<sonatagreen>
gosh, if only we had some sort of outage-resistant alternative website technology
<zignig>
noffle: needs to be in ipfs , harder to kill.
<zignig>
sonatagreen: indeed :)
akhavr has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
od1n has joined #ipfs
<noffle>
zignig: I was just thinking a few days ago about how we'd rig up an ipfs-based etherpad. fun project!
akhavr1 has joined #ipfs
jimbeam has quit [Ping timeout: 240 seconds]
<zignig>
it would have to be an ipns block chain thing.
akhavr has quit [Ping timeout: 255 seconds]
akhavr1 is now known as akhavr
<zignig>
noffle: do you code golang ?
<sonatagreen>
idea part one: a public/shared ipns channel can be achieved by publishing/sharing the private key
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<sonatagreen>
idea part two: to control spam, a channel should be able to mark itself as requiring proof-of-work for updates
<sonatagreen>
(also, we /really/ need append-only updatable sites)
<sonatagreen>
(or at least "a valid update must include a pointer to the previous version")
akhavr1 has joined #ipfs
od1n has quit [Ping timeout: 272 seconds]
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 268 seconds]
akhavr1 has joined #ipfs
od1n has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
akhavr1 is now known as akhavr
jfis has quit [Quit: s]
jfis has joined #ipfs
akhavr1 has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr has quit [Ping timeout: 255 seconds]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 240 seconds]
npatrick04 has quit [Ping timeout: 250 seconds]
<sonatagreen>
is putting an ipns address in the dns record less resilient than putting an ipfs address and remembering to update it manually every time?
<sonatagreen>
because it's a lot less work and it would be nice if it worked properly
<sonatagreen>
i really think ipfs needs to not try to solve all the problems here
<sonatagreen>
unix philosophy, small tools doing narrow jobs excellently
<sonatagreen>
not large suites doing all jobs mediocrely
<sonatagreen>
ipfs should focus on integrating smoothly with other systems that are designed from the ground up to address these kinds of issues
<davidar>
sonatagreen (IRC): yeah, but 1M nodes all responding to your query is just a ddos :)
dysbulic_ has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<davidar>
sonatagreen (IRC): also, the search engine would just be an app on top of ipfs
<davidar>
I don't think it should be built into ipfs itself, if that's your concern
<sonatagreen>
i feel like most interesting apps have as a prerequisite a certain specific tool that i'm not sure i can quite grasp clearly
<davidar>
However, I do think ipfs should support the general idea of aggregation, since it's necessary for building lots of interesting dynamic apps
<sonatagreen>
but something vaguely in the neighborhood of sifting good content from bad, discouraging spam, enabling strangers to discover and talk to each other
<davidar>
sonatagreen (IRC): yes, aggregation! ;)
<sonatagreen>
it's a common problem whose solution is needed by chat, forums, search engines, etc.
<sonatagreen>
wikis
amstocker_ has joined #ipfs
<davidar>
I think spam can be handled by blacklisting, bitswap credit, etc
<davidar>
sonatagreen (IRC): exactly
<sonatagreen>
and i feel like there needs to be a dedicated tool designed specifically and purely for this
goregrind has quit [Ping timeout: 265 seconds]
dysbulic_ has quit [Ping timeout: 240 seconds]
<sonatagreen>
and i'm not convinced that ipfs is the right place for that tool, at least at first
jimbeam_ has joined #ipfs
<sonatagreen>
there's a lot of cool projects out there, competing with each other. tor, i2p, gnunet, freenet, bitmessage, retroshare, ipfs.
<sonatagreen>
i want to aggressively avoid duplicating things.
<davidar>
sonatagreen (IRC): ipfs at least needs to be able to support such a tool though
<sonatagreen>
i want to not build something that already exists.
<sonatagreen>
absolutely, interacting with the thing is important
goregrind has joined #ipfs
<sonatagreen>
things need to fit together smoothly
<sonatagreen>
but don't build something until you're /sure/ it doesn't already exist
<davidar>
that's a little bit naive though
jimbeam has quit [Ping timeout: 252 seconds]
<davidar>
Yes, don't reinvent the wheel
<davidar>
But you need diversity of ideas, which unfortunately includes duplicated labor
<noffle>
davidar: yes! I've been following this excitedly. (CRDTs are super cool)
<noffle>
zignig: as of two weeks ago, yes :)
<davidar>
Good technology is born from the ashes of dozens of similar projects that were ultimately dead ends
<davidar>
noffle (IRC): awesome :)
jimbeam has joined #ipfs
<davidar>
Now somebody who knows crdts needs to actually build it ;)
* davidar
looks at jbenet
<noffle>
yes, we need to build up some solid primitives to enable these sorts of applications
<noffle>
it's just figuring out how it all fits into ipfs
akhavr1 has joined #ipfs
<bitemyapp>
davidar: there's an old implementation of CRDTs in Haskell. I could ping the author.
<davidar>
ping sargun
akhavr has quit [Ping timeout: 265 seconds]
akhavr1 is now known as akhavr
jimbeam_ has quit [Ping timeout: 246 seconds]
<davidar>
bitemyapp (IRC): that would be cool. Once we get ipld and crdts (and I eventually finish hs-ipfs-api...), I'd really like to start building some of this stuff in Haskell
<zignig>
davidar: is haskell your language of choice ?
<bitemyapp>
davidar: IPLD?
akhavr1 has joined #ipfs
<davidar>
bitemyapp (IRC): for storing arbitrary data structures in ipfs (transparently)
<multivac>
KeyError: u'extract' (file "/usr/local/lib/python2.7/dist-packages/sopel/modules/wikipedia.py", line 91, in mw_snippet)
<amstocker_>
uh oh
<amstocker_>
anyways, I think they also want to do in the typical 'multi-fashion' so that objects cant be equally represented in yaml, xml, json, etc
<amstocker_>
and also you would be able to seamlessly link very large objects that need to be composed of multiple ipfs objects under the hood
<amstocker_>
re sonatagreen
akhavr has quit [Ping timeout: 260 seconds]
akhavr1 is now known as akhavr
<davidar>
Sargun (IRC): ipld = dump a big json file into ipfs, and have it work nicely without you having to do all the manual link-walking
<davidar>
amstocker (IRC): jbenet tells me jsonld is no longer part of the plan (but will be supported on top of ipld)
<davidar>
jbenet: this ^ is exactly why I think the name IPLD is confusing :)
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<amstocker_>
so the schema thing is no longer a part of it?
<amstocker_>
davidar: its hard to keep track of all the action on github
jimbeam has quit [Ping timeout: 250 seconds]
<davidar>
amstocker (IRC): I know... and some of the info isn't even on github yet (just irc) :/
akhavr1 has joined #ipfs
<amstocker_>
I thought the data schema thing would be interesting. For example instead of the regular DAG objects having the two fields "Data" and "Links", you could define a schema that extends that to add metadata, etc
<amstocker_>
only having*
<amstocker_>
davidar: there needs to be like an IRC scribe bot for ipfs
<amstocker_>
it would be nice if there was a resource for critical conversations
<amstocker_>
you would say, "Scribe! Record the last 30 minutes of chat"
<amstocker_>
and then tag it with a subject line
<davidar>
i think multivac has a function like that, will have to check
* davidar
bbiab
<amstocker_>
or just someone who's job it is to read IRC and create a digest :P
rschulman has quit [Quit: Connection closed for inactivity]
dysbulic has quit [Ping timeout: 240 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
akhavr1 is now known as akhavr
akhavr has quit [Ping timeout: 256 seconds]
akhavr has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr1 has joined #ipfs
Soft has quit [Read error: Connection reset by peer]
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
dysbulic has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 244 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
akhavr has joined #ipfs
dysbulic has quit [Ping timeout: 246 seconds]
akhavr1 has quit [Ping timeout: 272 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Soft has joined #ipfs
Stard0g101 has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
akhavr1 is now known as akhavr
captain_morgan has joined #ipfs
akhavr1 has joined #ipfs
<davidar>
amstocker_ (IRC): are you volunteering? :p
<amstocker_>
hmmm
<amstocker_>
davidar: is there money in it? :)
akhavr has quit [Ping timeout: 246 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
<davidar>
amstocker_ (IRC): depends on your definition of money...
akhavr has quit [Ping timeout: 244 seconds]
<davidar>
.ask sargun to talk to jbenet about getting crdts and co into ipfs
<multivac>
davidar: I'll pass that on when sargun is around.
chriscool has joined #ipfs
akhavr1 has quit [Ping timeout: 272 seconds]
akhavr has joined #ipfs
sbruce has quit [Read error: Connection reset by peer]
sbruce has joined #ipfs
<Sargun>
davidar: Back.
