<orzo>
i do think the way go uses git for its package management is forward-looking
<whyrusleeping>
zignig_: yeah, it doesnt quite do what i want
<whyrusleeping>
it doesnt really allow you to vendor your libraries
<whyrusleeping>
which arguably 'doesnt need to happen'
<orzo>
i mean, maybe not the exact way, but the idea of git intead of debian's versioned tar ball assumption
<whyrusleeping>
its aimed at managing a single project
<whyrusleeping>
orzo: yeah, its nice, it also uses hg, and other vms's
<orzo>
its hard to get traction for such ideas in debian dev hangouts
<orzo>
source packages are obsolete at this point in my view
jbenet_ has joined #ipfs
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
<orzo>
debian's package management sufferes from legacy bloat
<orzo>
what distro is popular here among ipfs devs?
<zignig_>
orzo: it does , but it is a good system.
<zignig_>
I use debian and ubuntu at work and at home.
notduncansmith has joined #ipfs
<zignig_>
orzo: debian and ubuntu need a spring clean , and to a certain extent a seperation of server / desktop software.
notduncansmith has quit [Read error: Connection reset by peer]
<orzo>
why server/desktop
<orzo>
do you package any of your own work for debian, zignig_ ?
<zignig_>
I haven't , I just use other peoples packages.
<whyrusleeping>
orzo: i use arch, jbenet uses OSX
<orzo>
ipfs-capable apt could make forking debian cheaper
<zignig_>
then use config mangagement systems to set the machines up.
<whyrusleeping>
ipfs-capable apt would be very nice
<zignig_>
orzo: yes , been thinking about how to do that....
<zignig_>
whyrusleeping: did you get 'ipfs save' going ?
<zignig_>
whyrusleeping: I think it should be 'ipfs insert'
<orzo>
i've been essentially forking debian
<orzo>
even though we didn't originally set out to make a traditional debian-fork
<whyrusleeping>
zignig_: insert would be a good name for it
<whyrusleeping>
its still stuck right now because people have a thousand different ideas about how it should work
<zignig_>
indeed , inser a hash into a dag and get the new hash back , easy ! ;P
<zignig_>
*insert
<whyrusleeping>
yeah, it really should be transactional
<whyrusleeping>
i.e.
<whyrusleeping>
start with dag X
<whyrusleeping>
perform operations []Y
<whyrusleeping>
commit, get X'
<zignig_>
yes, and be able to pull out signed history.
<orzo>
i'm interested in your ideas about ipfs-capable apt, zignig_, i'm considering starting the project
<zignig_>
I think that local to your node mutable refs would be good too.
bengo has quit [Quit: Computer has gone to sleep.]
<zignig_>
orzo: luckily debian packages are static , the only thing you need to update is the package listing file.
<whyrusleeping>
Tv`: your idea of using bloom filters for indirect blocks is appealing
inconshreveable has joined #ipfs
<Tv`>
whyrusleeping: assuming there's a good enough construction that allows combining two filters
<Tv`>
don't want to have to look up in N filters
bengo has joined #ipfs
<orzo>
i'd like my ipfs-debian installs to be usable debian forks that other machines can install
<zignig_>
Tv`: hyperloglog(s) would work quite well too.
<orzo>
so the list of installed packages and the global list of packages become equivelent as far as the softwares are concerned
<zignig_>
orzo: sounds like a plan, have a look at astralboot , ( http://github.com/zignig/astralboot ) , its a boot server that uses ipfs , and does debian...
<orzo>
the global list is mere social convention
bengo has quit [Client Quit]
<orzo>
does "golang server" just mean a server written in go?
<orzo>
or is there a golang protocol?
<zignig_>
it's written in golang
<orzo>
is that the prefered name of the language
<zignig_>
yes , mainly because 'go' is too short to search on google... ;)
HastaJun has joined #ipfs
<orzo>
yeah, 'go' was a stupid name from the beginning
<orzo>
i don't liek the name of my project 'samizdat' but my partner is heavily attached to it
<zignig_>
they should have called it c_like_concurrent_garbage_collected_nifty_network_programming_language
<zignig_>
clcgcnnpl for short....
<orzo>
heh
<orzo>
ipfs is a good name
notduncansmith has joined #ipfs
<orzo>
do you use astraboot for anything practical?
notduncansmith has quit [Read error: Connection reset by peer]
<orzo>
we map our node ids into ipv6 addresses and make an ipsec vpn where the kernel does the encryption
<orzo>
how compatible with ipfs is that? Can ipfs use ipsec as the underlying transport?
