roguism has quit [Read error: Connection reset by peer]
roguism has joined #ipfs
jhulten has joined #ipfs
fleeky__ has quit [Quit: Leaving]
fleeky has joined #ipfs
roguism has quit [Ping timeout: 252 seconds]
jhulten has quit [Ping timeout: 240 seconds]
roguism has joined #ipfs
fleeky_ has joined #ipfs
konrad__ has quit [Ping timeout: 250 seconds]
NeoTeo has quit [Quit: ZZZzzz…]
Guest18335 is now known as sasha
simonv3 has quit [Quit: Connection closed for inactivity]
<whyrusleeping>
i'm getting 5-6% packet loss on my internet connection
<whyrusleeping>
this is so much fun
Dawn has joined #ipfs
Jeanne has quit [Ping timeout: 240 seconds]
jabberwocky has quit [Remote host closed the connection]
pfraze has quit [Remote host closed the connection]
merlijn_ has joined #ipfs
merlijn_ has quit [Ping timeout: 246 seconds]
vanila has joined #ipfs
<vanila>
hello
<vanila>
does anyoen have freeBSD conf files for starting up ipfs automatically?
jabberwocky has joined #ipfs
<whyrusleeping>
vanila: i dont know of any
<whyrusleeping>
if you write some definitely let us know
<vanila>
okay
jabberwocky has quit [Ping timeout: 272 seconds]
<davidar>
whyrusleeping (IRC): the other day I was getting better bandwidth on my connection going through a VPN than without it :/
jhulten has joined #ipfs
<whyrusleeping>
davidar: thats... concerning
fleeky has quit [Quit: Leaving]
<davidar>
whyrusleeping (IRC): yeah, I'm not even sure what to think anymore
<davidar>
It was probably slowed down by logging all my metadata. Hooray for democracy!
jhulten has quit [Ping timeout: 244 seconds]
<whyrusleeping>
lol
<whyrusleeping>
i'm supposed to have 12mbit of bandwidth
<whyrusleeping>
but if i try to push 12mbit of packets out to a server, i get over 80% of them dropped
<davidar>
millibit?
<davidar>
lol
<whyrusleeping>
megabit, lol
<lgierth>
davidar: try cjdns iptunnel :P
<davidar>
They didn't say *useful* bandwidth :p
<lgierth>
23:26 <@jercos> well, cjdns is now officially more reliable than the open internet for getting to my cheaper VPSes :|
<davidar>
lgierth (IRC): yeah, I keep meaning to set that up
<whyrusleeping>
if running a cjdns tunnel improves my connection, i'm going to be really upset
* whyrusleeping
sets up cjdns
<ansuz>
:D
<davidar>
Looks like a mild pita though :/
<whyrusleeping>
lgierth: stick around, i will need you for cjdns questions
<lgierth>
just came back from a party, will be here for another 2 hours or so
<davidar>
lgierth (IRC): can you run cjdns on Android?
<lgierth>
davidar: yeah!
<lgierth>
if you have root
<davidar>
Nope
<davidar>
Samsung :/
<lgierth>
non-root is totally doable too, someone just needs to do it
<whyrusleeping>
cjdns is written in node?
<lgierth>
the build system and tools are
<lgierth>
cjdns itself is c
<whyrusleeping>
ooooh, okay
<whyrusleeping>
i was about to say, i can make things faster >.>
<lgierth>
:)
<lgierth>
i've noticed a few times how my regular internet broke and cjdns continued to work
<whyrusleeping>
lgierth: once its installed, is it much more difficult to use it for tunneling my traffic?
<lgierth>
regular internet = tcp
<davidar>
lgierth (IRC): is it possible to have multiple cjdns iptunnels, so that it gets routed to the nearest node?
