infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
kthnnlg has quit [Remote host closed the connection]
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
infinity0 has quit [Remote host closed the connection]
Foxcool_ has quit [Ping timeout: 240 seconds]
infinity0 has joined #ipfs
infinity0 has quit [Changing host]
infinitesum has joined #ipfs
infinity0 has joined #ipfs
infinity0 has quit [Remote host closed the connection]
charley has joined #ipfs
infinity0 has joined #ipfs
infinity0 has quit [Remote host closed the connection]
elkalamar has quit [Ping timeout: 255 seconds]
bingus908 is now known as bingus
charley has quit [Remote host closed the connection]
ipfsrules has quit [Ping timeout: 245 seconds]
ipfsrules has joined #ipfs
infinitesum has quit [Quit: infinitesum]
jsgrant has joined #ipfs
droman has quit []
cyanobacteria has joined #ipfs
chris613 has joined #ipfs
btmsn has quit [Ping timeout: 240 seconds]
ipfsrules has quit [Ping timeout: 260 seconds]
ipfsrules has joined #ipfs
pawn has joined #ipfs
<pawn>
I'm putting together a quick keynote/talk about IPFS at one of my local tech communities. Is there any way I get get ahold of a keynote file from one of Juan's presentations? :D
chris613 has left #ipfs [#ipfs]
cyanobacteria has quit [Ping timeout: 240 seconds]
elkalamar has joined #ipfs
infinitesum has joined #ipfs
spossiba has quit [Quit: Lost terminal]
spossiba has joined #ipfs
_whitelogger has joined #ipfs
brianhoffman_ has joined #ipfs
brianhoffman has quit [Ping timeout: 255 seconds]
brianhoffman_ is now known as brianhoffman
GuestGuillev[m] has joined #ipfs
<ipfsrules>
Whats the general eta of ipns in js-ipfs?
<Guest61724>
Just using a home router, I don't know that make or model
<Guest61724>
Arris
<lgierth>
independent of what you're building there, does `ipfs id` show the correct external addresses if you're running a daemon?
<Guest61724>
I simply want to display the public address of my libp2p client
<Guest61724>
Which would be the address of my nat plus the port mapping
<lgierth>
the one that has your public ip address in? :)
<Guest61724>
Where is that function implemented? I'm only using libp2p.
<lgierth>
libp2p/go-libp2p-nat package
<Guest61724>
So this notifier is passed in as a parameter when instantiating a host?
<Guest61724>
basichost.New only takes two parameters from what I can see
<lgierth>
it takes 1+ arguments, the second one can be any number of arguments
<lgierth>
gotta go, bed's calling!
<lgierth>
check out first whether `ipfs id` comes up with your correct external addresses -- if it doesn't there's an issue with your NAT and/or libp2p
<lgierth>
and then independent of that it makes sense to make natManager in the basichost package a public thing
<lgierth>
so that you can pass it into basichost.NewHost()
maxguru[m]1 has left #ipfs ["User left"]
jaboja has joined #ipfs
jsgrant has quit [Ping timeout: 260 seconds]
libman has joined #ipfs
<libman>
Random thought of the day, via https://github.com/qntm/base65536 - why use base58 encoding for hash address strings when you just want to be able to copy-paste them while taking up as little screen space as possible?
ipfsrules has quit [Ping timeout: 260 seconds]
jsgrant has joined #ipfs
ipfsrules has joined #ipfs
lvernschrock has joined #ipfs
lvernschrock has left #ipfs [#ipfs]
cyanobacteria has quit [Quit: Konversation terminated!]