<orzo>
i think its superior to use the kernel over user space layering
<orzo>
os x, linux, and windows all have kernel support for ipsec
<zignig_>
orzo: yes I'm doing some investigations into coreos clusters for file processing.
<zignig_>
It makes deployment fast and simple for virtual machines.
<orzo>
i cloned it into my src/ directory again
<orzo>
sigh
Wallacoloo has quit [Ping timeout: 256 seconds]
nessence has quit [Remote host closed the connection]
bengo has joined #ipfs
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
LordBrain has joined #ipfs
bengo has quit [Quit: Computer has gone to sleep.]
lgierth_ has quit [Quit: Ex-Chat]
LordBrain has quit [Quit: Leaving]
neoteo has joined #ipfs
neoteo has quit [Ping timeout: 272 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<orzo>
is a clone of go-ipfs hosted on ipfs? If not, why not?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
tilgovi has quit [Ping timeout: 272 seconds]
bengo has joined #ipfs
domanic has joined #ipfs
EricJ2190 has quit [Ping timeout: 265 seconds]
Wallacoloo has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<Blame>
The only real answer to that is, it would require constant updates
therealplato1 has joined #ipfs
<Blame>
but that is what ipns is for
<Blame>
and heck, updating it could be automated
therealplato has quit [Ping timeout: 245 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
nessence has joined #ipfs
nessence has quit [Read error: Connection reset by peer]
bengo has quit [Quit: Computer has gone to sleep.]
nessence has joined #ipfs
bengo has joined #ipfs
bengo has quit [Client Quit]
nessence has quit [Ping timeout: 265 seconds]
<doublec>
A single ipfs daemon instance can't host multiple ipns keys though can it? It uses the key of the node itself?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping>
doublec: not yet
<whyrusleeping>
but that is planned
<doublec>
nice
notduncansmith has joined #ipfs
neoteo has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
neoteo has quit [Ping timeout: 276 seconds]
Wallacoloo has quit [Ping timeout: 264 seconds]
tperson has quit [Ping timeout: 264 seconds]
tperson has joined #ipfs
Wallacoloo has joined #ipfs
jbenet_ has quit [Ping timeout: 246 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping>
jbenet: i'm running into something weird
<whyrusleeping>
my new bitswap thing is trying to send a message to a peer that we dont have a connection to, or addresses for
<whyrusleeping>
im not sure how it knows about that peer... unless its connected to us
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
bengo has joined #ipfs
bengo has quit [Client Quit]
inconshr_ has joined #ipfs
inconshr_ has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
inconshr_ has joined #ipfs
inconshreveable has quit [Ping timeout: 256 seconds]
inconshr_ has quit [Ping timeout: 256 seconds]
<jbenet>
Hmmmmmmm yeah that's very odd.
<jbenet>
Maybe it found a provider record
<whyrusleeping>
oooh, maybe?
<whyrusleeping>
thats weird though
<whyrusleeping>
the things i'm looking for only exist on one machine
<whyrusleeping>
also, weird, calling 'connect' through the routed host *should* trigger a findpeer
<whyrusleeping>
but it doesnt appear to
neoteo has joined #ipfs
inconshreveable has joined #ipfs
neoteo has quit [Ping timeout: 240 seconds]
sharky has quit [Ping timeout: 272 seconds]
sharky has joined #ipfs
patcon has joined #ipfs
jbenet_ has joined #ipfs
<whyrusleeping>
note to self, it helps if the daemon youre requesting content from is on
<whyrusleeping>
marginally
<whyrusleeping>
yay progress!
Wallacoloo has quit [Quit: Leaving.]
pfraze_ has quit [Remote host closed the connection]
<whyrusleeping>
jbenet: from what i'm seeing here, bitswap is slow because for some reason, when we send out wantlist messages, it takes them a while to arrive on the other side...
<whyrusleeping>
or takes us a while to send them out
<whyrusleeping>
kinda weird.
<whyrusleeping>
still investigating
torpor has joined #ipfs
www1 has quit [Ping timeout: 265 seconds]
inconshreveable has quit [Remote host closed the connection]
<whyrusleeping>
this requires that bob knows alice and her public key beforehand, correct?