<davidar>
Gateway node closest to the clear net server I'm tunneling to, that is
<lgierth>
you'll need an iptunnel server, and configure your cjdns to use that server. it'll get addresses from the server, and you'll just have to point your default route to the tun0 interface
<lgierth>
davidar: no discrimnation between different iptunnel servers yet :(
<lgierth>
a more sophisticated iptunnel is very high on my wishlist
<lgierth>
i'd like cjdns to be able to replace bgp
<davidar>
lgierth (IRC): because that would be a killer feature :D
<lgierth>
word!
<lgierth>
thefinn93: got any cjdns node in the seattle area?
<ansuz>
ohai thefinn93
<davidar>
Especially if you could have gateway nodes in the same datacenter
<thefinn93>
lgierth: eh
<thefinn93>
seanode?
merlijn_ has joined #ipfs
<lgierth>
thefinn93: o/ is it any good for iptunnel?
<thefinn93>
ehh
<thefinn93>
proly not
<thefinn93>
it's OpenVZ
<lgierth>
otherwise i'd say i just set that up on the ipfs gateways
<davidar>
lgierth (IRC): that'd be cool
<davidar>
lgierth (IRC): how do I peer with the ipfs gateways BTW?
nicknikolov has joined #ipfs
<whyrusleeping>
alright, setting up cjdns on a machine in seattle
<lgierth>
right now you'd need access to the infrastructure secrets
<lgierth>
let me figure out how to give peering to friends\
merlijn_ has quit [Ping timeout: 250 seconds]
<lgierth>
(how to do it properly)
TheWhisper has quit [Quit: Leaving]
<lgierth>
the neat thing is that you don't need to be peered to your iptunnel server
<lgierth>
it's best for performance obviously, but it's not neccessary
<lgierth>
whyrusleeping: *you* can already peer with the gateways
<lgierth>
or any i.ipfs.io host for that matter
<whyrusleeping>
right, but we dont have any in seattle afaik
<whyrusleeping>
and SF adds a bit more latency than i'd like
<lgierth>
nope only sfo
<lgierth>
yeah
<whyrusleeping>
i'm 25ms to my seattle node
<lgierth>
25 ms in the same city!
<whyrusleeping>
lol, i'm not in seattle at the moment
<lgierth>
the dsl network here is just as bad...
<whyrusleeping>
not until january
<lgierth>
freifunk to the resque
<lgierth>
ah ok ;)
<whyrusleeping>
otherwise i'd go chill with mappum more often
TheWhisper has joined #ipfs
<davidar>
25ms would be a good ping for me :/
ekaron has joined #ipfs
<davidar>
45ms ping to the university (20 mins away)
<lgierth>
the local community network here (freifunk) is so much better than the dsl providers
<lgierth>
bandwidth a bit less stable, but the latency!
<davidar>
i wish we still had a meshnet :(
<lgierth>
35 ms to amsterdam vs. 18 ms
<davidar>
apparently we had a reasonably good one here like 10yrs ago
<lgierth>
3 hops to the local IX
<lgierth>
seattle?
<kpcyrd>
lgierth: nice
<lgierth>
kpcyrd: yeah i wanna hook up to it again :(
<lgierth>
i was hooked up right to the backbone last year, for one day
<whyrusleeping>
i need to replace cat with a version of cat that cowardly refuses to spit out binaries to my terminal
<whyrusleeping>
'cat cjd<TAB><ENTER>'
<davidar>
hehe
<whyrusleeping>
whoops, theres the entire binary, thanks. oh look, terminal control characters, great, huh, my terminal is now unusable
<davidar>
that's why you use less ;)
<kpcyrd>
whyrusleeping: type `reset`
nicknikolov is now known as nicknikolov2
<whyrusleeping>
kpcyrd: yeah, still sucks to have to wait for it to finish spitting things to the screen before anything i type gets accepted
<kpcyrd>
lgierth: we're peered to IX, but most nodes are not connected to the directional radio network
jabberwocky has joined #ipfs
<whyrusleeping>
lgierth: i have cjdns installed on my desktop and my server
<kpcyrd>
so it's going though the home uplinks to our servers first
<whyrusleeping>
how do i find out what port number to use?