cyanobacteria has joined #ipfs
cyanobacteria has quit [Client Quit]
cyanobacteria has joined #ipfs
ianopolous has quit [Read error: Connection reset by peer]
cyanobacteria has quit [Client Quit]
cyanobacteria has joined #ipfs
cyanobacteria has quit [Client Quit]
cyanobacteria has joined #ipfs
skeuomorf has quit [Ping timeout: 240 seconds]
infinity0 has joined #ipfs
ipfsrules has quit [Ping timeout: 260 seconds]
ipfsrules has joined #ipfs
jsgrant has quit [Ping timeout: 240 seconds]
JayCarpenter has quit [Quit: Page closed]
talonz has quit [Read error: Connection reset by peer]
talonz has joined #ipfs
Guest80280 has joined #ipfs
infinitesum has quit [Quit: infinitesum]
Foxcool has joined #ipfs
jaboja has quit [Ping timeout: 268 seconds]
Mateon3 has joined #ipfs
ipfsrules has quit [Ping timeout: 245 seconds]
Mateon3 is now known as Mateon1
Mateon1 has quit [Ping timeout: 245 seconds]
ipfsrules has joined #ipfs
talonz has quit [Remote host closed the connection]
talonz has joined #ipfs
kaiega has joined #ipfs
kaiega has left #ipfs [#ipfs]
gmoro_ has joined #ipfs
zuck05 has quit [Ping timeout: 240 seconds]
Guest80280 has quit [Ping timeout: 255 seconds]
gmoro has quit [Ping timeout: 255 seconds]
Aranjedeath has quit [Quit: Three sheets to the wind]
barnacs has quit [Ping timeout: 255 seconds]
kanzure has quit [Ping timeout: 255 seconds]
barnacs has joined #ipfs
ipfsrules has quit [Ping timeout: 260 seconds]
kanzure has joined #ipfs
akkad has quit [Ping timeout: 255 seconds]
ipfsrules has joined #ipfs
reit has quit [Ping timeout: 255 seconds]
reit has joined #ipfs
zuck05 has joined #ipfs
zuck05 has quit [Remote host closed the connection]
cehteh has quit [Ping timeout: 272 seconds]
sirdancealot has joined #ipfs
akkad has joined #ipfs
brianhoffman has quit [Read error: Connection reset by peer]
Guest61724 has quit [Quit: This computer has gone to sleep]
brianhoffman has joined #ipfs
Caterpillar has joined #ipfs
ipfsrules has quit [Ping timeout: 260 seconds]
cehteh has joined #ipfs
ipfsrules has joined #ipfs
espadrine has joined #ipfs
ipfsrules has quit [Ping timeout: 260 seconds]
ipfsrules has joined #ipfs
tilgovi has joined #ipfs
enzo__ has joined #ipfs
rendar has joined #ipfs
ipfsrules has quit [Ping timeout: 240 seconds]
ipfsrules has joined #ipfs
ipfsrule1 has joined #ipfs
ipfsrules has quit [Ping timeout: 260 seconds]
libman has quit [Quit: Connection closed for inactivity]
pawn has joined #ipfs
<pawn>
How does an IPFS directory name it's contents? What does that merkle tree look like?
ipfsrule1 has quit [Ping timeout: 255 seconds]
<lemmi>
pawn: ipfs object get $HASH
<lemmi>
pawn: this will give you a json representation of that entry.
<pawn>
So a directory is just a JSON file?
<pawn>
No no.
<pawn>
You said "representation"
<pawn>
So it's a file though right?
<pawn>
a directory is a file that contains a list of other hashes with names to those hashes?
<lemmi>
pawn: it can be a file, doesn't have to.
<pawn>
What I mean is, a directory hash changes when you change the directory in any way?
<lemmi>
pawn: yes
<pawn>
And it's a merkle tree?
<lemmi>
pawn: it's a directed acyclic graph. so it can be a tree, but also a little more
<pawn>
I meant merkle DAG hehe
m0ns00n has joined #ipfs
<lemmi>
pawn: fwiw, the ipfs object command is tailored towards files. the generalization of that is using the ipfs dag command to get access to arbitrary dags
<pawn>
So a DAG isn't a tree?
<pawn>
It's a graph?
<pawn>
And a tree is a type of graph, but a DAG isn't a type of tree?
