether you wanted me to.
<jbenet>
They don't fit together fully
<cryptix>
we had crazy ideas the other night about making up datastructres by describing them in small byte code languages
<Blame>
im fiine with it. Right now I have about 5 other things at higher "panick levels" but I want to make it happen.
<jbenet>
stackstream fit in relation to ipfs and with the use case I mentioned (persistent computable objects).
<jbenet>
Sounds good and no worries to already at an awesome level
<jbenet>
Idk how to pick these thing s or what best thing to do is
lgierth has quit [Quit: Ex-Chat]
<jbenet>
I'm not a PL person
<Blame>
yeah. Rather than a high level approach. "What is the simple thing that could do this job"
<jbenet>
Right
<Blame>
honestly an assembly avoids the programming language argument entirely. write a compiler that does what you want
<Blame>
i'll be back in an hour. sorry for just popping in
<jbenet>
Yeah that makes sense to me
<jbenet>
Ok metro time
<cryptix>
household time ^^
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
G-Ray has joined #ipfs
cryptix has quit [Quit: leaving]
domanic has quit [Ping timeout: 245 seconds]
cryptix has joined #ipfs
G-Ray has quit [Remote host closed the connection]
octalberry has joined #ipfs
<cryptix>
finally got my scuttlebot running where i wanted it to. add me if you can: VYAwnIIG6ImASEhrVEAVFJTucjG44xS+LTVfzUEybjA=.blake2s
octalberry has quit [Ping timeout: 272 seconds]
gatesvp has joined #ipfs
Darksecond has joined #ipfs
tilgovi has joined #ipfs
harlan| has joined #ipfs
harlan| has quit [Remote host closed the connection]
guest756 has joined #ipfs
Darksecond has quit [Quit: leaving]
daviddias has joined #ipfs
inconshreveable has quit [Read error: Connection reset by peer]
<Blame>
that looks nifty
<Blame>
the big issue I am having with stackstream is the entire "unlimited length" thing. I feel like there should be a more elegant way to handle that.
tilgovi has quit [Ping timeout: 245 seconds]
daviddias has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
lgierth has joined #ipfs
lgierth has quit [Read error: Connection reset by peer]
lgierth has joined #ipfs
inconshreveable has joined #ipfs
chriscool has joined #ipfs
tilgovi has joined #ipfs
tilgovi has quit [Ping timeout: 256 seconds]
patcon has quit [Ping timeout: 255 seconds]
daviddias has joined #ipfs
<cryptix>
from what i grok now, the id is usless on it's own..
<cryptix>
just found how to generate proper invites - "178.62.206.163,VYAwnIIG6ImASEhrVEAVFJTucjG44xS+LTVfzUEybjA=.blake2s,89lq3yvs0Zciym3+VLNxAcxaicY8Kv5JXLiBS9pGhfA="
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
<Blame>
scuttlebot docs are failing to explain what it actually does.
daviddias has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<cryptix>
whyrusleeping: which quic impl was that?
<jbenet>
rht: sorry i've delayed on review-- traveling atm. will take a look at your stuff in the next few days
notduncansmith has joined #ipfs
<wking>
jbenet, whyrusleeping: Ping me when you're ready to talk about an IPFS Docker-registry backend. I should be around for the next hour or two.
notduncansmith has quit [Read error: Connection reset by peer]
<jbenet>
whyrusleeping daviddias: you two may want to chat for a while
<daviddias>
jbenet do you have extra 5 mins?
<jbenet>
krl and i will finish up the meetup
<jbenet>
daviddias yeah
<daviddias>
Cool, I was looking for to talk about rough next steps
<jbenet>
yep sounds good let's chat on +hangouts, sec
<jbenet>
chriscool, rht, and others: if you have any outstanding issues add them to the etherpad-- will be turned into an issue in a few hours.