<lgierth>
whyrusleeping: UDPInterface.bind on the server
<davidar>
whyrusleeping (IRC): it's in the config
<lgierth>
kpcyrd: HHIX or something like that??
nicknikolov2 has quit [Remote host closed the connection]
<lgierth>
kpcyrd: i looked on the hh map the other day and it looked like there quite a few wifi links now
<ansuz>
germany has some problems with streaming
<ansuz>
if you're looking to put youtube on ipfs
<davidar>
ansuz: what kind of problems?
<ansuz>
all the problems
<davidar>
as in, technical, or geoblocking?
<ansuz>
"you can't stream this from your country" problems
<davidar>
haha
<davidar>
I'm in australia
<lgierth>
there's a non-governmental government agency that assumes it's the rightsholder for all music, by default
<davidar>
we get that message all the freaking time
<lgierth>
a.k.a GEMA
<ansuz>
<3
<davidar>
lgierth: isn't that the case in most countries?
<davidar>
RIAA, etc
<lgierth>
it's a bit worse. every artist is a member of gema by default, and you have to prove that the artist isn't if you wanna wand off their attacks
<lgierth>
artist name doesn't work, it has to be the full name
<lgierth>
so you can play a completely CC licensed party, and they'll still come to collect royalty
<ansuz>
guilty until proven innocent
<ansuz>
nice
<lgierth>
unless you get the fullname of every single artist, so that gema can check they're *not* members
<lgierth>
google basically said "ok fuck off then"
<davidar>
lgierth: surely it wouldn't be too hard to write a script to automate most of this? :/
<lgierth>
indeed
<lgierth>
:)
<davidar>
lgierth: pretty please write said script :)
<kandinski>
davidar: thanks. Difficult to meet then, particularly as my job is PST too, so my few overlapping hours are commited.
<kandinski>
But the py-ipfs plan seems good.
<whyrusleeping>
lgierth: i'm assuming i should replace eth0 with whatever interface gives me my external ip?
<lgierth>
whyrusleeping: yes
<kandinski>
Am I correct in understanding that libp2p is in js-ipfs, but go-ipfs still doesn't have libp2p refactored out yet?
<davidar>
kandinski: go-libp2p sort of exists
<davidar>
but go-ipfs isn't using it yet afaik
<lgierth>
you can do the whole dance with ipv6 instead of ipv4 too, that'll let you omit masquerading, if you can hand public ipv6 addresses to the clients
<whyrusleeping>
lgierth: okay, cool. now on the client one, the first two things it says are done automatically
<whyrusleeping>
where does the ip address in it come from?
<whyrusleeping>
is that the one that my tun0 locally has?
<lgierth>
from cjdroute.conf on the server
<lgierth>
oh, yeah
<whyrusleeping>
lgierth: what about the addresses in the UDP peering stuff?
<lgierth>
these are the addresses of you local node's peers
<lgierth>
so that you can still reach them
<whyrusleeping>
ah, okay cool
<whyrusleeping>
hrm...
<kandinski>
davidar: sorryfor the very random questions, I'm trying to get up to speed and still all over the place.
<kandinski>
so how important is base58 to IPFS?
cemerick has joined #ipfs
Senji has joined #ipfs
<kandinski>
I realise everything is swappable, hence multihash etc., but is base58 a preferred alphabet for encoding IPFS hashes?
<whyrusleeping>
if things work correctly, my external ip address (to the rest of the internet) should appear to be that of my server, right?
<lgierth>
yeah
<lgierth>
first milestone is pinging client<->server
<whyrusleeping>
hrm, my server has two tun interfaces now
<lgierth>
ps aux | grep cjdroute has two core processes?
<lgierth>
is that a linux over there?
<whyrusleeping>
hrm, gonna try and restart
<lgierth>
the run interface should always disappear if the process dies
<whyrusleeping>
tun1 is getting the cjdns fc00/8 addr
<whyrusleeping>
>.>
<lgierth>
and tun0 is openvpn or something else?