<lemmi>
pawn: use you google powers to clear that up :)
<pawn>
Excuse my rudimentary questions... :P
* pawn
needs to finish these slides and it's getting late
<lemmi>
pawn: np. but in that case there are better sources than me to clear things up
_whitelogger has joined #ipfs
X-Scale has quit [Quit: HydraIRC -> http://www.hydrairc.com <- In tests, 0x09 out of 0x0A l33t h4x0rz prefer it :)]
zuck05 has joined #ipfs
<engdesart>
Any siggestions for my new cactus wood necromancer tower?
<engdesart>
Whoops, wrong channel
<Mateon1>
pawn: Well, you can think of a DAG as a tree with free deduplication, but... It's somewhere between a directed graph and a tree
enzo__ has quit [Ping timeout: 255 seconds]
enzo__ has joined #ipfs
maxlath has joined #ipfs
enzo__ has quit [Ping timeout: 260 seconds]
enzo__ has joined #ipfs
mahloun has joined #ipfs
<_mak>
is there an ipfs command that will let me pin a list of hashes?
mahloun has quit [Client Quit]
<voker57>
_mak: ipfs pin accepts multiple hashes, I believe
<_mak>
but not external file?
<_mak>
I wonder how I can pipe 2 commands to do this
<voker57>
smth like cat file.txt | xargs ipfs pin
mahloun has joined #ipfs
<_mak>
oh nice, thanks
talis has joined #ipfs
mildred1 has quit [Read error: Connection reset by peer]
mildred2 has joined #ipfs
akt has joined #ipfs
espadrine has quit [Ping timeout: 240 seconds]
jungly has joined #ipfs
caladrius has joined #ipfs
<Mateon1>
_mak: `ipfs add` pins by default when you add a file
<Mateon1>
You have to manually specify --pin=false to prevent that
droman has joined #ipfs
pawn has quit [Remote host closed the connection]
kristjansson has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
MaybeDragon has joined #ipfs
tilgovi has quit [Ping timeout: 260 seconds]
mahloun has quit [Quit: WeeChat 1.8]
Muis has joined #ipfs
espadrine has joined #ipfs
rcat has joined #ipfs
Guest80280 has joined #ipfs
mahloun has joined #ipfs
skeuomorf has joined #ipfs
droman has quit []
Encrypt has joined #ipfs
Caterpillar2 has joined #ipfs
Caterpillar has quit [Ping timeout: 240 seconds]
Caterpillar has joined #ipfs
Caterpillar2 has quit [Ping timeout: 260 seconds]
mahloun has quit [Quit: WeeChat 1.8]
cyanobacteria has quit [Quit: Konversation terminated!]
Caterpillar has quit [Ping timeout: 272 seconds]
cyanobacteria has joined #ipfs
cyanobacteria has quit [Remote host closed the connection]
joocain2 has joined #ipfs
cyanobacteria has joined #ipfs
cyanobacteria has quit [Client Quit]
maxlath has quit [Ping timeout: 260 seconds]
<m0ns00n>
whyrusleeping: It's time to start idling at #friendup again :)
m0ns00n has quit [Quit: quit]
mahloun has joined #ipfs
enzo__ has quit [Ping timeout: 240 seconds]
maxlath has joined #ipfs
enzo has joined #ipfs
enzo is now known as Guest46309
Pat00[m] has joined #ipfs
Foxcool has quit [Ping timeout: 240 seconds]
Encrypt has quit [Quit: Quit]
Caterpillar2 has joined #ipfs
Caterpillar has joined #ipfs
Caterpillar2 has quit [Read error: Connection reset by peer]
Caterpillar has quit [Read error: Connection reset by peer]
Caterpillar has joined #ipfs
spacecat has joined #ipfs
Caterpillar2 has joined #ipfs
Caterpillar has quit [Read error: Connection reset by peer]
Guest46309 has quit [Quit: This computer has gone to sleep]
btmsn has joined #ipfs
talis has joined #ipfs
caladrius has quit [Quit: Leaving]
akt has quit [Quit: WeeChat 1.8]
spilotro has quit [Ping timeout: 260 seconds]