<rht_>
there is one I discovered very recently: ipfs cat <hash> (directly to stdout) can't be cancelled with C-c
<cryptix>
i need to shuffle a lot of $life annoyance in the next week or two, so i dont want to pick up big things i cant deliver - i think my time on ipfs is better spent giving feedback on issues/PRs and fixing triage
<cryptix>
the big thing i took from the weekend in berlin are the privacy converns which is why i want to push #1118 - i guess the client only mode would also be interesting wrt android
<whyrusleeping>
cryptix: it was devsisters/goquic
tilgovi has joined #ipfs
<whyrusleeping>
my availability is a bit limited today actually, may have to chat tomorrow
notduncansmith has quit [Read error: Connection reset by peer]
chriscool has quit [Ping timeout: 264 seconds]
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
<jbenet>
cryptix: we should try to repro a test for that
<jbenet>
then we'll know.
<cryptix>
how deep can we mock bitswap? if we can start to respond and then stall/disconnect this should be easy to reproduce
octalberry has joined #ipfs
tilgovi has quit [Ping timeout: 256 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
therealplato has quit [Read error: Connection reset by peer]
tibor has quit [Quit: Connection closed for inactivity]
torpor has joined #ipfs
<torpor>
hi guys - does anyone know how to do "ifps init" programmatically - i.e. not from the command line?
<Blame>
torpor: it is likely not what you want, but you could do a shell/system call in a language of your choice.
<torpor>
i want to do it in Go
<torpor>
so, programmatically in Go
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<torpor>
i found in go-ipfs/test/bench/benc_cli_ipfs_add some code that seems like it will work
void has joined #ipfs
<jbenet>
torpor: oooh good point not sure if we have init programmatically callable yet
<jbenet>
torpor: take a look at iptb
<jbenet>
github.com/whyrusleeping/iptb
<torpor>
ok thanks will
<torpor>
do
<torpor>
looks like it calls out to the binary too
<torpor>
but shouldn't i theoretically just be able to do the same things ipfs is doing when it handles init as an arg?
<torpor>
but through the Go module instead of syscmd?
<jbenet>
torpor: yes, you should be able to, but i think right now things may not be very happy.
<jbenet>
torpor: we need a nice programmatic interface, including "init"
<torpor>
well i will see if i can do it when i find out where the init entry point is ..
<torpor>
this is for ipfs on ios btw .. i'd rather like to not have to call out to the ipfs bin on ios ..
nessence has joined #ipfs
www has quit [Ping timeout: 246 seconds]
lgierth has quit [Quit: Ex-Chat]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has joined #ipfs
<cryptix>
torpor: i'd like to pick your brain on getting the toolchain working for that but i dont have time right now.. :/
<cryptix>
and yea - init is a bit burried ..
<torpor>
well i'll be around
<torpor>
i'm trying to build ipfs for iOS and its working - at least i can build ipfs and run it as a module on iOS, passing data back and forth between Obj-C and Go runtime envs
<torpor>
and now i'm just trying to build a little app that ipfs inits, takes a picture with the camera, and serves it up over ipfs ..
<torpor>
the toolchain is actually function but you have to have a bit of glue in place to do the obj-c/go bridging
<cryptix>
torpor: is the x/mobile bind stuff working already?
<torpor>
i don't know what x/mobile bind stuff means
<torpor>
but if you mean binding Go calls to Obj-C environment: yes.
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
<torpor>
i'm now writing an ipfs app in go for ios and building successfully, etc.
<whyrusleeping>
torpor: dhthell (one of my other programs) used to set up multiple ipfs instances in a single binary and run the,
<whyrusleeping>
them*
<whyrusleeping>
its definitely doable
<whyrusleeping>
i used the daemons themselves for iptb so i could inspect things more easily
<whyrusleeping>
using tools we already built (the CLI)
www has joined #ipfs
<torpor>
ok good to know
<torpor>
i'll have a closer look after this hack
Wallacoloo has joined #ipfs
<whyrusleeping>
torpor: awesome :)
<whyrusleeping>
torpor: also, i'm super happy youre doing iOS
<whyrusleeping>
can you post whatever you do to github?
<torpor>
yeah its already up there
<torpor>
i'm just trying to push forward to get to the point where there is a working ipfs example with an iOS gui before i promote it though
<torpor>
ultimately i'd like to have an ipfs-iOS native client working this week
patcon is now known as patcon|afk
<torpor>
i think that'd be good for the ipfs project
<torpor>
and if you want to add it to the project i'm happy to work in that direction too - i.e. include it as an example
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]