<whyrusleeping>
i've no idea really
<whyrusleeping>
it has the 172 address i gave it with the commands you linked
<lgierth>
oh sorry, the tun0 in these commands is the cjdroute tun interface
<davidar>
kandinski: not a problem :)
<lgierth>
so it seems you ran these before starting cjdroute, which saw the existing interface and made its own
<whyrusleeping>
yeah, i thought it was before
<davidar>
base58 is preferred, but not mandatory
<whyrusleeping>
lgierth: any way to reset tun0 back to blank?
<davidar>
well...
<davidar>
i think it's mandatory in /ipfs/... paths
<davidar>
whyrusleeping: ^ ?
<lgierth>
whyrusleeping: you can point cjdroute to an existing tun interface, router.interface.tunDevice
<whyrusleeping>
davidar: yeah, at least for current tooling
<whyrusleeping>
lgierth: no i dont have any reason for devices to exist already
<davidar>
kandinski: yeah, so just use base58 unless you have a good reason not to :p
<whyrusleeping>
i'd rather keep just one tun device
<davidar>
and/or bug whyrusleeping to support other encodings ;)
<whyrusleeping>
lol
<lgierth>
right, mh, ip tuntap del tun0, something like that?
<kandinski>
davidar: right. I asked because of the python multihash discussion
<kandinski>
davidar: (whether base58 should or not be part of the multihash library)
<davidar>
kandinski: yeah, i think encoding should be part of the library
<davidar>
(with the option to extend it to other encodings in the future)
<whyrusleeping>
lgierth: i can ping my servers cjdns ipv6
<lgierth>
and the ipv4?
<whyrusleeping>
which ipv6?
<whyrusleeping>
er 4
<lgierth>
172.23.23.1 in the tl;dr i linked
<whyrusleeping>
yes, thats pinging myself
<lgierth>
oh i mean from the client
<lgierth>
i.e. the network within the tunnel
<whyrusleeping>
the 172.23.23.1 was the ip given to my client wasnt it?
<whyrusleeping>
thats the one that shows up in my ip addr locally
<lgierth>
oh ok, you set .1 in router.ipTunnel.allowedConnections on the server? that should be anything but .1
<lgierth>
in the end it doesn't really matter that the server is exactly .1, just needs some address that the client can reach
<lgierth>
*can reach via the tunnel
<whyrusleeping>
okay, i get it
<whyrusleeping>
where does the server get the .1 from?
<lgierth>
by hand
<whyrusleeping>
ah
<whyrusleeping>
okay
<whyrusleeping>
messed that bit up then
<whyrusleeping>
WOO!
<whyrusleeping>
it appears to be working!
drathir has quit [Ping timeout: 264 seconds]
kpcyrd has quit [Ping timeout: 264 seconds]
d6e_ has quit [Ping timeout: 264 seconds]
pjz has quit [Ping timeout: 264 seconds]
sikander has quit [Ping timeout: 264 seconds]
wiedi has quit [Ping timeout: 264 seconds]
lgierth has quit [Ping timeout: 264 seconds]
kpcyrd has joined #ipfs
drathir has joined #ipfs
pjz has joined #ipfs
sikander has joined #ipfs
NightRa has joined #ipfs
jhulten has joined #ipfs
<whyrusleeping>
i also love that my ipfs node didnt care about all the network shit i was doing at all
jhulten has quit [Ping timeout: 240 seconds]
<kandinski>
davidar: the py-ipfs discussion was about the multihash API, people rightly wanted it to mimic the Standard Library's hashlib API
<kandinski>
davidar: so I guess it's a matter of how to extend that API gracefully to include IPFS-standard behaviours.
lgierth has joined #ipfs
<davidar>
kandinski: what does the stdlib do with encodings?
d6e_ has joined #ipfs
<kandinski>
the discussion was about multihashlib, and having it follow PEP247 (Python StdLibrary's hashlib) API.