droman has joined #ipfs
elkalamar has quit [Quit: Leaving]
charley has joined #ipfs
talis has quit [Ping timeout: 240 seconds]
mahloun has joined #ipfs
charley has quit [Ping timeout: 246 seconds]
akt has joined #ipfs
spilotro has joined #ipfs
aedigix has quit [Quit: disconnecting]
akt has quit [Changing host]
akt has joined #ipfs
brianhoffman has quit [Ping timeout: 240 seconds]
brianhoffman has joined #ipfs
Monokles has quit [Ping timeout: 240 seconds]
aedigix has joined #ipfs
m0ns00n has joined #ipfs
Encrypt has joined #ipfs
Monokles has joined #ipfs
jsgrant has joined #ipfs
owlet has joined #ipfs
dimitarvp has joined #ipfs
espadrine has quit [Ping timeout: 260 seconds]
Guest80280 has quit [Ping timeout: 255 seconds]
Pat00[m] has left #ipfs ["User left"]
larval has joined #ipfs
jsgrant has quit [Ping timeout: 258 seconds]
Boomeran1 has joined #ipfs
Boomerang has quit [Ping timeout: 255 seconds]
arkokoley[m] has joined #ipfs
Guest80280 has joined #ipfs
maxlath has quit [Ping timeout: 260 seconds]
lvernschrock has joined #ipfs
espadrine has joined #ipfs
infinitesum has joined #ipfs
cblgh_ is now known as cblgh
jaboja has joined #ipfs
spacecat has quit [Ping timeout: 255 seconds]
larval has quit [Ping timeout: 260 seconds]
spilotro has quit [Ping timeout: 255 seconds]
<lvernschrock>
when I leave ipfs daemon running, is benefit to the network if I'm not actually hosting anything?
<Mateon1>
lvernschrock: If you have a fast internet connection, you are helping the DHT network
droman has quit []
<Mateon1>
DHT is where IPNS records, provider records and node information is stored
<Mateon1>
So other people can find hashes quicker
<lvernschrock>
Mateon1: okay, that makes sense.
spilotro has joined #ipfs
maxlath has joined #ipfs
ilyaigpetrov has joined #ipfs
<vyzo>
that really depends on whether you are reachable through the network
<vyzo>
if you are NATed and not reachable, you are not really helping much
<vyzo>
although, you can conceivably host some of the data, which helps reachable nodes to which you are connected
maxlath has quit [Ping timeout: 255 seconds]
<lvernschrock>
vyzo: I'm definitely reachable through my NAT because I was able to view on the http://ipfs.io portal content that I had added via the command line
X-Scale has joined #ipfs
espadrine has quit [Ping timeout: 260 seconds]
mahloun has quit [Ping timeout: 246 seconds]
mahloun has joined #ipfs
mildred2 has quit [Ping timeout: 240 seconds]
jungly has quit [Remote host closed the connection]
blame has joined #ipfs
mildred2 has joined #ipfs
<blame>
hello world.
SuprDewd has quit [Quit: No Ping reply in 180 seconds.]
yodata[m] has joined #ipfs
bingus has quit [Ping timeout: 245 seconds]
SuprDewd has joined #ipfs
bingus has joined #ipfs
m0ns00n has quit [Quit: quit]
maxlath has joined #ipfs
mahloun has quit [Ping timeout: 255 seconds]
akt has quit [Quit: WeeChat 1.8]
infinitesum has quit [Quit: infinitesum]
infinitesum has joined #ipfs
infinitesum has quit [Client Quit]
infinitesum has joined #ipfs
espadrine has joined #ipfs
infinitesum has quit [Client Quit]
infinitesum has joined #ipfs
Foxcool has joined #ipfs
<Mateon1>
lvernschrock: Not neccesarily, you very likely connected to the gateways yourself. The ipfs.io gateways are the default bootstrap nodes, after all (`ipfs bootstrap`)
akt has joined #ipfs
<lvernschrock>
Mateon1: I hadn't considered that. Do you know of any portals that isn't a bootstrap node?