<multivac>
Sargun: 2015-10-07 - 05:56:23 <davidar> ask Sargun to talk to jbenet about getting crdts and co into ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<Sargun>
uh
<Sargun>
davidar: Via what path
<Sargun>
or medium
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 256 seconds]
akhavr1 is now known as akhavr
akhavr has quit [Ping timeout: 246 seconds]
akhavr has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
akhavr has joined #ipfs
akhavr has quit [Client Quit]
akhavr1 has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
akhavr1 is now known as akhavr
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed real-trailers from c01270d to c59631b: http://git.io/vZ2dl
<ipfsbot>
go-ipfs/real-trailers c59631b Jeromy: use go1.5 for circleCI (snippet from circle support)...
<ipfsbot>
go-ipfs/real-trailers 13998b7 Jeromy: update go version requirement in readme...
akhavr1 has joined #ipfs
Spinnaker has quit [Quit: sinked]
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
rendar has joined #ipfs
ygrek_ has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
mildred has joined #ipfs
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 265 seconds]
chriscool has quit [Ping timeout: 250 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 264 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr1 is now known as akhavr
akhavr has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
<gendale_>
is there a function in ipfs to return peers that have a certain hash?
akhavr has joined #ipfs
dignifiedquire has joined #ipfs
<sonatagreen>
there shouldn't be, because privacy
captain_morgan has quit [Ping timeout: 240 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
<whyrusleeping>
gendale_: there is!
<whyrusleeping>
thats the primary lookup mechanism ipfs uses to locate content
<whyrusleeping>
through the api it is 'ipfs dht findprovs <hash>'
<whyrusleeping>
in go code its routing.FindProviders()
akhavr has joined #ipfs
<gendale_>
that's what i thought
<gendale_>
are the hashes that get returned hashes that contain the hash given?
notduncansmith has joined #ipfs
akhavr1 has quit [Ping timeout: 240 seconds]
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping>
the hashes returned are the peer ID's of nodes who say they have that block
<gendale_>
ah, of course
<gendale_>
thanks!
<gendale_>
does it run indefinately?
<whyrusleeping>
no, it runs until it finds K (20) peers
<whyrusleeping>
which is kindof awkward right now because 20 is generally way more than the number of peers who have something
<gendale_>
gotcha
<gendale_>
so when i run ipfs id on those peers, it doesn't really return anything
<whyrusleeping>
what do you mean?
<gendale_>
"PublicKey": "",
<gendale_>
"Addresses": null,
<gendale_>
"AgentVersion": "",
<gendale_>
"ProtocolVersion": ""
<whyrusleeping>
huh
<whyrusleeping>
that feels like a bug
<gendale_>
that's what i assumed
<whyrusleeping>
yeah, generally if you think its a bug, its a bug. even if i dont think its a bug.
<gendale_>
theoretically, though, node info is similar to bittorrent, correct? e.g. by pinning or hosting a hash i'm exposing my ip to the swarm?
<gendale_>
lol
<whyrusleeping>
basically yeah
Guest73396 has joined #ipfs
<whyrusleeping>
thats a consequence of using the DHT as a routing system
<whyrusleeping>
we want to ship different routing systems (wayyy down the road) that avoid that
<whyrusleeping>
tor routing for example
<gendale_>
interesting
<gendale_>
cjdns would obfuscate that quite a bit too, right?
<whyrusleeping>
yeap, that would help too
<whyrusleeping>
and we're already working towards both of those ends
<gendale_>
because you can only be identified by first hop peers
<gendale_>
oh awesome
<gendale_>
i didn't know tor routing was something you were working on
<whyrusleeping>
we arent putting a ton of effort into it, but one contributor is building a tor dialer and listener
<whyrusleeping>
and once that works, it should be fairly easy to plug it into the rest of ipfs
<whyrusleeping>
and thats a very good start to a full tor integration
<whyrusleeping>
and then others are working on different levels of cjdns integration, which is something i dont know a lot about
<whyrusleeping>
unfortunately
<gendale_>
i was thrilled to see lgierth's gateways running cjdns
<gendale_>
it's an awesome protocol
<whyrusleeping>
yeap!
<whyrusleeping>
those are the gateways that you hit when you access ipfs.io
sonatagreen has quit [Ping timeout: 268 seconds]
<gendale_>
yep! they're really well organized
<gendale_>
i've been basically trying to emulate his workflow for the ipfs.alexandria.media gateways
akhavr1 has joined #ipfs
<gendale_>
he's a better devops than i am though. :)
akhavr has quit [Ping timeout: 264 seconds]
akhavr1 is now known as akhavr
<whyrusleeping>
haha, hes very meticulous and knowledgable in networking
<whyrusleeping>
very great person to have around IMO
<whyrusleeping>
i'm excited to get this utp stuff out the door and drop 0.3.8
<whyrusleeping>
once that lands, we have a short ramp to merging 0.4.0 (pending feature creep)
<whyrusleeping>
and then my primary focus is going to be transfer perf
<gendale_>
is there a public roadmap on github?
<whyrusleeping>
i've been sketching one out today
<whyrusleeping>
but other than that, not much
<whyrusleeping>
we're trying to ship 0.3.8 on thursday
<gendale_>
is 0.3.8 primarily utp then? or a bunch of other stuff too?
<gendale_>
what should i be looking out for?
<whyrusleeping>
which will fix the random api failures and also the adds failing due to too many file descriptors
<gendale_>
oh awesome
<whyrusleeping>
utp is slated for 0.3.8, but its not a hard requirement at the moment
<davidar>
Sargun (IRC): whenever he's on irc I guess
<whyrusleeping>
i might push for shipping 0.3.8, but not defaulting to using it
<whyrusleeping>
er, shipping utp
<whyrusleeping>
and then having people enable it and test it live for a week or so
<whyrusleeping>
then once that proves stabler, shipping utp in the default configs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<gendale_>
very cool
<gendale_>
i'm consistently blown away by how fast ipfs transfers are
<whyrusleeping>
yeah, theyre pretty quick right now. but they waste a good deal of bandwidth
<whyrusleeping>
which is my first target when improving perf
<gendale_>
right
<gendale_>
is the bootstrapping a big part of that? speed i mean not lots of bandwidth
devbug has quit [Ping timeout: 268 seconds]
<gendale_>
it seems like both peer discovery and transfer speeds are significantly and reliably faster than torrent
<davidar>
gendale_ (IRC): been meaning to look at Alexandria, would it be relevant to github.com/ipfs/archives?
<whyrusleeping>
yeah, i'm actually not sure about speed comparisons there
<whyrusleeping>
it could just be that mainline dht is huge and can get overloaded
<whyrusleeping>
it could also be (less likely) that our dht implementation kicks ass and just happens to be freaky fast
<gendale_>
davidar: yes absolutely
<gendale_>
i showed devon that github the other day and he was thrilled to see it
akhavr1 has joined #ipfs
<gendale_>
we're in the middle of a big push for a release in the next couple days
<gendale_>
but he was really excited about your work and really wanted to touch base
akhavr has quit [Ping timeout: 264 seconds]
<davidar>
gendale_ (IRC): awesome, happy to talk about any ways we can collaborate :)
<gendale_>
whyrusleeping: haha, gotcha, i'd have believed the latter if you told me. :P
<gendale_>
davidar: stop by the slack chat sometime
<gendale_>
a big goal of this next release is a daemon manager, which starts/stops ipfs and florincoind and libraryd (the alexandria api) from the systray
akhavr has joined #ipfs
<gendale_>
the practical purpose of which is that people can publish things to alexandria without manually managing a bunch of daemons
<gendale_>
but there's some really neat front end stuff happening too
akhavr1 has quit [Ping timeout: 264 seconds]
<gendale_>
and the overall project goal is to build a library of alexandria that can't burn down. :) so yeah archival. :P
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
<whyrusleeping>
gendale_: you should check out the changes being made to our app
<davidar>
gendale_ (IRC): sounds great, will do :)
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
* davidar
wonders if it's possible to bridge to slack from matrix (Matthew/Erik?)
akhavr1 has quit [Ping timeout: 250 seconds]
akhavr has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
akhavr has joined #ipfs
elavoie has quit [Quit: Connection closed for inactivity]
akhavr1 has quit [Ping timeout: 260 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 244 seconds]
distro has joined #ipfs
<distro>
Hi there guys. Quick question. When I add a file in IPFS and then I shutdown the computer after let's say one minute, can others still get the file?
<whyrusleeping>
distro: only if someone else grabbed it from you before you shut down your computer
akhavr has joined #ipfs
<distro>
So it's like BitTorrent..