<domanic>
whyrusleeping, the client needs to know the server's pubkey
<domanic>
but the client sends her pubkey
<domanic>
consider the server's pubkey to be an access capability
<whyrusleeping>
okay
<domanic>
that is important, because it allows the client to prove they did not dial a wrong number (or are just war dialing)
<whyrusleeping>
yeah, i like
<domanic>
I just need to write up exactly what happens in a mitm attack and at what point the client/server realize they are being middled
www has joined #ipfs
mildred has joined #ipfs
www has quit [Ping timeout: 244 seconds]
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to refactor/bitswap: http://git.io/vUuy1
<ipfsbot>
go-ipfs/refactor/bitswap 006b124 Jeromy: correctly buffer work messages
<whyrusleeping>
get it ipfsbot
<whyrusleeping>
you go girl
bengo has joined #ipfs
nessence has joined #ipfs
Tv` has quit [Quit: Connection closed for inactivity]
konubinix has quit [Remote host closed the connection]
konubinix has joined #ipfs
domanic has quit [Ping timeout: 246 seconds]
neoteo has joined #ipfs
nessence has quit [Read error: Connection reset by peer]
neoteo has quit [Ping timeout: 246 seconds]
Kuba has quit [Remote host closed the connection]
www has joined #ipfs
nessence has joined #ipfs
atomotic has joined #ipfs
Wallacoloo has quit [Ping timeout: 265 seconds]
chrisr_ has joined #ipfs
Wallacoloo has joined #ipfs
<cryptix>
gmorning
<cryptix>
sorry guys, my alarm clock and i failed terribly yesterday
<jbenet>
cryptix o/ morning!
<jbenet>
no worries-- it happens :) -- may want to check in with whyrusleeping re sprint stuff
<jbenet>
btw whyrusleeping this listener handshake accept problem is so complicated.
<jbenet>
preserving the exact listener / accept semantics is non trivial. i may do the easy thing-- which is what golang/crypto/tls does -- defer running the handshake until someone reads or writes to the conn.
<whyrusleeping>
that would probably fix it
<whyrusleeping>
cryptix: heyo! hope youre well rested :P
www has quit [Ping timeout: 245 seconds]
<whyrusleeping>
if you could take a look at https://github.com/ipfs/pm/issues/8, i put a couple issues under your name, let me know if you can get to those this week
<cryptix>
i just updated my freebsd ws to latest master
<AtnNn>
It looks similar to ipfs, but it's interesting how ipfs promotes itself as "the permanent web" but that demo is called "ephemeralp2p"
<headbite>
so the idea with ephemeralp2p is you paste in some html or whatever and it makes a hash. that stays live as long as one person is viewing the page and after the last person closes their browser the page goes away.
torpor has left #ipfs [#ipfs]
<okket_>
but the data always goes through the main central server, it only fetches the content from a different source instead of its own db / filesystem
<AtnNn>
it could use some webrtc
<AtnNn>
ipfs seems to have the same ephemeral nature: if everyone viewing the page unpins + gcs, or disconnects, it goes away
<okket_>
maybe, but currently it is only a hypes varnish with a decentralized backend
<okket_>
s/hypes varnish/hyped web proxy/
<krl>
ipfs should really have something similar to a local hostfile
pfraze_ has joined #ipfs
patcon has quit [Ping timeout: 265 seconds]
pfraze_ has quit [Remote host closed the connection]
Tv` has joined #ipfs
<grawity>
wouldn't that end up like the centrally-distributed host files i2p has
<grawity>
1980's and <SRI-NIC>HOSTS.TXT all over again
patcon has joined #ipfs
mildred has quit [Ping timeout: 276 seconds]
<whyrusleeping>
krl: what do you mean by a local hostfile?
<krl>
like etc/hosts that can map a string to a hash
<whyrusleeping>
oh, i was going to implement aliases
<krl>
at best, so you can have local aliases, and also exported aliases in a public /ipns/names/
warner has quit [Quit: ERC (IRC client for Emacs 24.5.1)]
<whyrusleeping>
yeah, i want to have something like that too
<whyrusleeping>
jbenet: good job on the listener fix :)
<wking>
how are "exported aliases" different from /ipns/<you>/<alias> being a link to an IPNS entry? That's the CA-like setup outlined in the ipfs-cap2pfs paper
warner has joined #ipfs
<jbenet>
Yeah, I think the only difference may be implementation details / use case
<jbenet>
We were discussing details before around remotes, and yesterday about making the repo an object. I think we should have the remotes listed on "<repo>/remotes"
<jbenet>
(Let's borrow from git here too if we can. May be useful to be able to set preferred addrs for hosts)
<jbenet>
whyrusleeping: thanks! Do you feel it's better now?