<Mateon1>
Hmm
<Mateon1>
I think so, but I need to dig in my history
<lvernschrock>
I suppose that I could just install IPFS on a VPS
<lvernschrock>
and then see if the VPS can see the stuff that my computer at home is supposedly hosting
<vyzo>
in the sense that you might be storing records for the DHT
<vyzo>
which an adjacent node can find from you
<vyzo>
if you are connected to it
<vyzo>
since it can't reach you
<vyzo>
*can reach you
<Mateon1>
Well, storing records doesn't do much if other people can't connect to you
<vyzo>
right, i am saying that it might help nodes you have connected to
<Mateon1>
You will only provide DHT records for those you are connecting to
<vyzo>
since they send the queries to you too
<vyzo>
so it might conceivably reduce latency
<lvernschrock>
I can download my content from https://ipfs.jes.xxx/ipfs/. 'ipfs.jes.xxx/ipfs' resolves to '178.62.243.230' which is not in the list of boostrap notes as far as I can tell
<lvernschrock>
Mateon1: also `ipfs swarm peers` returns over 200 peers
infinitesum has quit [Read error: Connection reset by peer]
infinitesum has joined #ipfs
rozie has joined #ipfs
jaboja has quit [Ping timeout: 240 seconds]
m0ns00n has joined #ipfs
m0ns00n has quit [Quit: quit]
joocain2 has quit [Ping timeout: 248 seconds]
joocain2 has joined #ipfs
jaboja has joined #ipfs
akt has quit [Ping timeout: 260 seconds]
lvernschrock has quit [Ping timeout: 260 seconds]
m0ns00n has joined #ipfs
espadrine has quit [Ping timeout: 255 seconds]
cwahlers_ has joined #ipfs
cwahlers has quit [Ping timeout: 260 seconds]
webdev007 has joined #ipfs
cwahlers has joined #ipfs
cwahlers_ has quit [Ping timeout: 255 seconds]
espadrine_ has joined #ipfs
e0f has quit [Quit: Bye]
e0f has joined #ipfs
jaboja has quit [Remote host closed the connection]
M-Sonata1 has joined #ipfs
gully-foyle has joined #ipfs
Guest46309 has joined #ipfs
nekomune has joined #ipfs
Aranjedeath has joined #ipfs
spacecat has joined #ipfs
ygrek__ has joined #ipfs
Guest46309 has quit [Quit: This computer has gone to sleep]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
joocain2 has quit [Ping timeout: 248 seconds]
omega3[m] has joined #ipfs
elkalamar has joined #ipfs
Encrypt has joined #ipfs
Foxcool has quit [Ping timeout: 240 seconds]
joocain2 has joined #ipfs
Foxcool has joined #ipfs
Encrypt has quit [Quit: Quit]
robattila256 has joined #ipfs
hoboprimate has joined #ipfs
djangod has joined #ipfs
m0ns00n has quit [Quit: quit]
<djangod>
if you want to use ipfs to host listing content for a marketplace application, it would be necessary to purchase computing space with amazon or someone else to host the nodes right?
<djangod>
its not expected for users to run their own node right? That would make it very un accessible for nearly everyone
Boomerang has joined #ipfs
chungy has joined #ipfs
<SchrodingersScat>
I guess that's up to how available you want the files to be. Hypothetically even one slow machine hosting something could get a job done. right?
<djangod>
SchrodingersScat but if its treated as the hosting provider, for lets say 10000 users, the only option is to use a centrilized provider
<djangod>
so it will be a centrilized host like amazon, hosting your virtual decentrilized node
<djangod>
is that correct?
<xelra>
djangod: That is what filecoin will be for. You can host your content in the ipfs-filecoin-cloud then. But it doesn't exist yet.
<MaybeDragon>
sounds like you want filecoin.io ; pay others to host your files?
