<pinbot>
now pinning /ipfs/Qme8HUxpr6QMcDzd3U1SzgUqtnoFQDdxZUZC69VMpi4LaL
Matoro has quit [Ping timeout: 260 seconds]
JohnClare has joined #ipfs
trochosphere has quit [Ping timeout: 272 seconds]
pinbot has quit [Ping timeout: 252 seconds]
<cryptix>
somehow pinbot doesnt like my hashes :x
<demize>
hah
JohnClare has quit [Ping timeout: 240 seconds]
Matoro has joined #ipfs
water_resistant has quit [Quit: Leaving]
fingertoe has quit [Ping timeout: 246 seconds]
<victorbjelkholm>
The gateway is ok currently? Trying to load QmS5BHZknKbjFqqJarfyCz6JdWiCrQpAAG1nmQKRZnMkJY but nothing happens... Works on other machines I've tested on
<victorbjelkholm>
huh, seems like something is wrong with how firefox connects to the gateway or how the gateway handles firefox. In Chrome is working
__mek__ has quit [Ping timeout: 240 seconds]
cSmith_ has quit [Ping timeout: 244 seconds]
__mek__ has joined #ipfs
hellertime has joined #ipfs
vanila has quit [Quit: Leaving]
ygrek has joined #ipfs
f[x] has quit [Ping timeout: 265 seconds]
__mek__ has quit [Ping timeout: 260 seconds]
cSmith_ has joined #ipfs
jfred__ is now known as jfred
<ipfsbot>
[js-ipfs-api] harlantwood created travis-config (+1 new commit): http://git.io/v8ZSg
<ipfsbot>
js-ipfs-api/travis-config 4f56715 Harlan T Wood: test under multiple node versions
<davidar>
does the gateway do gzip compression?
<multivac>
davidar: 2015-11-06 - 23:18:34 <jbenet> tell davidar to talk to __mek__ at some point
<davidar>
__mek__: hi :)
pinbot has joined #ipfs
<davidar>
or, at least, set an appropriate content-encoding header like s3
guest234234 has joined #ipfs
<davidar>
(for already gzip'd files)
<davidar>
s/does/can
<ipfsbot>
[js-ipfs-api] harlantwood opened pull request #108: test under multiple node versions (master...travis-config) http://git.io/v8ZHU
Guest23423 has joined #ipfs
nonofficially has joined #ipfs
guest234234 has quit [Ping timeout: 265 seconds]
ygrek_ has joined #ipfs
ygrek has quit [Ping timeout: 260 seconds]
Guest23423 has quit [Read error: Connection reset by peer]
border0464 has joined #ipfs
r04r is now known as zz_r04r
JohnClare has joined #ipfs
JohnClare has quit [Ping timeout: 250 seconds]
<fleeky_>
wow the new imgur sucks
<fleeky_>
i need to get the ipfs imgur installed soon
<ipfsbot>
[js-ipfs-api] harlantwood deleted circleci at db5d731: http://git.io/v8ndW
dignifiedquire_ has quit [Quit: dignifiedquire_]
compleatang has joined #ipfs
compleatang has quit [Client Quit]
anticore has joined #ipfs
e-lima has quit [Ping timeout: 240 seconds]
dignifiedquire has quit [Quit: dignifiedquire]
dignifiedquire has joined #ipfs
domanic has quit [Ping timeout: 246 seconds]
amstocker has joined #ipfs
rendar has joined #ipfs
<M-davidar>
Stskeeps (IRC): multicast in what way?
amstocker has quit [Ping timeout: 246 seconds]
cemerick has joined #ipfs
<ion>
Stskeeps: It would be cool if go-ipfs tried to use multicast for pubsub and used P2P as a fallback (since it will unfortunately fail on most of the public Internet).
<ion>
Everyone managing receive the multicast would of course also participate in the P2P distribution.
<ion>
to
Encrypt has joined #ipfs
dignifiedquire has quit [Quit: dignifiedquire]
NightRa has joined #ipfs
NeoTeo has joined #ipfs
dignifiedquire has joined #ipfs
dignifiedquire has quit [Client Quit]
slothbag has joined #ipfs
<ipfsbot>
[go-ipfs] rht opened pull request #1949: Wire offline routing for ipns in core.Resolve (master...wire-offline-routing) http://git.io/v8cks
zuchto-mob has left #ipfs [#ipfs]
__mek__ has quit [Ping timeout: 255 seconds]
jamie_k has joined #ipfs
guest234234 has quit [Ping timeout: 240 seconds]
RX14 has quit [Quit: Fuck this shit, I'm out!]