<whyrusleeping>
yeap.
<whyrusleeping>
very similar distribution mechanisms
<distro>
So it's just a wrapper around a BitTorrent fork? Because one could create the same thing with BitTorrent, right?
<distro>
(If you exclude IPNS)
<whyrusleeping>
haha, i encourage you to read the whitepaper
akhavr1 has quit [Ping timeout: 240 seconds]
<whyrusleeping>
its nearly 1am here and i'm really sleepy
<distro>
Sure, thanks
distro has quit [Client Quit]
<whyrusleeping>
no probs! (hopefully someone else is around if you have more questions)
NeoTeo has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<shyamsk>
hey is there something like IRC/Slack built on Slack? or does anyone plan to build something like that?
<shyamsk>
bah... built on IPFS
akhavr has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
<locusf>
yo dawg ..
<davidar>
shyamsk (IRC): chat.ipfs.io
akhavr1 has joined #ipfs
<davidar>
Not fully on ipfs though (yet ;)
akhavr has quit [Ping timeout: 246 seconds]
akhavr1 is now known as akhavr
<davidar>
distro (IRC): cf project maelstrom
<davidar>
IPFS is better though ;)
<davidar>
locusf (IRC): we were just talking about directed acyclic word graphs the other day :p
<davidar>
.w DAWG
<multivac>
[WIKIPEDIA] Dawg | "Dawg or DAWG may refer to:Directed acyclic word graph, a computer science data structureDirected acyclic word graph, an alternative name for the Deterministic acyclic finite state automaton computer science data structureDawg, the nickname of American mandolinist David GrismanDawg, a fictional dog..." | https://en.wikipedia.org/wiki/Dawg
<_p4bl0>
hello
<_p4bl0>
i think i see a security problem in IPFS API because of CORS
<davidar>
_h3ll0 _p4bl0
voxelot has joined #ipfs
<_p4bl0>
davidar: ^^
<_p4bl0>
is it possible to access-control-allow-origin not of a domain but of a subsirectory of a domain?
<_p4bl0>
like "127.0.0.1:8080/ipns/something"
<_p4bl0>
becausse I can't get that to work
<_p4bl0>
and that means that if I allow 127.0.0.1:8080 then any IPFS loaded web page can call the API using some JavaScript
<_p4bl0>
and for instance the webpage could pin somme huge file on my node
<davidar>
gendale_ (IRC): how do i actually join the slack chat? :/
<davidar>
_p4bl0 (IRC): sounds like you're talking about path-based security, it's a known issue
<_p4bl0>
so it is not solvable using CORS headers?
<davidar>
not that I'm aware of, browsers are still stuck with this idea that everything within a single domain is fine...
<davidar>
but I believe that's changing currently
<_p4bl0>
ok
akhavr1 has joined #ipfs
<_p4bl0>
davidar: thx
akhavr has quit [Ping timeout: 244 seconds]
akhavr1 is now known as akhavr
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
devbug has joined #ipfs
<_p4bl0>
davidar: oh, it seems CORS can request an Authorization header with a custom token
<_p4bl0>
that should be enough to provide security :)
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
akhavr1 is now known as akhavr
devbug has quit [Ping timeout: 264 seconds]
<davidar>
_p4bl0 (IRC): I believe there's also going to be more going into securing the api (to the extent that you'll be able to open it to the world without it being abused)
<_p4bl0>
that would be cool :)
akhavr has quit [Ping timeout: 265 seconds]
<_p4bl0>
I think a good first step would be to have a way to specify a token in the .ipfs/config file that should always be in the Authorization headers of the http api request
akhavr has joined #ipfs
<_p4bl0>
that way you could give your token only to web page that you trust
bedeho has quit [Ping timeout: 244 seconds]
<_p4bl0>
mh no actually I'm not sure this CORS feature is supposed to be used that way
<davidar>
_p4bl0 (IRC): yeah, jbenet probably knows more about that than i do
akhavr1 has joined #ipfs
<_p4bl0>
davidar: ok
<_p4bl0>
anyway, I should get back to my actual work now ^^
<_p4bl0>
bye :)
akhavr has quit [Ping timeout: 244 seconds]
akhavr1 has quit [Ping timeout: 264 seconds]
akhavr has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ygrek_ has quit [Ping timeout: 272 seconds]
jhulten_ has quit [Ping timeout: 264 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
akhavr1 is now known as akhavr
akhavr has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
<locusf>
davidar: LOL
akhavr1 has joined #ipfs
<ion>
“<gendale_> i'm consistently blown away by how fast ipfs transfers are” I'm curious, what kind of an Internet connection do you have?
<davidar>
locusf (IRC): zignig and i want to write a dawg library purely for the meme value :p
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
<davidar>
ion: yeah, on my connection there's still a lot of room for improvement :/
<davidar>
cf my periodic raging at pinbot :p
bashrc has joined #ipfs
zz_r04r is now known as r04r
voxelot has quit [Ping timeout: 272 seconds]
voxelot has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<rendar>
DHT works like this: i contact ipfs.io, asking for the file with hash:45gtr3 then the server does: server_addr = DHT_VECT[ 45gtr3 % DHT_VECT_SIZE ]; and sends me the 'server_addr' that i can finally connect to -- is this right?
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 272 seconds]
<clever>
rendar: not really
<clever>
rendar: first, all nodes have an id, which is the same length as the hash, and the network has a distance function to measure the difference between 2 things (a nodeid and a hash for example)
akhavr1 has quit [Ping timeout: 265 seconds]
<clever>
and then a node will search the network, for a node that is near the hash it wants to access
<clever>
i believe IPFS saves hash=<ownid> pairs into the DHT, for every piece of every file it has
<rendar>
clever: hmmm, ok
<clever>
there is also some complixity in how it finds a given node faster
<rendar>
clever: how a node id is calculated?
<clever>
for ipfs, its a hash of your public key
<rendar>
so every server has a different public key
<clever>
yeah
<rendar>
then you'll have a pair of { hash(node_public_key), distance } in the DHT
Soft has quit [Quit: WeeChat 1.4-dev]
<clever>
the distance isnt stored in the dht, its just computed as needed
<rendar>
ok
akhavr has joined #ipfs
ianopolous2 has quit [Ping timeout: 256 seconds]
<rendar>
and the DHT is replicated in *each* node everytime?
<clever>
no, the DHT is how you store a small number of key=value pairs on each node
<clever>
so the data is spread over all nodes
<clever>
no node holds the entire DHT, thats what the D stands for
<clever>
distrobuted
<rendar>
right
<rendar>
so DHT is basically just an hash table, locally
<clever>
yeah, and only with the hash can you know which node to query for the data
<rendar>
let's assume the hash of the file i'm requesting isn't present in the local node DHT, what happens then?
<clever>
it will go thru the peer list, and find the peer that is nearest to the file hash
<clever>
and then ask him if he has the value, or knows of another peer that is closer
<rendar>
ok, and going thru the peer list means simply enumerating a list of peers the node is connected to, in this time?
<clever>
yep
<rendar>
and when it receives the response for one peer, it updates the DHT
<rendar>
so the next request will be faster
<rendar>
s/for/from/
<multivac>
rendar meant to say: and when it receives the response from one peer, it updates the DHT
<clever>
the peer list is setup with several buckets, to keep the speed fast all the time
<rendar>
yep
notduncansmith has joined #ipfs
<clever>
the first level will have a small number of peers, that are near the current node
notduncansmith has quit [Read error: Connection reset by peer]
<clever>
the 2nd level will have more peers, at a farther distance
<clever>
and so on
<r1k0>
is there any design solution for POSTing data to an ipfs site? Or is ipfs restricted to languages that can be run client side only?
<rendar>
clever: yeah but having a query in the local DHT will be many orders of magnitute faster than querying one peer, so when the node get responses from *remote* peers, it updates the *local* DHT, right?
<clever>
rendar: so for example, the level 10 bucket has 50 peers, that are very far from the current node, but equaly spread about the network
compleatang has quit [Quit: Leaving.]
<clever>
which can act as shortcuts to get to any point in the dht
<r1k0>
where can I find a documentation list of supported events? like Bitswap.Rebroadcast.*/handleFindPeerBegin/etc...
<clever>
rendar: yeah, it likely keeps a cache of replies localy, just for speed
<rendar>
clever: isn't that cache the DHT itself?!
<clever>
rendar: the DHT is data that peers are storing for a set time, on behalf of others
<clever>
while the cache is more of something you keep localy for speed, and dont have to keep long-term
compleatang has joined #ipfs
<rendar>
clever: ok but both those strcutures (DHT and the local cache) say basically the same thing: file hash XYZ? you will find it at the node ABC -- so what is the difference?