<djangod>
yea I guess so
<djangod>
I saw a dapp advertising that they will do this, I guess they are waiting for it to be created
<djangod>
So really the only part of dapps that is ready right now is the payment aspect
<blame>
You can also just bypass the whole p2p market thing and use the p2p market of offering to pay people to host your stuff.
<djangod>
and the open source philosophy that comes with it
<djangod>
blame like paying people to be your own miners for your coin?
<blame>
I meant more along the lines of: https://ipfsstore.it/ or just offering to pay in a public forum (like this one)
<blame>
but yes
<xelra>
The problem is that it's not robust enough.
<djangod>
it seems like sooner or later the big guys will come in and host all of it
<djangod>
surprised that every coin is mined by amazon yet
Boomerang has quit [Remote host closed the connection]
<djangod>
maybe the community will limit them, but it seems inevitable with their offer of scale
<blame>
well yes, honestly the only reason bitcoin has avoided that taint is being effectively being money laundering as a service.
<SchrodingersScat>
if they want to host things, let them?
<xelra>
That's the one thing I'm worried about with filecoin too. I mean one guy puts a PB rack in the datacenter and hosts it on 10 GbE with high performance Xeon processors. Invests $100k and the other is running it from home and plugs a few USB drives. The coin definitely needs to take into account those differences.
<M-Sonata1>
yeah, providers are nonfungible
<M-Sonata1>
reputation &c.
<blame>
the only way to prevent something like that is to reduce the scalability of the system in datacenter context without hurting it on small scale. Which we don't have a solution yet.
<djangod>
so you guys are here in #ipfs, what are you using it for now?
<xelra>
blame: Not happening. Going professional is part of lowering the average price.
<M-Sonata1>
well, presumably there'd be some sort of deposit that the hoster loses if they shut down?
<M-Sonata1>
so there's a concrete financial risk to getting too big
<blame>
I don't actually want it to happen. It is the only way to prevent a decentralized system from centralizing.
<M-Sonata1>
i.e. the provider has to shoulder the single-point-of-failure risk
<M-Sonata1>
there's a huge difference between "have 10m downtime and lose 1% of your customers" vs. "have 10m downtime and lose 1% of your customers and 10% of your security deposits"
<xelra>
It will always centralize because of real world requirements. Power, fire departments, cooling, monitoring (real persons) and stuff like that.
<M-Sonata1>
To some extent, sure. But that can be limited by imposing hard incentives on the providers to have redundancy.
droman has quit []
<xelra>
I hope that it actually will be the other way around. That companies like Netflix or Amazon buy filecoins to outsource their storage needs.
<SchrodingersScat>
djangod: I have two nodes that I host random stuff on. ipfs is my pastebinit essentially. but not like the actual ipfs pastebin that someone in here made, I mean I use it for temporary file hosting.
<M-Sonata1>
I expect it'll happen both ways
<M-Sonata1>
big companies cross-sourcing to each other to reduce their downtime risk
<xelra>
I expect Exabyte scale within the first year anyway.
<djangod>
SchrodingersScat do both instances download/upload each change to always have an updated copy on both devices?
<xelra>
When filecoin goes live, I'm planning to put one 4 unit storage server in the datacenter.
<xelra>
I've already been saving. Will cost me $25,000 and §300 per month. I hope traffic won't eat me up.
<xelra>
And hopefully be profitable over a 2 year course.
<djangod>
does each node have to keep an entire copy of all of the files?
<SchrodingersScat>
djangod: not always, sometimes I have junk laying around that I don't want manually synced. I have a script that uses ssh to have one machine match the pins of the other though.
<djangod>
is p2p a replication of the entire data set x times, where x is each user?
<xelra>
djangod: You can only have parts of it too.
<xelra>
It goes down to block level.
<M-Sonata1>
is § filecoin? what do the units mean?
<djangod>
but when you request a file, does it grab a tiny piece from each node, or the entire file from your own node that has been syncing on every change
<xelra>
djangod: It works very much like torrents. It grabs it from everywhere where it can get it.
