Adbray has quit [Remote host closed the connection]
larpanet has quit [Ping timeout: 240 seconds]
jamiew has quit [Ping timeout: 250 seconds]
jamiew has joined #ipfs
<massino>
It doesn't look like the multiaddr spec supports local domain sockets... Does that mean it's not possible to eg. configure p2p streams to connect to, and listen on, domain sockets?
lanzafame has quit [Ping timeout: 248 seconds]
<whyrusleeping>
massino: we can definitely add unix domain sockets to multiaddr
<whyrusleeping>
We just havent done it yet
* massino
adding to my contribution todo-list :)
lanzafame has joined #ipfs
<lgierth>
they are specced but not implemented, iirc
<lgierth>
it's /unix/path/to/socket.sock
noobineer has joined #ipfs
noobineer has quit [Max SendQ exceeded]
<whyrusleeping>
lgierth: how would one do ipfs over that?
<lgierth>
doesn't have to be ipfs over that
<lgierth>
can be /http or anything
<lgierth>
i was just saying you can have /unix multiaddrs
<whyrusleeping>
lgierth: i know, i'm asking how you would nest them?
<whyrusleeping>
thats the tricky one as i remember it
<lgierth>
they way we do /ws currently can't address a ws endpoint at a specific subpath
<lgierth>
same for addressing something *within* a unix socket
<lgierth>
i've become a multiaddr nerd over the years
<whyrusleeping>
i love it
<lgierth>
ah, and there's a third option, simply escaping slashes
<lgierth>
so we have a type option, a syntax option, and an escaping option
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
dhruvbaldawa has joined #ipfs
shoogz has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
f33d has quit [Quit: Leaving...]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
dhruvbaldawa has quit [Ping timeout: 248 seconds]
teod has joined #ipfs
<whyrusleeping>
lgierth: how hard would it be for me to deploy a custom version of ipfs to biham?
toxync01 has quit [Ping timeout: 240 seconds]
<lgierth>
not hard at all
<lgierth>
just need a hash
<whyrusleeping>
mmmkay
erictapen has quit [Ping timeout: 240 seconds]
erictapen has joined #ipfs
erictapen has quit [Remote host closed the connection]
erictapen has joined #ipfs
<whyrusleeping>
lgierth: mind doing the exp/sync-concurrency branch?
<lgierth>
ok!
<lgierth>
that one could be worthy of a config setting (with boundaries)
<whyrusleeping>
yeah, if this pans out then definitely
<whyrusleeping>
either that, or an option some other way (like a flag?)
<whyrusleeping>
ideally we shouldnt have to tweak it and it should just be smart about things
Alpha64 has joined #ipfs
<lgierth>
better be able to tweak it, the best value for it is dependent on what data you're dealing with and what your resources are
<lgierth>
biham: ref changed from b5794c4 to 22255db
Mateon1 has quit [Ping timeout: 268 seconds]
Mateon3 has joined #ipfs
Reinhilde is now known as Ellenor
Mateon3 is now known as Mateon1
larpanet has joined #ipfs
<whyrusleeping>
lgierth: is 172.31.32.20 a public range ip?
Xiti has quit [Ping timeout: 240 seconds]
<Icefoz_>
whyrusleeping: Nope.
lexruee has joined #ipfs
<whyrusleeping>
hrm...
<Icefoz_>
172.16.0.0 - 172.31.255.255 is private
toxync01 has joined #ipfs
anewuser has joined #ipfs
<whyrusleeping>
lgierth: if youre up, i pushed another change to that branch. we were limiting the concurrency elsewhere too
gozala has quit [Ping timeout: 240 seconds]
gozala has joined #ipfs
<lgierth>
biham: ref changed from 22255db to f1af17c
<lgierth>
whyrusleeping: ^ :)
<whyrusleeping>
lgierth: <3
<whyrusleeping>
so, when using a bitswap session, the optimal strategy is to shove as many requests into it as possible and let it handle the rate limiting
<whyrusleeping>
previously, the code calling bitswap had to kinda manage that to prevent things from getting out of hand
<whyrusleeping>
so the next step is to make the merkledag.GetMeAGreatNumberOfThings type calls just call `GetBlocks` as fast as possible
<whyrusleeping>
then we need to have blocks we receive get batched and written in parallel somehow
larpanet has quit [Remote host closed the connection]
larpanet has joined #ipfs
jaboja has quit [Remote host closed the connection]
larpanet has quit [Remote host closed the connection]
larpanet has joined #ipfs
<drew55>
been thinking lots on to dovetail pubsub equitable machiny into filecoin. cachers get rewarded methinks? a speculation on what traffic will be most popular in the near future
erictapen has quit [Ping timeout: 248 seconds]
<drew55>
caching is a market?
<drew55>
caching is an important and resource requiring utility
<drew55>
* machinery
anewuser has quit [Ping timeout: 248 seconds]
dhruvbaldawa has joined #ipfs
anewuser has joined #ipfs
Jesin has joined #ipfs
Alpha64 has joined #ipfs
larpanet has quit [Remote host closed the connection]
larpanet has joined #ipfs
anewuser has quit [Read error: Connection reset by peer]
vindelschtuffen has joined #ipfs
todder has quit [Quit: is such a quitter...]
todder has joined #ipfs
anon136 has joined #ipfs
<anon136>
oh hey cool. so glad this is a thing!