<clever>
the difference is that the DHT only stores data for hashes near the current nodeid
<clever>
which allows others to know you will have an answer
<rendar>
clever: hmm so the DHT only stores data of the *next* nodes, and not of the next->next->...next nodes
gilgongo has joined #ipfs
<clever>
as an example, if a file had the hash 1234, it would be stored as 1234=<nodelist> on the nodes 1233 and 1235
<rendar>
ok
<clever>
and then a user at node 1 would keep connections to nodes 2,3,4 10,20,30, 400,500,600, and 1000,2000,3000 for example
<clever>
node 1 would then ask node 1000 for info, and 1000 would keep data on 998,999,1001,1002, 800,900,1100,1200 and it would relay you to 1200
nagydani has left #ipfs [#ipfs]
<rendar>
clever: hmm i see
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
akhavr1 is now known as akhavr
HostFat has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 244 seconds]
akhavr has joined #ipfs
<r1k0>
where can I find a documentation list of supported events? like Bitswap.Rebroadcast.*/handleFindPeerBegin/etc...
akhavr1 has quit [Ping timeout: 264 seconds]
<r1k0>
does it exist?
gilgongo has quit [Ping timeout: 264 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
<Sargun>
davidar: Cool
<Sargun>
.ask jbenet Talk to me about CRDTs in IPFS - IRC, or Email - sargun@sargun.me.
<multivac>
Sargun: I'll pass that on when jbenet is around.
<Sargun>
Is this connected to a slack channel or something
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
<davidar>
Sargun (IRC): why do you ask?
<davidar>
if by "or something" you mean matrix.org, then yes :)
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 272 seconds]
akhavr1 has joined #ipfs
amstocker_ has quit [Ping timeout: 244 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr has quit [Ping timeout: 244 seconds]
akhavr1 is now known as akhavr
locusf has quit [Ping timeout: 246 seconds]
j0hnsmith has joined #ipfs
<kandinski>
clever, thanks for the explanation about the DHT (and thanks rendar for being the socratic interlocutor). Is there a canonical paper that we can read?
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 264 seconds]
devbug has joined #ipfs
arrdem has quit [Remote host closed the connection]
akhavr1 has joined #ipfs
therealplato1 has joined #ipfs
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
devbug has quit [Ping timeout: 264 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
therealplato has quit [Ping timeout: 264 seconds]
akhavr has quit [Ping timeout: 240 seconds]
therealplato has joined #ipfs
therealplato1 has quit [Ping timeout: 256 seconds]
mildred has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
voxelot has quit [Ping timeout: 252 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
akhavr has quit [Ping timeout: 240 seconds]
mildred has joined #ipfs
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 260 seconds]
akhavr1 has joined #ipfs
Guest73396 has quit [Ping timeout: 260 seconds]
akhavr has quit [Ping timeout: 265 seconds]
akhavr1 is now known as akhavr
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
HostFat has quit [Read error: Connection reset by peer]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 268 seconds]
locusf has joined #ipfs
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 240 seconds]
akhavr has quit [Ping timeout: 244 seconds]
akhavr has joined #ipfs
akhavr1 has joined #ipfs
atgnag has quit [Read error: Connection reset by peer]
akhavr has quit [Ping timeout: 252 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
<locusf>
how do I put raw data to a new ipfs object?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<multivac>
[WIKIPEDIA] Convergent encryption | "Convergent encryption, also known as content hash keying, is a cryptosystem that produces identical ciphertext from identical plaintext files. This has applications in cloud computing to remove duplicate files from storage without the provider having access to the encryption keys. The combination of..."
<achin>
if you want to encrypt your data, i think you are better off doing so before you put it into IPFS
<stick`>
this would need the user to have two hashes - one hash of the encrypted file stored in ipfs, and one hash of the plaintext file (which was used for encrypting the plaintext)
mildred has quit [Ping timeout: 240 seconds]
<stick`>
achin: right, but when 2 people encrypt the same file you would end up with different ciphertext
<achin>
yes, and that might be a good thing?
<stick`>
in convergent encryption you have the same ciphertext, because the key depends on the plaintext
<achin>
to clarify one thing: if all of the blocks for a given file are stored on a node, you will be able to scan all of the blocks and recover the file (even if you don't know the hash ahead of time)
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
<stick`>
that's ... unfortunate
<achin>
but all of the blocks on yoru own node should be for files that you've fetched
zero-ghost has quit [Quit: Leaving.]
akhavr has joined #ipfs
nicolagreco has joined #ipfs
akhavr1 has quit [Ping timeout: 265 seconds]
<rschulman>
stick`: Not sure why that should be unfortunate... what's your use case for not wanting a user to be able to look at the blocks on their node?
nicolagreco_ has joined #ipfs
akhavr1 has joined #ipfs
nicolagreco has quit [Read error: Connection reset by peer]
nicolagreco_ is now known as nicolagreco
<ion>
Would you want your web browser cache to be encrypted from you?
akhavr has quit [Ping timeout: 244 seconds]
akhavr1 is now known as akhavr
mildred has quit [Ping timeout: 264 seconds]
mildred has joined #ipfs
pfraze has quit [Read error: Connection reset by peer]
pfraze has joined #ipfs
akhavr has quit [Ping timeout: 255 seconds]
akhavr1 has joined #ipfs
notduncansmith has joined #ipfs
jimbeam has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<achin>
i might want it encrypted so that others can't read it, though. in that case i might rely on something like full-disk encryption
akhavr has joined #ipfs
<ion>
And I do.
akhavr1 has quit [Ping timeout: 255 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 260 seconds]
<achin>
so that just means to get at your data, i have to attack you and your computer while it is turned on and you are at its terminal
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 240 seconds]
jimbeam has quit [Ping timeout: 260 seconds]
nicolagreco has quit [Quit: nicolagreco]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 252 seconds]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 240 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
akhavr1 is now known as akhavr
<stick`>
rschulman: gateway nodes
<M-matthew>
* davidar wonders if it's possible to bridge to slack from matrix (Matthew/Erik?)
<M-matthew>
yes, it is
<M-matthew>
but it's pretty beta atm
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
pfraze has quit [Remote host closed the connection]
<rschulman>
stick`: What about them, though?
<stick`>
i might want to use them
<stick`>
but don't want them to see my data
<ion>
But it's the beta uprising.
<achin>
stick`: so the data itself will be encrypted?
<stick`>
yes
<achin>
so then it's encrypted end-to-end?
<stick`>
read the issue i wrote :)
<ion>
stick: Get it through an anonymizing layer such as tor.
<stick`>
tor is not a silver bullet
jimbeam has joined #ipfs
<stick`>
and it's slllllloooooooo
<stick`>
wwww
hellertime has quit [Quit: Leaving.]
<achin>
i guess i'm still confused about the usecase :(
<rschulman>
stick`: Why is gpg not the solution for your problem?
<rschulman>
or are you just saying we should build gpg into ipfs?
<stick`>
have you read the issue?
<rschulman>
yep
<stick`>
gpg does notdeduplicate ciphertezt
<rschulman>
and I get what you're saying about convergent encryption
<rschulman>
right
<rschulman>
that's not a large problem for ipfs, though
<rschulman>
who would want to deduplicate?
<stick`>
dropbox
<stick`>
for example :)
<rschulman>
you're presuming that multiple different people often encrypt the exact same bitstream
<rschulman>
I'm not sure I see that happening a lot.
<ion>
>encrypt the plaintext file with AES, using H_p as the encryption key
nicolagreco has joined #ipfs
<stick`>
this will happen omce everythibg is encrypted
<ion>
Would it be useful to use e.g. scrypt(H_p) instead?