<kandinski>
the hashlib already offers one standard encoding, which is a hex digest
<davidar>
you can always convert to b58 afterwards too i guess
<davidar>
would be nicer to have it in the lib though i agree
jamie_k has quit [Quit: jamie_k]
<kandinski>
something like m.b58digest(), yeah
<kandinski>
that was my point
<kandinski>
rather, that was why I was asking how privileged base58 was. If it's compulsory in /ipfs/... filepaths, then very
merlijn_ has joined #ipfs
reit has joined #ipfs
Senji has quit [Ping timeout: 265 seconds]
<davidar>
yeah
merlijn_ has quit [Ping timeout: 250 seconds]
<kandinski>
I see a difference between the repo structure in py-ipfs and js-ipfs
pfraze has quit [Remote host closed the connection]
pfraze has joined #ipfs
s_kunk has joined #ipfs
merlijn_ has joined #ipfs
pfraze has quit [Remote host closed the connection]
mildred has joined #ipfs
mildred has quit [Ping timeout: 260 seconds]
anticore has joined #ipfs
arrdem has quit [Ping timeout: 260 seconds]
hoony has quit [Remote host closed the connection]
wiedi has joined #ipfs
jhulten has joined #ipfs
jhulten has quit [Ping timeout: 240 seconds]
amade has joined #ipfs
e-lima has joined #ipfs
jamie_k_ has joined #ipfs
rendar has joined #ipfs
baselab has joined #ipfs
NeoTeo has joined #ipfs
M-eternaleye is now known as eternaleye
eternaleye has joined #ipfs
eternaleye has quit [Changing host]
slothbag has joined #ipfs
anticore has quit [Quit: bye]
<ipfsbot>
[go-ipfs] rht opened pull request #1968: Add gc auto test (master...gc-auto-test) http://git.io/v43fF
M-osener has joined #ipfs
ungojellym is now known as mungojelly
Encrypt has joined #ipfs
JohnClare has joined #ipfs
<JohnClare>
Hi all. Please forgive my ignorance - but is there any way to 'discover' content on remote nodes in IPFS?
dignifiedquire has joined #ipfs
dignifiedquire has quit [Client Quit]
<victorbjelkholm>
JohnClare, no, you'll ask the swarm and find peers who has the content you are requesting. You cannot see what content your peers already have
<JohnClare>
victorbjelkholm: OK - so I couldn't search through connected peers for files like: grep searchParams ipfs/connectedPeer
<JohnClare>
i'm trying to understand how I can discover content
<victorbjelkholm>
JohnClare, no, discovering would need some sort of index of files -> hashes
<daviddias>
the bitswap errors are cause I'm not connected to more peers. But whyrusleeping said that is ok :)
Dawn has quit [Killed (Sigyn (Spam is off topic on freenode.))]
conservatory has joined #ipfs
<ion>
Btw, you might want to have the npm thing use --chunker=rabin from the start (pending it becoming the default).