<wking>
jbenet: link to remotes/repo-objects discussion?
<jbenet>
wking yesterday we were talking about it in the sprint. I can write something up today-- gist is to try and represent the entire repo as an ipfs object (or as close as we can) so that state could be a single hash (or close to)
<wking>
re, #1208, I'm jumping through a few hoops to avoid implementing per-protocol namespaces ;). "We'll deal with this later" is fine, but it feels sort of like "we'll do more work now to avoid moving in the direction we agree is useful" ;)
<jbenet>
It's all in the timing
<jbenet>
It's not important enough yet, other things are much higher prio to get to
<jbenet>
(And the protofs stuff is large in its own right)
inconshreveable has joined #ipfs
hellertime has quit [Quit: Leaving.]
hellertime has joined #ipfs
gatesvp has quit [Quit: Page closed]
nessence_ has quit [Remote host closed the connection]
zignig_ has quit [*.net *.split]
nemik has quit [*.net *.split]
amiller has quit [*.net *.split]
lachenmayer has quit [*.net *.split]
hij1nx has quit [*.net *.split]
hij1nx has joined #ipfs
zignig has joined #ipfs
lachenmayer has joined #ipfs
nemik has joined #ipfs
Guest24344 has joined #ipfs
Guest24344 is now known as amiller_
amiller_ has joined #ipfs
amiller_ is now known as amiller
inconshreveable has quit [Remote host closed the connection]
<rschulman>
o/ everyone
<rschulman>
been out of pocket, on paternity leave. :)
<whyrusleeping>
rschulman: heyo! welcome back
<rschulman>
thanks. :)
<rschulman>
how are you doing?
<whyrusleeping>
doing pretty well!
<whyrusleeping>
fixing bugs, making things fast
<whyrusleeping>
how about you?
<rschulman>
good stuff
<rschulman>
busy
<rschulman>
trying to move my family across the country
<rschulman>
(do we need a Washington, DC IPFS meetup?)
<whyrusleeping>
haha, if you want to set one up that would be cool
<whyrusleeping>
although, i might try setting one up in seattle first
<whyrusleeping>
and see how that goes
<rschulman>
hehe
<rschulman>
why not?
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to refactor/bitswap: http://git.io/vUaS0
<ipfsbot>
go-ipfs/refactor/bitswap af66214 Jeromy: more cleanup, and improve test perf
<whyrusleeping>
i'm not sure how to go about setting them up, really
<whyrusleeping>
like, how do you advertise to people in the area that arent in this channel?
<rschulman>
oh, I would probably just use meetup.com
<rschulman>
already built in.
<whyrusleeping>
ooooooh
<whyrusleeping>
yeah
<whyrusleeping>
thats a thing isnt it
<rschulman>
totally
<whyrusleeping>
well, the more meetups, the merrier
<rschulman>
that's my approach!
cow_2001 has quit [Quit: ASCII Muhammad - @o<-<]
jbenet_ has joined #ipfs
Wallacoloo has joined #ipfs
jbenet_ has quit [Ping timeout: 250 seconds]
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to refactor/bitswap: http://git.io/vUVG6
<ipfsbot>
go-ipfs/refactor/bitswap b9063b4 Jeromy: remove unneccessary test
bengo has joined #ipfs
bengo has quit [Client Quit]
bengo has joined #ipfs
hellertime has quit [Read error: Connection reset by peer]
hellertime has joined #ipfs
bengo has quit [Client Quit]
hellertime has quit [Client Quit]
hellertime has joined #ipfs
bengo has joined #ipfs
bengo has quit [Client Quit]
bengo has joined #ipfs
bengo has quit [Quit: Computer has gone to sleep.]