<stick`>
yes, i forgot about keystretching
<stick`>
the only thing that matters that it is determinisric
<stick`>
i wil create a prototype i guess
<achin>
so maybe we can update this issue to be clearer about the use cases
Spinnaker has joined #ipfs
<stick`>
yep
<achin>
for some use cases, the user (the producer of the data) really wants gpg for end-to-end encryption
<stick`>
gpg has lots of flaws
<achin>
but that's not what you are talking about; i think you're targeting a different use-case
<stick`>
but noone is stopping you from using that
<ansuz>
aren't you able to pipe into ipfs?
hellertime has joined #ipfs
<achin>
well, s/gpg/encryption of your choice/
<stick`>
you still can send H_p via gpg
<ansuz>
echo "pewpewpew" | gpg <flags> | ipfs add
akhavr1 has joined #ipfs
tsp has quit [Ping timeout: 272 seconds]
<achin>
yeah, that would work fine, ansuz
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
devbug has joined #ipfs
tsp has joined #ipfs
hellertime has quit [Ping timeout: 244 seconds]
akhavr has quit [Ping timeout: 240 seconds]
hellertime has joined #ipfs
akhavr has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
devbug has quit [Ping timeout: 272 seconds]
gamemanj has joined #ipfs
<achin>
stick`: i just added a comment to your issue. can you have a look to see if i got it right?
pfraze has joined #ipfs
pfraze has joined #ipfs
<achin>
(i think it's a fairly obvious re-wording of what you already said, but i just want to be clear)
r04r is now known as zz_r04r
zz_r04r is now known as r04r
jimbeam has quit [Ping timeout: 256 seconds]
nicolagreco has quit [Quit: nicolagreco]
nagydani has left #ipfs [#ipfs]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Stard0g101 has quit [Quit: Leaving.]
Stard0g101 has joined #ipfs
Stard0g101 has quit [Client Quit]
devbug has joined #ipfs
bedeho has joined #ipfs
<stick`>
achin: yep, correct
amade has joined #ipfs
jager_ has joined #ipfs
<achin>
thanks
Guest73396 has quit [Ping timeout: 244 seconds]
jager has quit [Read error: Connection reset by peer]
ZioFork has quit [Ping timeout: 265 seconds]
<achin>
you could for-sure put together a POC-demo of this. might be neat
Myagui has quit [Excess Flood]
ZioFork has joined #ipfs
Myagui has joined #ipfs
hellertime has quit [Quit: Leaving.]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
hellertime has quit [Read error: Connection reset by peer]
<giodamelio>
gamemanj: Can you link this script?
<Gisle>
What are the conditions for ipfs mount to work? I've tried changing mountpoint owner (chown), restarting ipfs, and adding my user to a group named fuse.
voxelot has joined #ipfs
voxelot has joined #ipfs
<Gisle>
time="17:12:02:000" level=error msg="error mounting: fusermount: exit status 1 fusermount: exit status 1" module="core/commands"
<Gisle>
Error: fuse failed to access mountpoint /ipfs
voxelot has quit [Client Quit]
<ion>
Run id. If it does not list fuse in your groups, you’ll need to log in again.
voxelot has joined #ipfs
devbug has quit [Remote host closed the connection]
Encrypt has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
<achin>
for my fuse mount i just had to chown achin:achin
<gamemanj>
giodamelio: It's two scripts: one to read "ipfs diag net" results, and the script called by the first to run the "resolve" in the background, with a 20-sec timeout
<Gisle>
It does show fuse.
<gamemanj>
it only does one at a time, the background-ing is needed because more than 20 seconds is somewhat impractical for a scan of the whole network
<gamemanj>
so it needs to timeout, so it needs to background, so it needs to be written in shell...
<gamemanj>
the sub-script is in the comments: /ipfs/QmUafRDofKaxJTysoWPMVy3kmwTXcvFsmYhq1ezq4B7Phx
<Gisle>
I have the permissions set to the user with ipfs, and ssh'd out and back in. Is fuse a kernel module that need a reboot?
<giodamelio>
gamemanj: Thanks
akhavr1 has joined #ipfs
ygrek_ has joined #ipfs
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
Encrypt has quit [Quit: Quitte]
akhavr1 has joined #ipfs
notduncansmith has joined #ipfs
jimbeam has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr has quit [Ping timeout: 260 seconds]
akhavr1 is now known as akhavr
<gamemanj>
giodamelio: you probably will want to pipe the output to a file, which leaves you with a simple index of IPNS addresses and IPFS hashes
bedeho has joined #ipfs
sseagull has joined #ipfs
Tv`_ has joined #ipfs
Tv`_ is now known as Tv`
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 272 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
hellertime1 has quit [Quit: Leaving.]
akhavr has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
<gamemanj>
Seems someone else has also been mapping the IPNS space: QmZXxbfUdRYi578pectWLFNFv5USQrsXdYAGeCsMJ6X8Zt /ipfs/QmSUVCsBWgLCYJWagXrQkEWFUkhcpDo9kt35VRkASkMxeP
hellertime has joined #ipfs
akhavr1 has quit [Ping timeout: 264 seconds]
jimbeam has quit [Ping timeout: 240 seconds]
akhavr1 has joined #ipfs
<ion>
>nostril/gorf
akhavr has quit [Ping timeout: 250 seconds]
akhavr1 is now known as akhavr
<gamemanj>
ion: I don't understand it either
<gamemanj>
maybe they didn't want people to look
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Spinnaker has quit [Quit: sinked]
warner has quit [Quit: ERC (IRC client for Emacs 24.5.1)]
akhavr1 has joined #ipfs
nskelsey has joined #ipfs
jimbeam has joined #ipfs
<ion>
When referring to IPNS hashes, you could say /ipns/Qm...
akhavr has quit [Ping timeout: 264 seconds]
akhavr1 is now known as akhavr
<gamemanj>
ion: I assumed it was implied by the other hash being marked with /ipfs/, but I'll note that in future
<giodamelio>
whenever I see a hash by itself, I assume it's ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 244 seconds]
akhavr1 is now known as akhavr
Spinnaker has joined #ipfs
josephdelong has joined #ipfs
jimbeam has quit [Ping timeout: 240 seconds]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
<gamemanj>
Well, here's my results: /ipfs/QmbTQPUFFEPxnRk5sxpf3k8yevqDoJucJXyxjzds9DMcuS
<ion>
My IPNS entry seems to be missing.
<gamemanj>
I'm still not sure if the script is omitting peers for some reason or another...
<whyrusleeping>
what did you use to generate that?
akhavr1 has quit [Ping timeout: 240 seconds]
<whyrusleeping>
the nostril thing is zignig
bedeho has quit [Ping timeout: 272 seconds]
sseagull has quit [Quit: Lost terminal]
<gamemanj>
whyrusleeping: Script included. You pipe the output of "ipfs diag net" into a file, then pipe that file into the script. (That's what I did, anyway - it should directly, but untested)
<ion>
Do you see my hash in ipfs diag net?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<ion>
And should cover both the sawtooth case and the huge-wantlist-in-the-beginning case.
<dignifiedquire>
daviddias: found a way to solve the horrible error popups on shutdown :) now cleanly shutting down the daemon before quitting the application and all is well
jhulten_ has joined #ipfs
Spinnaker has joined #ipfs
notduncansmith has joined #ipfs
Spinnaker has quit [Remote host closed the connection]
notduncansmith has quit [Read error: Connection reset by peer]
HostFat has joined #ipfs
mildred has quit [Ping timeout: 260 seconds]
akhavr1 has joined #ipfs
Spinnaker has joined #ipfs
akhavr has quit [Ping timeout: 264 seconds]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 240 seconds]
akhavr1 has joined #ipfs
jimbeam has joined #ipfs
akhavr has quit [Ping timeout: 264 seconds]
akhavr1 is now known as akhavr
gamemanj has joined #ipfs
jimbeam has quit [Ping timeout: 255 seconds]
akhavr1 has joined #ipfs
jimbeam has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
akhavr1 is now known as akhavr
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
akhavr1 is now known as akhavr
<whyrusleeping>
ion: the communicating weights part is done currently with a 'priority' on each wantlist entry
Poefke has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
akhavr1 is now known as akhavr
thomasreggi has joined #ipfs
Spinnaker has quit [Quit: sinked]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 268 seconds]
akhavr1 is now known as akhavr
<whyrusleeping>
as for estimating how likely a user is to fetch a block, i think we can make a few assumptions there to make our life easier
akhavr1 has joined #ipfs
<whyrusleeping>
if the user is reading a file, we can assume that they are going to keep wanting consecutive blocks
bashrc has quit [Quit: Lost terminal]
therealplato has quit [Ping timeout: 244 seconds]
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
<rendar>
whyrusleeping: let's consider a big .avi file, it will be broke in several objects, with their own hash, right?
<whyrusleeping>
currently, the reason we're seeing the wantlists shrinking down to zero is because we only fetch a single subtree at a time
<whyrusleeping>
(this happens in unixfs/io/dagreader.go)
<whyrusleeping>
we could probably just tweak that code to start fetching the next subtree once we're a certain percentage of the way through the current one
<NeoTeo>
Right, I've decided to organize an IPFS meetup in Copenhagen. If anyone here is around please check it out :) meetu.ps/2PrWWB
<whyrusleeping>
rendar: yeah, files are chunked into 256k blocks and built into a merkledag
akhavr1 has joined #ipfs
<rendar>
whyrusleeping: so every object in ipfs will have a maximum size of 256k ?