<victorbjelkholm>
daviddias, sweet! Let me know once I can try install some modules :)
<daviddias>
ion: I want to do both, because it will be sweet to have a graph comparing how much storage we can save from the original
<daviddias>
I'm getting a couple of modules, once in a while that are corrupt inside npm
<daviddias>
2015-11-15T14:10:00.001Z static [err] [4] file appears to be corrupt, skipping.. /npkg/-/npkg-0.0.6.tgz
<daviddias>
2015-11-15T14:10:02.131Z static [err] shasum check failed for /npkg/-/npkg-0.0.5.tgz
<daviddias>
2015-11-15T14:10:02.131Z static [err] found dd0f406e88cbb84c37d35e5a2568f0a39633f997 expected 4f9795ccca52474e6e4deae6e51b8400079ed0a4
<daviddias>
wonder what happened :)
merlijn_ has quit [Ping timeout: 250 seconds]
NeoTeo has quit [Quit: ZZZzzz…]
mungojelly has quit [Remote host closed the connection]
NeoTeo has joined #ipfs
<victorbjelkholm>
daviddias, was thinking yesterday, there is probably a lot of package you could skip
<victorbjelkholm>
like packages with one commit more than one year ago without any downloads this year
<victorbjelkholm>
there is a lot of inactive packages no one is using
<daviddias>
agree, there are indeed some optimizations that can be done that would speed things up a lot
<daviddias>
but having the information that the whole NPM is X Gb, NPM on IPFS is Y (Y<X) and that NPM on IPFS with rabin is Z (Z<Y<X) is very valuable information to show how this can be a powerful addition
jabberwocky has joined #ipfs
<daviddias>
so, I'm downloading everything for now :)
david___ has joined #ipfs
merlijn_ has joined #ipfs
<ion>
daviddias: Does the code unpack the tarballs, btw? If not, chunking by a rolling hash being useful depends on the creator of the tarball having used gzip --rsyncable (or whatever the parameter was). Pristine-tar stores files in a Git repository along with inferred parameters to gzip to achieve as close a result as possible and a binary diff to reproduce the upstream tarball exactly.
Encrypt has joined #ipfs
<david___>
i can't execute ipfs command through php on cloud ? any suggestions ?
JohnClare has joined #ipfs
JohnClare has quit [Remote host closed the connection]
JohnClare has joined #ipfs
david___ has quit [Quit: Page closed]
rombou has quit [Remote host closed the connection]
rombou has joined #ipfs
dignifiedquire- has joined #ipfs
baselab has quit [Quit: Leaving]
<dignifiedquire->
daviddias, making good progress today on npm mirror, with my performance tweaks I got a massive speed improvement and was already >10% after an hour
<ion>
nice
jhulten has joined #ipfs
dignifiedquire- has quit [Client Quit]
rombou has quit [Quit: Leaving.]
rombou has joined #ipfs
jhulten has quit [Ping timeout: 264 seconds]
Senji has joined #ipfs
rombou has quit [Quit: Leaving.]
rombou has joined #ipfs
rombou has quit [Quit: Leaving.]
rombou has joined #ipfs
rombou has quit [Client Quit]
rombou has joined #ipfs
rombou has quit [Client Quit]
rombou has joined #ipfs
JohnClare has quit [Remote host closed the connection]
merlijn_ has quit [Ping timeout: 272 seconds]
federico2 has left #ipfs ["WeeChat 1.1.1"]
<ipfsbot>
js-ipfs-api/greenkeeper-concurrently-1.0.0 b946686 greenkeeperio-bot: chore(package): update concurrently to version 1.0.0...
<ipfsbot>
[js-ipfs-api] Dignifiedquire created greenkeeper-concurrently-1.0.0 (+1 new commit): http://git.io/v4sBF
merlijn_ has joined #ipfs
<ipfsbot>
[js-ipfs-api] greenkeeperio-bot opened pull request #118: Update concurrently to version 1.0.0
border0464 has joined #ipfs
merlijn_ has quit [Ping timeout: 264 seconds]
pfraze has joined #ipfs
jhulten has joined #ipfs
conservatory has quit [Ping timeout: 260 seconds]
e-lima has joined #ipfs
jhulten has quit [Ping timeout: 276 seconds]
Pterodactylus has joined #ipfs
pfraze has quit [Remote host closed the connection]
atrapado has joined #ipfs
rombou has quit [Quit: Leaving.]
rombou has joined #ipfs
merlijn_ has joined #ipfs
merlijn_ has quit [Ping timeout: 240 seconds]
cemerick has joined #ipfs
reit has quit [Quit: Leaving]
reit has joined #ipfs
merlijn_ has joined #ipfs
Gaboose has joined #ipfs
<Gaboose>
whyrusleeping: didn't quite get your comment on https://github.com/jbenet/go-multiaddr-net/pull/12 "golang.org/x/crypto/sha3 expects to be imported by that exact path, unless its in a packages vendor directory then it doesnt matter"
<Gaboose>
it seems to me that the same rules apply to $GOPATH/src and vendor
<Gaboose>
be it $GOPATH/src/sha3 or $GOPATH/src/golang.org/x/crypto/sha3
<whyrusleeping>
Gaboose: the go1.5 vendor experiment appears to loosen the rules for canonical package names when they are in the vendor directory
<Gaboose>
you mean that vendor/golang.org/x/crypto/sha3 can be placed in vendor/sha3?