<djangod>
xelra but if I looked at 1 instance, would it have like 20% of that image, 1% of that thing, ect,
<SchrodingersScat>
djangod: it does appear similar to torrents, each connectable machine that has a part should hypothetically pitch in
<SchrodingersScat>
djangod: now that we've accessed that file, I think it's now on at least one of the ipfs.io gateway machines, so I have one of my machines + 1 ipfs.io machine hosting it until they do their garbage collection
<djangod>
wont this cause a need for everyones devices to step up their resources? If you want to visit a page on your smart phone, you need to be hosting some part of it
<kythyria[m]>
Browsers already cache.
<kythyria[m]>
But yes, IPFS uses more resources than plain HTTPS would.
<kythyria[m]>
That's just a P2P thing.
<M-Sonata1>
probably more upload bandwidth
<xelra>
djangod: Yes, it's not good for devices that run on batteries.
<M-Sonata1>
but very possibly much reduced download
<djangod>
kythyria[m] my concern is with large data sets like, craigslist for example. The amount of content is enormous, and any of it may be accessed at any time
<SchrodingersScat>
djangod: but if your smartphone went through an ISP that used ipfs, then would it decrease their bandwidth from external site to them?
<xelra>
djangod: For mobile, you'd always want to go through some kind of proxy or gateway.
<kythyria[m]>
djangod: If you're careful you can split up the data so you only download parts of it that you need.
<xelra>
That's a good question. You mean if you access something though ipfs that is bigger than all your available local storage. I actually don't know.
<M-Sonata1>
it seems like it should be possible to run an ipfs node in a sort of minimal-networking mode where it leeches the data you want but otherwise declines to participate in the bitswap whatnot
<SchrodingersScat>
would garbage collection take into effect then? but obviously if you need the entire larger set you'd be in tears ;(
<blame>
my view of ipfs is more long term. filecoin is a small thing. It has the facility to be a highly effective p2p distributed local cache that supports ad-hoc isolation and re-connection (and reconciliation!)
<djangod>
xelra right, like when you visit zillow there are millions of bytes of possible content that you could access, based on the filtering you choose
<kythyria[m]>
If you were reading the data from beginning to end I'd hope the start gets pushed out of cache by the time you get to the end.
<xelra>
kythyria[m]: Do you know what happens if I access a 5 TB RAW video file through ipfs, but only have 200 GB local storage?
<SchrodingersScat>
nightmares
jsgrant has joined #ipfs
infinitesum has quit [Quit: infinitesum]
<kythyria[m]>
No idea what it actually does, but I'd expect that it'd try and fill all that space with video.
<djangod>
maybe for the near future, p2p on devices is impossible. We can achieve some hybrid by using amazon-esc hosting environments, with the benefit of encryption and blockchain
<blame>
p2p on devices is perfectly possible. It just needs to be local and not global.
<kythyria[m]>
djangod: P2P on mobile will always suffer compared to putting the full P2P implementation elsewhere and having something much more lightweight on the constrained device.
<blame>
inverse square law sucks.
maxlath has quit [Ping timeout: 260 seconds]
owlet has quit [Quit: Leaving]
<M-Sonata1>
Mmm, MANETs.
<djangod>
yuo can achieve some real decentralization by having a pool of node hosting providers using ipfs, instead of being dependent on a singular one
<kythyria[m]>
(the mobile side wouldn't have to be completely dumb; it could very well be made to have a cache of its own and pin things to it, but would delegate to the cloud side to do some or all of the heavy lifting of maintaining a DHT and the like)
<djangod>
these dapps im looking at all advertise some cool stuff, but after this chat ive realized they are selling technology that is not built yet
<djangod>
thats not really clear when looking at these landing pages asking for an ico
<kythyria[m]>
Or that exists only with massive caveats, yeah
<blame>
well, you real use case of this is robustness not access speed/bandwidth. p2p+decentralized can be really slow, but it has an unmatched level of robustness.