<whyrusleeping>
NeoTeo: awesome! we have a meetups page you can post on too
<whyrusleeping>
rendar: by default yes
<rendar>
whyrusleeping: ok
akhavr has quit [Ping timeout: 260 seconds]
jimbeam has quit [Ping timeout: 260 seconds]
akhavr1 is now known as akhavr
<whyrusleeping>
the upper limit is actually around 1MB, and blocks larger than that are thrown out
notduncansmith has joined #ipfs
<NeoTeo>
Cool, will do.
notduncansmith has quit [Read error: Connection reset by peer]
<rendar>
whyrusleeping: so now let's assume you want to play a very big .avi file: if you broke that in 256k chunks, you have to call a bunch of open() read() close()
<rendar>
whyrusleeping: for bittorrent which basically first download all the file chunks in the user machine, creating the big .avi file, ipfs will be able to reproduce that .avi in real time, so it means that ipfs in real time will open, read and close a bunch of file descriptors
sonatagreen has joined #ipfs
<rendar>
whyrusleeping: and, for a *very* big file, let's say, a 2GB video, it will generate many many little chunks, so the parent node must store those 256k nodes, which for a very big file, it will became very huge...
nicolagreco has joined #ipfs
<whyrusleeping>
rendar: i'm not sure what youre getting at. but yeah, we write a bunch of blocks to disk
<rendar>
whyrusleeping: i'm simply thinking possible inefficencies in doing that
<NeoTeo>
Done. Hopefully this will also provide a place for the IPFS devs to contact if you should find yourselves in the area and need a venue (or a place to crash :) )
<whyrusleeping>
storing a 2GB file will have an overhead of maybe a couple MB
<rendar>
whyrusleeping: ok
<sonatagreen>
it's not very /much/ inefficiency, a hash is a tiny fraction of 256k, it's logarithmic
<whyrusleeping>
NeoTeo: :) noted!
<NeoTeo>
Copenhagen is nice :)
<rendar>
sonatagreen: sure, but for a very big file? the node will be quite big
<whyrusleeping>
NeoTeo: so i've heard, i think mafintosh is there too. might have to come visit sometime
nicolagreco has quit [Remote host closed the connection]
<NeoTeo>
That would be most welcome.
<sonatagreen>
for a very big file, the overhead is (wild guess) like 1% of the actual data
<rendar>
sonatagreen: ok
<sonatagreen>
the most inefficiency would actually come from a directory containing lots of small files
<sonatagreen>
like, if you have a text file containing only "hello world" then its hash would actually be longer than its content
<whyrusleeping>
rendar: do you know much about inode pointer structures?
<mafintosh>
whyrusleeping: you should swing by and bring jbenet + team
<Gisle>
I get "Error: fuse failed to access mountpoint /ipfs" even when following the commands to make and mount them here https://ipfs.io/docs/commands/#ipfs-mount
<whyrusleeping>
mafintosh: that would be a ton of fun! We just might have to plan something
<rendar>
whyrusleeping: yep
<NeoTeo>
Would be awesome.
<sonatagreen>
Gisle, did you mkdir /ipfs beforehand? And set the permissions?
<whyrusleeping>
rendar: ipfs works very much the same way as that
<rendar>
whyrusleeping: i see
<Gisle>
Yeah. mkdir, chown, ipfs daemon and ipfs mount
<whyrusleeping>
Gisle: 'ls -l /dev/fuse'
<sonatagreen>
also 'ls -ld /ipfs'
<sonatagreen>
you might need to chmod it
<Gisle>
ls: cannot access /dev/fuse/: Not a directory
therealplato has joined #ipfs
<sonatagreen>
try 'ls -l /dev/fuse' instead of 'ls -l /dev/fuse/'
<Gisle>
Without the slash at the end, it's: crw-rw-rw- 1 root root 10, 229 Jul 7 09:00 /dev/fuse
<whyrusleeping>
okay, that looks fine
<whyrusleeping>
and /ipfs?
Not_ has quit [Ping timeout: 250 seconds]
<Gisle>
drwxr-xr-x 2 gisle root 4096 Oct 7 19:10 ipfs
<Gisle>
drwxr-xr-x 2 gisle root 4096 Oct 7 19:10 ipns
<Gisle>
They're both empty
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
dysbulic has joined #ipfs
captain_morgan has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
dysbulic_ has joined #ipfs
akhavr1 has quit [Ping timeout: 246 seconds]
foobear has joined #ipfs
dysbulic has quit [Ping timeout: 246 seconds]
<whyrusleeping>
Gisle: hrm... i'm not sure what the issue is then
<whyrusleeping>
maybe file an issue for us with that info
dysbulic_ is now known as dysbulic
<ion>
whyrusleeping: You said something about memory consumption in the smarter wantlist manager. This algorithm should use very little memory with the cost of every object with n children storing 2·n additional hashes on disk.
captain_morgan has quit [Ping timeout: 240 seconds]
akhavr has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<ion>
whyrusleeping: If senders already get priority information, why does the second graph show the files being downloaded throwing barely at all until a huge burst in the very end? That would be the expected result from getting the blocks in a random order.
akhavr1 has joined #ipfs
<whyrusleeping>
ion: because we dont request the next subtree until the current one is complete
akhavr has quit [Ping timeout: 264 seconds]
akhavr1 is now known as akhavr
<ion>
whyrusleeping: I meant the other case where it generates a 8k-object wantlist in the beginning.
<whyrusleeping>
ooooh, that one.
ianopolous has joined #ipfs
akhavr has quit [Ping timeout: 265 seconds]
<ion>
Uh. s/throwing/growing/
akhavr has joined #ipfs
<whyrusleeping>
i think that the best way to do this will be to take the current method of one subtree at a time, and just augment it to start the next subtree before this one has finished
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 255 seconds]
akhavr1 is now known as akhavr
<ion>
That will cover the sawtooth case but not the problem with the excessively long wantlist.
<ipfsbot>
[go-ipfs] ForrestWeston opened pull request #1811: Added an xml decoder, Fixes #1612 (master...master) http://git.io/vCeLu
ljfa has joined #ipfs
akhavr has quit [Ping timeout: 252 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
kitcambridge has joined #ipfs
yuvipanda has quit [Quit: WeeChat 1.0.1]
denizen has quit [Ping timeout: 268 seconds]
captain_morgan has joined #ipfs
amstocker_ has joined #ipfs
therealplato has quit [Ping timeout: 265 seconds]
<whyrusleeping>
i need a cloud to butt type extension that replaces 'big data' with 'cat pictures'
<achin>
mrow
f[x] has joined #ipfs
<sonatagreen>
foxreplace
<sonatagreen>
(@ whyrusleeping)
notduncansmith has joined #ipfs
akhavr has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ygrek_ has quit [Ping timeout: 244 seconds]
akhavr has quit [Ping timeout: 240 seconds]
ianopolous has quit [Ping timeout: 244 seconds]
<whyrusleeping>
sonatagreen: oooo, that looks nice
<whyrusleeping>
need to find a similar one for chromium
amstocker has joined #ipfs
akhavr has joined #ipfs
amstocker_ has quit [Ping timeout: 244 seconds]
Eudaimonstro has quit [Ping timeout: 268 seconds]
f[x] has quit [Ping timeout: 252 seconds]
<amstocker>
does anyone have a big (~1Gb) file on ipfs that I could test with?
simonv3 has joined #ipfs
rendar has quit [Ping timeout: 256 seconds]
<whyrusleeping>
amstocker: i do... but its on my lan with a 2mbps upload...
<noffle>
is jbenet flying / otherwise out of contact? he seems to be *everywhere*
<whyrusleeping>
hes in washington right now
<whyrusleeping>
but not my part of washington
<whyrusleeping>
he wont be available on irc until i think friday
<whyrusleeping>
if you need to contact him, just send me a PM and i can relay
<noffle>
awesome, thanks!
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
<whyrusleeping>
this soylent stuff is pretty nice
foobear has quit [Remote host closed the connection]
<gamemanj>
"soylent"?
<achin>
powered food
<whyrusleeping>
yeah, its basically a smoothie that doesnt taste like anything, smells like pancake batter, and is a complete 'meal'
ianopolous has joined #ipfs
amstocker has quit [Ping timeout: 255 seconds]
<gamemanj>
"Some developers eat meat..." "Some developers eat bacon..." "IPFS developers eat people ..." "The lesson is: Don't judge a product by the food the devs eat/drink. It's usually better than that estimate would give." Or is something called soylent not made of people?