<whyrusleeping>
or vendor/QmWhatever
<Gaboose>
and you're certain that $GOPATH/src/QmWhatever doesn't work?
<whyrusleeping>
works fine there, but if you copy everything from vendor into $GOPATH/src/ it complains about 'this package expects to be imported as...'
pfraze has joined #ipfs
<Gaboose>
weird
<Gaboose>
k, thanks
<whyrusleeping>
Gaboose: yeah, its kindof annoying :/
<whyrusleeping>
if you have any other ideas, input is great
<Gaboose>
right, because ipfs-hashed packages would be nice to have globally, that's the goal right?
<whyrusleeping>
ion: ipfs has the ability to unpack tars (and repack) on the fly
<whyrusleeping>
Gaboose: yeap! thats the goal
<whyrusleeping>
the idea is that we will eventually (through ipfs) never have to worry about fetching dependencies
<whyrusleeping>
they just exist and we read them
<ion>
whyrusleeping: That doesn’t directly lead to reproducing pristine upstream tar.gz files, though, does it?
<whyrusleeping>
ion: no, not necessarily
<whyrusleeping>
at least not the zip part
<ion>
That is what I was talking about.
grahamperrin has left #ipfs ["Leaving"]
JohnClare has quit [Remote host closed the connection]
JohnClare has joined #ipfs
JohnClare has quit [Remote host closed the connection]
JohnClare has joined #ipfs
JohnClare has quit [Remote host closed the connection]
<Gaboose>
whyrusleeping: it works fine for me!
<whyrusleeping>
putting them in the global namespace does?
<Gaboose>
cd $GOPATH/src; git clone git@github.com:jbenet/go-multiaddr.git
<Gaboose>
and then import ma "go-multiaddr"
<Gaboose>
which package did you try?
<whyrusleeping>
one with a canonical import path
<whyrusleeping>
like the crypto/sha3 one
<Gaboose>
go-multiaddr's canonical path is github.com/jbenet/go-multiaddr, isn't it?
<Gaboose>
or am i misunderstanding something
jhulten has joined #ipfs
JohnClare has joined #ipfs
<Gaboose>
you want import "golang.com/x/crypto/sha3" to work, when you only have $GOPATH/src/QmWhatevs/sha3?
jhulten has quit [Ping timeout: 250 seconds]
JohnClare has quit [Remote host closed the connection]
JohnClare has joined #ipfs
<Gaboose>
sorry, i should have looked up canonical paths...
<Gaboose>
it's when they have a line like this: package main // import "golang.org/x/tools/cmd/gorename"
JohnClare has quit [Remote host closed the connection]
merlijn_ has quit [Ping timeout: 240 seconds]
<whyrusleeping>
yeap
<whyrusleeping>
its pretty annoying
<victorbjelkholm>
woho! Managed to mirror ALL the nodejs binaries and added them to ipfs now
<whyrusleeping>
o.o
<whyrusleeping>
whoa
<victorbjelkholm>
5 seconds to install a new node version, woho!
<whyrusleeping>
hahaha, awesome!