RX14 has joined #ipfs
dignifiedquire has joined #ipfs
<Vyl>
Multicast could be a very useful thing on public networks - wifi hotspots and such. But right now, IPFS is so niche that the chances of you happening to connect to the same hotspot as another IPFS node used are basically zero.
<Vyl>
We're still in the chicken-and-egg stage.
<Vyl>
No site is going to utilise IPFS because practically no-one would see any benefit, and they'd be dependant on the gateway staying up. But no-one is going to install IPFS because there's no point to a protocol that only enthusiasts use.
<victorbjelkholm>
Godmorgon!
<Vyl>
This is unlikely to change without a 'killer app' of some sort that is IPFS-exclusive. But I can't think of anything that would fit that criteria. Not anything legal, anyway.
<Vyl>
And if our killer site is 'getfreemp3z' then that creates a whole new set of problems.
<Vyl>
But if IPFS does somehow gain popularity, then I can certainly imagine multicast node location (IPv6 link-local is ideal for this) becoming a powerful feature.
<Stskeeps>
M-davidar: well, one usecase i could see it would be content distribution on let's say, in flight entertainment, as an example
<Vyl>
I see it more as a performance boost for popular content. When your OS needs to fetch some updates or you want to watch the latest viral video that everyone has seen, there's a decent chance that one of the people on that LAN segment will have it. So why wait for it to slowly download from a node at the other end of the congested internet link?
<davidar>
Stskeeps (IRC): ipfs already peers with nodes over LAN automatically. I guess multicast could be supported once bitswap gets more intelligent
<Vyl>
Oh, it already does that?
<Vyl>
Well, that's that conversation over!
<davidar>
yeah
<davidar>
lol
<Vyl>
How does it locate them?
<davidar>
I'm not sure tbh, ask whyrusleeping ;)
slothbag has quit [Quit: Leaving.]
dignifiedquire has quit [Quit: dignifiedquire]
<davidar>
actually, it might use multicast for that part already
<davidar>
just not for actually distributing content, iirc
<Stskeeps>
davidar: if it's the thing about that peers has /192.168.foo in their addresses in the public peer list, it can probably be done more intelligently, as i'm quite sure my local LAN doesn't have peers like that .. :) something like mdns to look would probably make more sense
ianopolous has quit [Read error: Connection reset by peer]
ianopolous has joined #ipfs
anticore has quit [Remote host closed the connection]
* ion
runs make test_all_commits and twiddles thumbs
guest234234 has quit [Ping timeout: 250 seconds]
Guest23423 has joined #ipfs
pfraze has joined #ipfs
sunnymilk has joined #ipfs
rombou has quit [Ping timeout: 272 seconds]
HostFat has joined #ipfs
Guest23423 has quit [Ping timeout: 260 seconds]
dignifiedquire has joined #ipfs
rombou has joined #ipfs
arpu has joined #ipfs
rombou has quit [Ping timeout: 255 seconds]
jamie_k_ has quit [Quit: jamie_k_]
jamie_k has joined #ipfs
dignifiedquire has quit [Quit: dignifiedquire]
jamie_k___ has joined #ipfs
jamie_k has quit [Ping timeout: 250 seconds]
jamie_k___ is now known as jamie_k
vanila has joined #ipfs
dignifiedquire has joined #ipfs
<vanila>
so what's the procedure if an illegal image was hosted on ipfs?
<victorbjelkholm>
vanila, hosted on ipfs? Can only be hosted by nodes that want to host that image
loki has joined #ipfs
loki is now known as Guest63605
<Guest63605>
is it necessary to open a port in firewall to run a ipfs node ?
nessence has quit [Remote host closed the connection]
nessence has joined #ipfs
pfraze_ has joined #ipfs
<victorbjelkholm>
Guest63605, for it to connect to the swarm, yes, you should have port 4001 open
pfraze has quit [Ping timeout: 240 seconds]
<SoreGums>
putting together a project that uses firebase/pubnub and was going to use cdnjs/rawgit to host the actual client - then remembered ipfs - might go the ipfs route then - want to ensure longevity of the client code ;)
<demize>
cjdns would provide the network that the client communicates over, it won't actually.. "host" it.
<demize>
You can use ipfs just fine over cjdns though.
nessence has quit [Ping timeout: 272 seconds]
<SoreGums>
i'm not clear - cjdns was for the JS libs
<SoreGums>
github was where i'll post the client - it'll be published under MIT
<SoreGums>
cdnjs**
<demize>
Oh, I thought cjdnjs was a typo for cjdns, sorry.