<anon136>
i was like hey you know i bet there is a freenode chatroom for ipfs and sure enough
bloodsta1ker has quit [Ping timeout: 268 seconds]
bloodstalker has joined #ipfs
<anon136>
well anyway i need a hash to test out to see if my server is working properly. does anyone know a good resource for this?
bloodsta1ker has joined #ipfs
bloodstalker has quit [Ping timeout: 248 seconds]
infinity0_ has joined #ipfs
infinity0_ has quit [Changing host]
infinity0 has joined #ipfs
infinity0 has quit [Killed (card.freenode.net (Nickname regained by services))]
larpanet has quit [Ping timeout: 248 seconds]
lanzafame has quit [Ping timeout: 248 seconds]
lanzafame has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
<Icefoz_>
anon136: Just add a file or two to IPFS and get the hash from that.
<anon136>
Yea but I dont wanna pull a file from my own machine. Whats the fun in that? Anyway I found something. QmWenbjgZnA6UguLtmUYayS6e7UQM7woB15zuEymSRRMoi im guessing its that free blender video with the rabbits
<miko>
Another question i have is, are there any file size limits?
erictapen has joined #ipfs
<JCaesar>
non that would concern you, I think. There are multi-TB-files somewhere out there…
<miko>
wow thats impressive
<miko>
so potentially you could use ipfs as an infinite storage system.
<JCaesar>
well, you still have to have some node that pins those files.
<miko>
How would you know how many seeds, or replications your file has?
<JCaesar>
I dunno about that. I suspect that you will know if you read about filecoin. (it's not so easy to prove that you actually have a copy of a file.)
<miko>
Ahh yea I was just starting to look into that now.
larpanet has quit [Remote host closed the connection]
<engdesart>
IIRC, they're broadcasted when pinned/added, to all nodes that one is connected to.
ZaZ1 has joined #ipfs
rpost__ has quit [Ping timeout: 248 seconds]
<JCaesar>
That would seem like a rather stupid idea then…
rpost__ has joined #ipfs
graffen has quit [Remote host closed the connection]
ZaZ has quit [Ping timeout: 248 seconds]
ZaZ has joined #ipfs
ZaZ1 has quit [Ping timeout: 240 seconds]
graffen has joined #ipfs
<raziellight>
lol it's funny you ask miko, because i just realized the potential to browze through some ideas and applying those ideas to stuff like zeronet and ipfs
<raziellight>
in other words, it's something that i've just begun trying to wrap my head around
<raziellight>
a system that is
etched has quit [Quit: etched]
Ellenor is now known as Reinhilde
GeoCode has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
revisor has joined #ipfs
rpost__ has quit [Ping timeout: 268 seconds]
ZaZ has quit [Read error: Connection reset by peer]
rpost__ has joined #ipfs
ilyaigpetrov has joined #ipfs
someone_ has joined #ipfs
someone_ is now known as someone42
someone42 is now known as someoneelse
someoneelse is now known as too_lazy_to_choo
<too_lazy_to_choo>
Hi all
<too_lazy_to_choo>
I am working on a libp2p project
<too_lazy_to_choo>
Is there a way to replace the secio part?
<too_lazy_to_choo>
Aka implementing my version
<too_lazy_to_choo>
Planning to do something like implementing a transport wrapper, but maybe there is a better way?
Encrypt has joined #ipfs
speaker[m] has joined #ipfs
blocklock has joined #ipfs
too_lazy_to_choo has quit [Ping timeout: 260 seconds]
blocklock has quit [Quit: Page closed]
Encrypt has quit [Quit: Quit]
<miflow[m]>
too_lazy_to_choo: I have little expertise in libp2p, but wouldn't you write an own connection module then
PyHedgehog has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
clickjac_ has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
dhruvbaldawa has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
clickjack has quit [Ping timeout: 248 seconds]
f33d has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
qnix has joined #ipfs
qnix_ has joined #ipfs
larpanet has quit [Ping timeout: 248 seconds]
toxync01 has quit [Ping timeout: 268 seconds]
rpost__ has quit [Ping timeout: 240 seconds]
qnix has quit [Ping timeout: 248 seconds]
rpost__ has joined #ipfs
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
qnix_ is now known as qnix
rcat has joined #ipfs
nek0 has quit [Ping timeout: 252 seconds]
pat36 has quit [Read error: Connection reset by peer]
PyHedgehog has quit [Quit: Connection closed for inactivity]
erictapen has quit [Ping timeout: 258 seconds]
plexigras has quit [Quit: WeeChat 1.9.1]
plexigras has joined #ipfs
chris613 has quit [Quit: Leaving.]
chris613 has joined #ipfs
<raziellight>
i'm trying to understand merkle trees better. anyone know them?
erictapen has joined #ipfs
<r0kk3rz>
it a tree made of merkles, whats to understand
}ls{ has quit [Ping timeout: 248 seconds]
dhruvbaldawa has quit [Remote host closed the connection]
}ls{ has joined #ipfs
Wikipedia[m] has joined #ipfs
<Kythyria[m]>
raziellight: The basic principle is that it's a tree where each node refers to its children by hash (or including a hash).