<achin>
i have some, but it doesn't quite fit perfectly on my 'taste-versus-easy-of-use' curve
rendar has joined #ipfs
<M-rschulman1>
Yeah, I would completely hate to eat something that doesn't taste like anything. What's the damn point of that?
<whyrusleeping>
gamemanj: lol, i dont think its made of people
<whyrusleeping>
M-rschulman1: i can eat this now, not take any time out of my programming schedule
<noffle>
soylent !=== soylent green
<whyrusleeping>
and then use the money i save on food to buy a $30 steak for dinner at a nice restaraunt
<gamemanj>
meals that don't taste of anything are pretty rubbish "meals"... but I understand the purpose behind it
<gamemanj>
noffle: Why, do they offer a green variant?
<gamemanj>
wait, you actually somehow save money on food via this?
<M-rschulman1>
Yeah, also a very wierd choice of names.
Eudaimonstro has joined #ipfs
<whyrusleeping>
gamemanj: yeah, it costs me like, $1.40 or so a meal
<whyrusleeping>
and as far as different foods go, it makes me feel pretty good
<whyrusleeping>
especially compared to things like pizza and microwave burritos
notduncansmith has joined #ipfs
<M-rschulman1>
god I feel bad for you youngsters.
<M-rschulman1>
whyrusleeping: I just want to cook you a good meal now.
notduncansmith has quit [Read error: Connection reset by peer]
<gamemanj>
whyrusleeping: Do you at least add sugar or something
<whyrusleeping>
M-rschulman1: hahaha, i wont say no. But I do enjoy cooking every so often myself, made proscuitto tortellini with a nice white wine sauce and some fresh italian sausage from the butcher
<whyrusleeping>
as well as a goat cheese, arugula, toasted hazelnut and grape salad
<multivac>
[WIKIPEDIA] Soylent Green | "Soylent Green is a 1973 American science fiction film directed by Richard Fleischer and starring Charlton Heston, Leigh Taylor-Young, and, in his final film, Edward G. Robinson. The film combines the police procedural and science fiction genres, depicting the investigation into the murder of a wealthy..."
<whyrusleeping>
with a nice vinaggrette
<gamemanj>
whyrusleeping: Ah, so you do use your taste buds then
<whyrusleeping>
of course :)
<whyrusleeping>
but when i do so less frequently it becomes more of a treat
<gamemanj>
whyrusleeping: I was worried... I thought I might have to hide in a blanket of "the future is not tasteless"
<whyrusleeping>
lol
<gamemanj>
the future is distributed, relatively easy to use, and not tasteless
<gamemanj>
that's good
hummus has joined #ipfs
akhavr1 has joined #ipfs
<M-rschulman1>
whyrusleeping: Ok, phew.
<whyrusleeping>
the future is really just thai food though
<whyrusleeping>
get me some chicken pad see ew any day of the week
<M-rschulman1>
well, that I can agree with.
<noffle>
blade runner style
akhavr has quit [Ping timeout: 256 seconds]
akhavr1 is now known as akhavr
<M-rschulman1>
There's a new burmese place a few blocks from my house that I want to try
<whyrusleeping>
yeah, soylent is really just about not interrupting my productive hours with a big distracting lunch break
<whyrusleeping>
if i'm hacking away on something, i dont want to stop
HostFat has quit [Read error: Connection reset by peer]
<noffle>
whyrusleeping: big lunches tend to lead to big sleepiness, too. how does soylent rank on that scale?
<gamemanj>
presumably low, if it's a smoothie
<noffle>
right; cheap to digest
<whyrusleeping>
yeap
<whyrusleeping>
i havent noticed any sluggishness caused by it
<whyrusleeping>
i also generally have it alongside a small cup of coffee
<gamemanj>
is that any increased sluggishness?
<noffle>
is one sufficient for a meal?
<whyrusleeping>
i normally drink about one 6-8oz glass
<whyrusleeping>
and then a cup of coffee and about 12oz of water
<hummus>
no need to go in depth on this, but could you somehow enumerate over all the files in ipfs, like for a search index?
<whyrusleeping>
hummus: nope, best you can do is listen on the dht for provider messages
<gamemanj>
hummus: Probably can't enumerate over everything in IPFS, but A. you can do what whyrusleeping is suggesting, and B. in order to find a good place to start, get every peerid from "ipfs diag net"
<gamemanj>
and do ipns lookups
<gamemanj>
admittedly that will become impractical as the network gets bigger
<hummus>
okay. just had an interesting idea for an automated backup system and was curious how much of a project it would be
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 246 seconds]
akhavr1 is now known as akhavr
Xubaka has joined #ipfs
Xubaka has quit [Client Quit]
Xubaka has joined #ipfs
Xubaka has quit [Client Quit]
<whyrusleeping>
pleaseeeee dont abuse 'ipfs diag net'
<whyrusleeping>
its set to be removed when the network gets large enough and the bandwidth it consumes gets large enough
akhavr1 has joined #ipfs
<sonatagreen>
perhaps it might be more sensible to just cap how much it can do?
<sonatagreen>
so it returns an appropriately modestly sized subset of the network
akhavr has quit [Ping timeout: 240 seconds]
akhavr1 is now known as akhavr
<whyrusleeping>
yeah, thats going to be implemented soon
<whyrusleeping>
well at least rate limiting is
<whyrusleeping>
giving less than the whole picture of the network would make the tool not do what it was made to do, lol
<gamemanj>
Which is absolutely fine when "everybody" is < 100.
<sonatagreen>
yeah, that is. not feasible.
<gamemanj>
Oh, it's quite feasible, at least right now . Sane? No.
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<hummus>
it would also be fine if it's not done that often, i.e. have a public index
<whyrusleeping>
its absolutely not sane :D
hellertime has quit [Read error: Connection reset by peer]
<whyrusleeping>
yeah, one option was 'respond to diags from trusted peers'
<whyrusleeping>
such as whyrusleepings node ;)
hellertime has joined #ipfs
<M-rschulman1>
psh, like I would trust that whyrusleeping guy
voxelot has quit [Ping timeout: 272 seconds]
<gamemanj>
well, it seems whyrusleeping is unlikely to try sending us all diags for the purposes of evilness
<whyrusleeping>
yeah, i'm trustworthy!
<whyrusleeping>
now if you could all run this random unsigned binary for me...
<whyrusleeping>
and when it asks for your password, thats totally legit
<whyrusleeping>
</sarcasm>
<sonatagreen>
i mean, /i/ didn't personally review the ipfs code before running it on my machine
<whyrusleeping>
sonatagreen: yeah, ideally we get it into package managers, so the package maintainers can do the reviews for you
<whyrusleeping>
and also provide signatures and our public keys for verification
<sonatagreen>
pubsigs are good
akhavr has quit [Ping timeout: 246 seconds]
<whyrusleeping>
but in general, i'm against just arbitrarily trusting someone like that
<whyrusleeping>
or rather, special casing node behaviour for certain people
akhavr has joined #ipfs
<sonatagreen>
right
<sonatagreen>
don't want an aristocracy
<sonatagreen>
although f2f wot is a thing
Spinnaker has joined #ipfs
<sonatagreen>
which is not really the same philosophically but involves treating your friends' nodes differently
Brender has joined #ipfs
<gamemanj>
the user deciding explicitly that someone is special is fine, the user being told "$NAME is your friend" not so much
<sonatagreen>
right
<gamemanj>
but... couldn't the same thing be applied to content? Talking about the webui here, ofc...
<sonatagreen>
there's a difference between trusting a file and trusting a person i think
<sonatagreen>
the webui points to an ipfs address not an ipns address
<sonatagreen>
you can, in principle, inspect the code to make sure it's not doing anything sinister
<sonatagreen>
and that risk of detection keeps people honest
<sonatagreen>
trusting someone that they're not /going to/ do anything evil is pretty different
<sonatagreen>
because there's no way to verify
<sonatagreen>
until it's too late
<konubinix>
whyrusleeping: I don't need to debug anything in particular. I just wanted to dig into the code
<whyrusleeping>
gamemanj: yeah, the webui is basically part of the ipfs daemon itself
therealplato has joined #ipfs
<whyrusleeping>
we just didnt want to embed it into the binary
<whyrusleeping>
its hash is hardcoded in the binary though
<whyrusleeping>
(you can change that out if you want too)
akhavr has quit [Quit: akhavr]
<konubinix>
whyrusleeping: In most of the projects I contributed to, I figured out that running the code step by step was a good introduction
akhavr has joined #ipfs
<whyrusleeping>
konubinix: huh. That gets a little difficult in ipfs, you can use delve, but ipfs ends up with looooots of different goroutines running around
<whyrusleeping>
so its hard to follow down a sequential code path
<whyrusleeping>
once the daemon initializes, the main code path just sits and waits for things to close down
<konubinix>
whyrusleeping: That is not generally problematic, since once one sees the goroutine launch in the main code, he can add a breakpoint in the goroutine to see what happens
<konubinix>
At least as far as I can understand
<konubinix>
Actually, I understand that ipfs is highly asynchronous
<whyrusleeping>
okay, then i would try out delve
rschulman has quit []
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
f[x] has joined #ipfs
<konubinix>
whyrusleeping: Thank you for the hint. I never managed to build it, but I will definitely try again :-)
NeoTeo has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
j0hnsmith has joined #ipfs
j0hnsmith has quit [Client Quit]
Brender has quit [Remote host closed the connection]
unjust has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
akhavr1 is now known as akhavr
<gamemanj>
I've found a problem... the latest release of the Java IPFS API isn't actually usable, as things like "IPFSObject.put" aren't accessible, because the IPFSObject class isn't public... so it's functions can't be accessed outside of the org.ipfs package. Unless this is somehow intentional?