spacecat has quit [Ping timeout: 246 seconds]
<kythyria[m]>
It's particularly annoying when the system in question is already useful, but not living up to the grand vision.
olivernyc has quit [Ping timeout: 255 seconds]
<djangod>
what do you think about the de-centralized philosophy? A foundation, with demographic ETH based voting and open source
<djangod>
the business philosophy that is
<kythyria[m]>
No idea there
<M-Sonata1>
I'm skeptical of voting as an organizational mechanism. Afaict it's a patch for situations where people have been disenfranchised; if you don't have a monopoly on anything – and you shouldn't – then there's nothing to correct.
<M-Sonata1>
Exit rights are necessary and sufficient.
<djangod>
I wonder where you draw the line. What the foundation decides in a dictator or board of directors fashion, and what the community votes on
<M-Sonata1>
As long as each member of the community has both the legal right and the technical capacity to fork the code and tell the foundation to go fuck themselves, the "dictator" can at most issue suggestions.
<djangod>
haha good point
<djangod>
thats unless you offer it as a public protocol
<djangod>
like how linux functions
<blame>
different government mechanisms are systems to solve different problems. No single system is a viable global solution. Even the ability to change systems gracefully is potentially no optimal. A diverse method of governance methods is key.
<mantycore[m]>
In the other news, hetzner just mistook ipfs daemon looking for peers for netscan attempt and asked me to put it down :7
<mantycore[m]>
olivernyc: I was able to create a post, edit my profile
<mantycore[m]>
but search doesn't seem to work
<olivernyc>
Search not implemented yet
<lemmi>
mantycore[m]: hetzner doesn't seem to like the mDNS traffic in particular
<olivernyc>
I'm thinking it's too hard to do a full global search, so the approach I'm going to take is crawling the people you are following up to one or two degrees
<lemmi>
mantycore[m]: you can turn that off
hoboprimate has quit [Quit: hoboprimate]
<olivernyc>
So each node builds up a search index of the network from it's own point of view
<olivernyc>
And you trust the indices of people you follow up to a few degrees
<mantycore[m]>
Cool!
espadrine_ has quit [Ping timeout: 240 seconds]
<mantycore[m]>
Initial test was on chrome (58.0.3029.81); in firefox it doesn't work, but I have rather old firefox build (42)
<mantycore[m]>
I wonder if ipfs works in it at all :)
chris613 has joined #ipfs
Guest80280 has quit [Ping timeout: 260 seconds]
Boomerang has joined #ipfs
Boomerang has quit [Quit: Lost terminal]
<olivernyc>
Works great in firefox and chrome
<olivernyc>
Only safari has issues
Mateon1 has quit [Remote host closed the connection]
maxlath has joined #ipfs
btmsn has quit [Ping timeout: 255 seconds]
m0ns00n has quit [Quit: quit]
olivernyc has quit [Ping timeout: 260 seconds]
maxlath has quit [Quit: maxlath]
owlet has joined #ipfs
Guest80280 has joined #ipfs
owlet has quit [Client Quit]
ilyaigpetrov has quit [Quit: Connection closed for inactivity]
lvernschrock has joined #ipfs
tilgovi has quit [Ping timeout: 246 seconds]
owlet has joined #ipfs
tilgovi has joined #ipfs
webdev007 has quit [Ping timeout: 272 seconds]
grumble has quit [Remote host closed the connection]
grumble has quit [Quit: The evaluation period for grumble IRC bot v2.9.7 (Trial version) has expired. Please contact the sales department of Chatbot Industries International if you wish to keep using grumble.]
grumble has joined #ipfs
infinitesum has joined #ipfs
Caterpillar2 has quit [Quit: You were not made to live as brutes, but to follow virtue and knowledge.]
rcat has quit [Quit: leaving]
infinitesum has quit [Client Quit]
Mateon1 has joined #ipfs
webdev007 has joined #ipfs
webdev007 has quit [Remote host closed the connection]