inconshreveable has joined #ipfs
tilgovi has quit [Ping timeout: 255 seconds]
inconshreveable has quit [Read error: Connection reset by peer]
inconshr_ has joined #ipfs
jbenet_ has joined #ipfs
Wallacoloo has quit [Ping timeout: 256 seconds]
nessence has joined #ipfs
bren2010 has quit []
grncdr has joined #ipfs
rschulman1 has joined #ipfs
rschulman1 has quit [Client Quit]
bren2010 has joined #ipfs
bren2010 has quit [Client Quit]
inconshr_ has quit [Remote host closed the connection]
bren2010 has joined #ipfs
bengo has joined #ipfs
inconshreveable has joined #ipfs
tilgovi has joined #ipfs
grncdr has quit [Ping timeout: 272 seconds]
amiller has quit [Excess Flood]
RzR has quit [Excess Flood]
hellertime has quit [Quit: Leaving.]
inconshreveable has quit [Remote host closed the connection]
amiller has joined #ipfs
hellertime has joined #ipfs
RzR has joined #ipfs
bengo has quit [Quit: Computer has gone to sleep.]
<whyrusleeping>
i can tell you where it lives from what api call its making
<whyrusleeping>
but i am unfamiliar with the webui codebase
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to refactor/bitswap: http://git.io/vUwgZ
<ipfsbot>
go-ipfs/refactor/bitswap cece493 Jeromy: revert changes to bitswap message, they werent necessary
pfraze_ has joined #ipfs
<whyrusleeping>
jbenet: on your protocol streams thing, you should include both an EOF signal, and a 'suspend' signal
<whyrusleeping>
somehow
<whyrusleeping>
well... suspend would be on the connection as a whole
<whyrusleeping>
also, ive been thinking about the EOF problem, and its easier than we thought
ehmry has joined #ipfs
inconshreveable has quit [Read error: Connection reset by peer]
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
inconshreveable has joined #ipfs
<doublec>
How does one build/run ipfs from git master? I assume there's some go command to do it? I see instructions for "got get -u ..." but I want to test local changes.
timgws has joined #ipfs
<wking>
doublec: 'make install' or 'make build' work for me, but you need to put your Git repository into the right location in your GOPATH ($GOPATH/src/github.com/ipfs/go-ipfs/)
cow_2001 has joined #ipfs
<whyrusleeping>
doublec: after doing 'go get github.com/ipfs/go-ipfs'
<whyrusleeping>
the source will be in your GOPATH (as wking says)
<whyrusleeping>
the go command, if you want to avoid makefiles, is 'go install ./cmd/ipfs'
<whyrusleeping>
or, you can 'go build ./cmd/ipfs' to just make a binary next to you
<doublec>
ah, the gopath stuff is what I was missing - thanks all!
timgws has quit [Quit: Computer has gone to sleep.]
<Blame>
no. We need a browser that allows a webpage to open and listen on a port on the client (and also support UPnP and maybe nat busting)
<Blame>
WebRTC still needs a meet-in-the-middle server
<Blame>
Its the only way to have truly decentralized broswer applications
<doublec>
Blame: is it the use of the stun server you find centralized in webrtc?
<Blame>
essentially. I'm ok with an inital "meet in the middle" but I need to be able to create my own "meet in the middle" that can join the same network
<Blame>
so essentially I think the solution is making a decentralized nework of stun servers that allow anoybody to connect to anybody
inconshreveable has quit [Remote host closed the connection]
<Blame>
Another option is to jsut never try and put a server in a browser. Establish a client-only level of involvment in a p2p system and let browser be at that tier
<Blame>
which I feel is more practical.
<Blame>
but runs affoul of same-origin policy
<Blame>
Browser security as evolved (likely the correct way) with an assumption of centrality that leads to treating true p2p applications as insecure.
<Blame>
(which, in a lot of ways, they are)
<Blame>
which leaves: make a new browser or make everybody run a proxy.
<Blame>
ipfs chose the latter
<whyrusleeping>
'proxy'
jbenet_ has joined #ipfs
<whyrusleeping>
i dont know if you guys have kept up on servo and browser.html
<Blame>
it really is a proxy (from the browser point of view)
<whyrusleeping>
but they are doing some really neat things that we will be able to take advantage of
<whyrusleeping>
yeah, it is, i agree
<doublec>
whyrusleeping: what sort of things?
<doublec>
I've used servo-shell but haven't followed browser.html work
<whyrusleeping>
the ability to open up TCP sockets and link binaries in
<Blame>
oooh
<whyrusleeping>
you get direct access to the network stack in javascript
bengo has quit [Quit: Computer has gone to sleep.]