Encrypt has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 256 seconds]
akhavr1 is now known as akhavr
Encrypt has quit [Client Quit]
martinBrown has quit [Ping timeout: 240 seconds]
yosafbridge has quit [Ping timeout: 240 seconds]
Gisle has quit [Quit: Gisle]
akhavr1 has joined #ipfs
martinBrown has joined #ipfs
<dignifiedquire>
whyrusleeping: just pushed the first working jellyfish if you want to try it out ;)
yosafbridge has joined #ipfs
<dignifiedquire>
though you’ll only see it if you are initializing a new node
<M-rschulman1>
is the electron-app usable by mortals or still more in beta?
<M-rschulman1>
or alpha?
akhavr1 has joined #ipfs
Xe has joined #ipfs
dysbulic__ has joined #ipfs
akhavr has quit [Ping timeout: 268 seconds]
akhavr1 is now known as akhavr
dysbulic has quit [Ping timeout: 246 seconds]
dysbulic__ is now known as dysbulic
<dignifiedquire>
M-rschulman: hmm hard work in progress, so no packaged version yet but it’s running quite well for me, so you can try the version from the linked PR above
l has joined #ipfs
l is now known as Guest64533
dysbulic_ has quit [Ping timeout: 240 seconds]
jimbeam has joined #ipfs
ljfa has quit [Disconnected by services]
akhavr1 has joined #ipfs
Guest64533 is now known as ljfa
<M-rschulman1>
dignifiedquire: Cool
<M-rschulman1>
I'm pretty content just running it in the term for now too.
<dignifiedquire>
M-rschulman1: and please report all the bugs :)
<multivac>
[WIKIPEDIA] Magnet URI scheme#URN.2C containing hash .28xt.29 | "The Magnet URI scheme is a de facto standard defining a uniform resource identifier (URI) scheme for Magnet links, which mainly refer to resources available for download via peer-to-peer networks. Such a link typically identifies a file not by location, but by content—more precisely, by the content's..."
akhavr has quit [Ping timeout: 264 seconds]
akhavr1 is now known as akhavr
<substack>
considering having ipfs links work with magnet:?xt=urn:ipfs:... in hyperboot
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<sonatagreen>
yes please
<giodamelio>
That is pretty neat
<sonatagreen>
especially since a magnet link can specify multiple sources
voxelot has joined #ipfs
voxelot has quit [Changing host]
voxelot has joined #ipfs
<ion>
“<@whyrusleeping> sonatagreen: yeah, ideally we get it into package managers, so the package maintainers can do the reviews for you” Yeah, I'm sure the package maintainers have audited every single line of code I have installed.
bsm1175321 has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 272 seconds]
akhavr1 has quit [Ping timeout: 255 seconds]
akhavr has joined #ipfs
amade has quit [Quit: leaving]
jimbeam has quit [Ping timeout: 260 seconds]
akhavr has quit [Ping timeout: 265 seconds]
ianopolous has quit [Ping timeout: 250 seconds]
<whyrusleeping>
ion: lol, its a *little* better than curling into sudo bash
<ion>
Oh, right, i had that ipfs diag net running. I killed it after 4 hours, let me restart the daemon and try again.
<ion>
whyrusleeping: Fair enough. :-)
<whyrusleeping>
ion: you can set a timeout with --timeout=20m
<whyrusleeping>
or something
<ion>
Immediately after restarting the daemon, it finished in 19 seconds.
<whyrusleeping>
yeah
<whyrusleeping>
it shouldnt take long at all...
akhavr has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
hellertime has quit [Ping timeout: 260 seconds]
<gamemanj>
Good news: If you follow the instructions as written in README.md ("npm i" followed by "npm start"), the IPFS interface works. Downside is that it decided to install a second copy of Electron (and presumably just about every other NPM package being used) to do so. And I'm still waiting for the tray icon to show up.
captain_morgan has quit [Ping timeout: 240 seconds]
<gamemanj>
s/IPFS interface/electron-app/
<multivac>
gamemanj meant to say: Good news: If you follow the instructions as written in README.md ("npm i" followed by "npm start"), the electron-app works. Downside is that it decided to install a second copy of Electron (and presumably just about every other NPM package being used) to do so. And I'm still waiting for the tray icon to show up.
NeoTeo has joined #ipfs
<ion>
Has that multivac function ever been helpful? :-P
<whyrusleeping>
lol, that one might be a bit overboard
amade has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 252 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 256 seconds]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 268 seconds]
akhavr1 has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr has quit [Ping timeout: 264 seconds]
<dignifiedquire>
gamemanj: has it started for you? have you stopped your ipfs daemon before?
<gamemanj>
Wait, I need to stop the ipfs daemon before using it?
atrapado has joined #ipfs
akhavr1 has quit [Ping timeout: 272 seconds]
akhavr has joined #ipfs
Xubaka has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 272 seconds]
akhavr1 is now known as akhavr
josephdelong has quit [Quit: josephdelong]
akhavr1 has joined #ipfs
<dignifiedquire>
gamemanj: yes, it starts its own daemon
akhavr has quit [Ping timeout: 252 seconds]
akhavr1 is now known as akhavr
captain_morgan has joined #ipfs
faddat has joined #ipfs
ianopolous has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 264 seconds]
akhavr1 is now known as akhavr
amade has quit [Quit: leaving]
dignifiedquire has quit [Quit: dignifiedquire]
akhavr has quit [Ping timeout: 250 seconds]
akhavr has joined #ipfs
jimbeam has joined #ipfs
akhavr1 has joined #ipfs
gamemanj has quit [Ping timeout: 256 seconds]
akhavr has quit [Ping timeout: 250 seconds]
akhavr1 is now known as akhavr
ljfa has quit [Quit: Leaving]
Xubaka has quit [Quit: Leaving]
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
akhavr1 is now known as akhavr
akhavr1 has joined #ipfs
akhavr has quit [Ping timeout: 250 seconds]
NeoTeo has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
akhavr has joined #ipfs
akhavr1 has quit [Ping timeout: 240 seconds]
<SWingedSeraph>
I'm looking to build a set of links to interesting content in IPFS, do any of you know of any good entries?
jimbeam has quit [Ping timeout: 255 seconds]
<faddat>
I do not, BUT I think that's a really great idea.
<SWingedSeraph>
I was hoping it would drive a little more interest in IPFS. I think I may publish my GNU/Linux tutorials exclusively in IPFS (aside from the GitHub repo, or course)
<faddat>
That's a really cool idea!
<faddat>
have you published anything into IPFS yet? If so, can you link me so I can havea look-see at the changes needed to host content in IPFS?
<SWingedSeraph>
I haven't, mostly because I'm currently thinking out the best way to do so. I think what I may do is publish an immutable front page with IPNS links to the actual tutorials and perhaps a small homepage, which will let me have a bit more flexibility
<SWingedSeraph>
The main problem is that I don't have a server from which to host, just a Ras Pi that's already running a Mumble server and an IRC bouncer and wolfram mathematica
<SWingedSeraph>
Thanks achin, I'll take a look
<faddat>
oh!
<faddat>
holy crap!
<ion>
What was the name of the program someone was developing for publishing web pages on IPFS? ipsurge?
captain_morgan has quit [Ping timeout: 240 seconds]
<faddat>
@SWingedSeraph, I happen to be commercially solving that problem right as we speak, do you want to be free beta tester?
<SWingedSeraph>
@faddat I would love to be
<SWingedSeraph>
Also, has anyone considered making a browser plugin that would replace "ipfs.io/ipfs" with "localhost:8080/ipfs" to reduce load on ipfs.io?