<SoreGums>
ahh, never heard of cjdns, looks interesting ;)
<vanila>
here's an attack scenario: nasty person makes a page that appears beigning but includes within it an illegal image, lots of nice persons view this page (e.g. it was linked on reddit), now they are all hosting that illegal image and may suffer consequences
jamie_k has quit [Ping timeout: 240 seconds]
<Vyl>
That's a legal attack, not a purely technological one. If fetching is done without user knowledge though, they should have plausible deniability. In theory that can get them off the hook legally except for strict liability offenses, and even most of those. Though in practice law enforcement can be rather overzealous in certain crimes, and will quite happily destroy a suspect's life before concluding
<Vyl>
they are not guilty after all.
nessence has joined #ipfs
jamie_k_ has joined #ipfs
dignifiedquire has quit [Quit: dignifiedquire]
RX14 has left #ipfs ["Leaving"]
cemerick has joined #ipfs
pfraze_ has quit [Remote host closed the connection]
<vanila>
i think this could be a problem for the safety of IPFS users
chriscool has quit [Ping timeout: 252 seconds]
_jkh_ has joined #ipfs
<victorbjelkholm>
vanila, just because you view the file doesn't mean you'll store and share it
<ion>
In the current implementations it does.
<victorbjelkholm>
ion, oh, I was sure you'd have to pin content for your node to start sharing it
Guest63605 has quit [Quit: Page closed]
<ion>
Anything you load is cached. Pinning just prevents garbage collection from deleting the data.
ianopolous has quit [Ping timeout: 244 seconds]
<Stskeeps>
could be nice to have a flag to not 'seed' when done downloading, though
<codehero>
Stskeeps: i think automatic seeding is a major advantage, because popular files will have faster retrieval times
<Stskeeps>
it does
<Stskeeps>
i mean, seeding -after- downloading it all
<Stskeeps>
so there's still seeding while downloading as an advantage
<codehero>
hrm
pfraze has joined #ipfs
<codehero>
well, i'm all for more options, but if everyone turns off seeding after download
<codehero>
then downloads will be slower
<Stskeeps>
nod
<Stskeeps>
there's always a weighing i guess
nessence has quit [Remote host closed the connection]
nessence has joined #ipfs
locusf has joined #ipfs
nessence has quit [Ping timeout: 240 seconds]
NeoTeo has joined #ipfs
munkey has joined #ipfs
<ipfsbot>
[js-ipfs-api] diasdavid closed pull request #108: test under multiple node versions (master...travis-config) http://git.io/v8ZHU
nessence has joined #ipfs
dignifiedquire has joined #ipfs
jamie_k has joined #ipfs
<ipfsbot>
[js-ipfs-api] diasdavid deleted travis at 2a9cc63: http://git.io/v8WeE
bussiere_ has joined #ipfs
<bussiere_>
hi
<bussiere_>
please can anyone remind me the command to get a folder and write it in the current system please
<ion>
ipfs add -r directory
<ion>
ipfs get /ipfs/hash
<bussiere_>
ok thanks it writes it and appears as a folder after in y filesysteme ?
<ion>
yes
<ion>
named after the hash
<ion>
ipfs get /ipfs/hash/subdirectory will create “subdirectory”
anticore has joined #ipfs
<bussiere_>
ion thanks
voxelot has joined #ipfs
<rendar>
codehero, what you mean with 'seeding'?
<ion>
Serving the objects
<rendar>
ion, hmm, objects = chunks of a big file, split in littler objects?
<ion>
Any IPFS objects
<rendar>
so, seeding improve downloading because you can demand IPFS objects from several different servers
<ion>
Yes, potentially even from within your local network.
<rendar>
right
<rendar>
but why you have only 1 server? it will decay to a normal download i guess
<rendar>
s/why/when/
<ion>
Someone closer to you than the original server can still have it.
<ion>
Perhaps a fast cache box on your ISP’s premises.
<ion>
Or perhaps the original server is just slow compared to everyone else.
<rendar>
hmm
domanic has joined #ipfs
<rendar>
ion, in the hypothetical case only 1 server has those objects, it will be a normal file download, like everyone else, right?
sharky has quit [Quit: Leaving]
<ion>
A big file will consist of many chunks. As soon as one node has downloaded one of them, that one is already being served by two nodes in total and soon even more.
<anticore>
hi guys. is it possible to send an ipfs hash to someone without ipfs as a link?
<multivac>
anticore: 2015-10-25 - 02:14:43 <whyrusleeping> tell anticore use 0.0.0.0 isntead of 127.0.0.1 as the ip
<ion>
anticore: What do you mean by that?
cemerick has joined #ipfs
<dignifiedquire>
daviddias: did you find anything more?
atgnag has quit [Ping timeout: 264 seconds]
dPow has quit [Quit: No Ping reply in 180 seconds.]
<anticore>
ion: a proxy i guess
<anticore>
like ipfsproxy.com/<hash> would retrieve the object