<Kythyria[m]>
Has the nice property that you can sign the root and that indirectly encompasses all the other nodes. You can also extend the principle to a DAG; Git, Matrix, and IPFS all do that.
<Kythyria[m]>
The resulting structure is immutable but can reuse identical subgraphs: if any node changes all the nodes pointing to it have to change.
jaboja has quit [Ping timeout: 248 seconds]
jaboja has joined #ipfs
Encrypt has joined #ipfs
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
Alpha64 has quit [Ping timeout: 248 seconds]
larpanet has joined #ipfs
Alpha64 has joined #ipfs
jaboja has quit [Ping timeout: 248 seconds]
lassulus has quit [Ping timeout: 240 seconds]
larpanet has quit [Ping timeout: 240 seconds]
lassulus has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
vivus has joined #ipfs
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
jkilpatr has quit [Ping timeout: 240 seconds]
shoogz has joined #ipfs
shoogz has quit [Max SendQ exceeded]
jaboja has joined #ipfs
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
bsm117532 has quit [Ping timeout: 258 seconds]
jkilpatr has joined #ipfs
Encrypt has quit [Quit: Quit]
nek0 has joined #ipfs
dhruvbaldawa has joined #ipfs
dhruvbaldawa has quit [Ping timeout: 268 seconds]
<raziellight>
ya so there needs to be a set number of children before the root merkle is created
<raziellight>
and one problem i'm facing is how to validate root merkles if you're going to advertise them as a location
larpanet has joined #ipfs
rpost__ has quit [Ping timeout: 248 seconds]
rpost__ has joined #ipfs
<raziellight>
so in a DAG you have several children. they can converge or diverge, in IPLD there are links. the links go down? but that will constantly change the root hash, which doesn't make sense
Tolik[m] has joined #ipfs
<raziellight>
but it is a lot more clear than it was
<raziellight>
or there are a bunch of links are created, and then the root is calculated, and it remains permanent
apiarian has joined #ipfs
<raziellight>
another thing about dags i don't understand is when you link down(seemingly) you are branching out, so the root must be at the top. but how do you have the root before you know the children
Alpha64_ has joined #ipfs
Alpha64 has quit [Ping timeout: 240 seconds]
<Kythyria[m]>
You have to start at the leaves and work up towards the root when generating the graph, yes.
<Kythyria[m]>
However, once you've done that you can start at the root and descend.
someone_lazy has joined #ipfs
larpanet has quit [Ping timeout: 240 seconds]
DuClare has quit [Ping timeout: 240 seconds]
someone_lazy has left #ipfs [#ipfs]
someone_lazy has joined #ipfs
DuClare has joined #ipfs
Encrypt has joined #ipfs
hmagellan has joined #ipfs
robattila256 has quit [Quit: WeeChat 1.9.1]
onabreak has joined #ipfs
fredthomsen has joined #ipfs
robattila256 has joined #ipfs
someone_lazy has quit [Quit: WeeChat 1.9.1]
jaboja has quit [Ping timeout: 248 seconds]
noperope has joined #ipfs
noperope has quit [Client Quit]
jzeus has joined #ipfs
Jesin has quit [Quit: Leaving]
<pjz>
directionality is confusing.
<pjz>
I think of new nodes as 'children' of old ones; so child nodes contain the hash of their parents.
plexigras has quit [Ping timeout: 248 seconds]
pat36 has quit []
erictapen has quit [Ping timeout: 248 seconds]
<massino>
From the point of view of an ipfs user, how will filecoin affect me? Is it an optional add-on, a new subcommand? Will it require a breaking upgrade? Will it wrap ipfs completely, bifurcating its community? Or hook into it?
jaboja has joined #ipfs
brabo has joined #ipfs
onabreak has quit [Ping timeout: 260 seconds]
Intensity has quit [Ping timeout: 246 seconds]
Caterpillar has quit [Quit: You were not made to live as brutes, but to follow virtue and knowledge.]
erictapen has joined #ipfs
<r0kk3rz>
it'll be seperate and hook into ipfs afaik
onabreak has joined #ipfs
Caterpillar has joined #ipfs
Intensity has joined #ipfs
larpanet has joined #ipfs
vivus has left #ipfs ["Leaving"]
GeoCode has joined #ipfs
Alpha64_ has quit [Ping timeout: 248 seconds]
code-glitch has quit [Ping timeout: 248 seconds]
Encrypt has quit [Quit: Quit]
code-glitch has joined #ipfs
<anon136>
When I started the daemon it gave me a command like ipfs cat /ipfs/?????????????????/readme. How do I find that hash again? And where is the data that it references being stored on my machine? Is it abstracted out of the files in ~/.ipfs and so purely virtual?
<anon136>
Like what is the actual location of the readme that it opens?
GeoCode has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]