<victorbjelkholm>
got it working with nvm, should be simple to get it working with n as well
<victorbjelkholm>
since they are both just using nodejs.org/dis
<victorbjelkholm>
t
JohnClare has joined #ipfs
JohnClare has quit [Remote host closed the connection]
JohnClare has joined #ipfs
merlijn_ has joined #ipfs
JohnClare has quit [Remote host closed the connection]
<ion>
gore name
jhulten has joined #ipfs
JohnClare has joined #ipfs
simpbrain has joined #ipfs
JohnClare has quit [Ping timeout: 260 seconds]
simonv3 has joined #ipfs
Whispery is now known as TheWhisper
reit has quit [Ping timeout: 265 seconds]
mildred has joined #ipfs
fazo has joined #ipfs
fazo has quit [Changing host]
fazo has joined #ipfs
fazo has joined #ipfs
border0464 has quit [Quit: sinked]
<lgierth>
daviddias: yeah castor and pluto are running dev0.4.0
stilt has joined #ipfs
Pterodactylus has quit [Ping timeout: 252 seconds]
merlijn_ has quit [Ping timeout: 264 seconds]
cemerick has quit [Ping timeout: 276 seconds]
<fazo>
I finally wrote my experimental discussion board hosting web app thingy built on IPFS
<fazo>
I need a volunteer or two to try it out
<fazo>
you just need to have a go-ipfs node running
<whyrusleeping>
we're using it in a few different projects
<whyrusleeping>
and a bootstrap node is running it to
<fazo>
very nice! Thanks for taking the time to answer. I have another question: are deterministic node identities planned? I mean will I be able at some time to start a node with an ID generated by a password?
elima_ has quit [Ping timeout: 260 seconds]
<whyrusleeping>
fazo: eh... you could probably do that manually if you like
<whyrusleeping>
but its like bitcoin brain wallets, theyre horribly insecure
<fazo>
yeah, that's true
<fazo>
but for building applications that require authentication such as mine, it could be useful. Of course insecure, but at least convenient for the average joe end user
<dignifiedquire>
daviddias: looks like I can’t use threads :( the code is structured so that it’s executing a function passed through a config object so I can’t distribute that easily :(
domanic has quit [Ping timeout: 250 seconds]
<daviddias>
oh :( I see. But let's not give up on that idea
<daviddias>
we can certainly make it a lot faster
<dignifiedquire>
for sure
merlijn_ has joined #ipfs
jhulten has joined #ipfs
nonmoose has joined #ipfs
merlijn_ has quit [Ping timeout: 276 seconds]
jhulten has quit [Ping timeout: 276 seconds]
<whyrusleeping>
Gaboose: godeps gets around that canonical imports problem by just removing the canonical import declaration... lol
nonmoose has quit [Ping timeout: 265 seconds]
domanic has joined #ipfs
NeoTeo has quit [Quit: ZZZzzz…]
atrapado has quit [Quit: Leaving]
rombou has quit [Read error: No route to host]
rombou has joined #ipfs
mungojelly has joined #ipfs
rombou has quit [Ping timeout: 252 seconds]
rombou has joined #ipfs
rombou has quit [Ping timeout: 260 seconds]
NeoTeo has joined #ipfs
david__ has joined #ipfs
konrad__ has quit [Ping timeout: 240 seconds]
NeoTeo has quit [Client Quit]
<david__>
daemon doesn't gives any error but stops working on vps, can anyone help
<dignifiedquire>
daviddias: :( not seeing a way to parallelize properly without rewriting registry-static, just too many assumptions about serial execution in there -.-
<dignifiedquire>
daviddias: but hey, at least it’s a decent download spped now, I’m at 100k packages now :)
<daviddias>
other than the time required to accomplish that, would there any reason why not too?
<dignifiedquire>
you mean rewriting registry-static?
<daviddias>
yep
<dignifiedquire>
if I had the time I would rewrite it immediately, but it will take some time especially as the functionality is stretched over a couple of modules, but it could be a lot faster when properly paralellized as the inherent problem of fetching all modules and storing them in flat fs, is fully parallelizable
<dignifiedquire>
but not sure it’s acutally a good idea to do this in nodejs as the existing threads implementations are pretty limited :(
Encrypt has joined #ipfs
<dignifiedquire>
and I don’t want to use anything around cluster because spawning mutliple node processes would be just a waste of resources