ShalokShalom_ has quit [Read error: Connection reset by peer]
ShalokShalom has quit [Ping timeout: 246 seconds]
ShalokShalom has joined #ipfs
arcaline1 is now known as arcalinea
galois_dmz has joined #ipfs
wrouesnel has joined #ipfs
galois_dmz has quit [Client Quit]
mbags has quit [Remote host closed the connection]
palkeo has quit [Ping timeout: 240 seconds]
infinity0 has quit [Ping timeout: 240 seconds]
infinity0 has joined #ipfs
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
infinity0 has quit [Remote host closed the connection]
subtrakt_ has joined #ipfs
infinity0 has joined #ipfs
infinity0 has quit [Remote host closed the connection]
subtraktion has quit [Ping timeout: 256 seconds]
infinity0 has joined #ipfs
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]
infinity0 has joined #ipfs
bwn has quit [Ping timeout: 240 seconds]
bwn has joined #ipfs
bwn has quit [Ping timeout: 264 seconds]
skeuomorf has joined #ipfs
shizy has joined #ipfs
subtrakt_ has quit [Remote host closed the connection]
zabirauf_ has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
zabirauf has joined #ipfs
tmg has quit [Ping timeout: 240 seconds]
skeuomorf has quit [Ping timeout: 246 seconds]
galois_dmz has quit []
galois_dmz has joined #ipfs
skeuomorf has joined #ipfs
dimitarvp has quit [Ping timeout: 260 seconds]
bwn has joined #ipfs
jsgrant- has quit [Quit: Peace Peeps. o/ If you need me asap, message me at msg[at]jsgrant.io & I'll try to get back to you within 24 hours.]
jsgrant- has joined #ipfs
Oatmeal has quit [Quit: Suzie says, "TTFNs!"]
zabirauf_ has joined #ipfs
zabirauf has quit [Ping timeout: 264 seconds]
galois_dmz has quit [Remote host closed the connection]
<deltab>
I'm seeing "Path Resolve error: no link named "ipfs" under Qmdf6iGuPoQ7jmGM48nWoy46F5EwMiXYVif968cpwQXDrU" from the gateway for everything, it seems
tilgovi has quit [Remote host closed the connection]
tilgovi has joined #ipfs
tilgovi has quit [Remote host closed the connection]
tilgovi has joined #ipfs
tmg has joined #ipfs
muvlon has quit [Ping timeout: 256 seconds]
tilgovi has quit [Remote host closed the connection]
tilgovi has joined #ipfs
muvlon has joined #ipfs
zabirauf_ has quit [Remote host closed the connection]
zabirauf has joined #ipfs
tilgovi has quit [Ping timeout: 246 seconds]
fleeky has joined #ipfs
zabirauf has quit [Ping timeout: 260 seconds]
tilgovi has joined #ipfs
wallacoloo____ has joined #ipfs
koshii has joined #ipfs
antiantonym has quit [Ping timeout: 246 seconds]
_whitelogger has joined #ipfs
_whitelogger has joined #ipfs
vandemar has quit [Remote host closed the connection]
skeuomorf has quit [Ping timeout: 258 seconds]
Grallae has quit [Remote host closed the connection]
Caterpillar has joined #ipfs
tilgovi has quit [Ping timeout: 246 seconds]
tmg has quit [Ping timeout: 260 seconds]
onabreak has quit [Ping timeout: 260 seconds]
ylp has joined #ipfs
rendar has joined #ipfs
ShalokShalom has quit [Ping timeout: 256 seconds]
Gytha has joined #ipfs
Akaibu has quit [Quit: Connection closed for inactivity]
zabirauf_ has joined #ipfs
zabirauf_ has quit [Ping timeout: 240 seconds]
s_kunk has quit [Ping timeout: 260 seconds]
BanJo[m] has joined #ipfs
<BanJo[m]>
Thanks for kicking me xD
<BanJo[m]>
I just got my new setup, will try out 0.4.7 this weekend :)
kvda has joined #ipfs
gwillen has quit [Ping timeout: 256 seconds]
chrisdickinson has quit [Ping timeout: 260 seconds]
chrisdickinson has joined #ipfs
tcrypt has quit [Ping timeout: 260 seconds]
gwillen has joined #ipfs
tcrypt has joined #ipfs
Gruu has joined #ipfs
s_kunk has joined #ipfs
Gytha has quit [Ping timeout: 246 seconds]
inetic has joined #ipfs
tmg has joined #ipfs
tmg has quit [Ping timeout: 246 seconds]
espadrine_ has joined #ipfs
Mitar has quit [Ping timeout: 240 seconds]
Mitar has joined #ipfs
gmcabrita has joined #ipfs
palkeo has joined #ipfs
wallacoloo____ has quit [Quit: wallacoloo____]
mg_ has quit [Ping timeout: 256 seconds]
maxlath has joined #ipfs
mg_ has joined #ipfs
zabirauf has joined #ipfs
zabirauf has quit [Ping timeout: 240 seconds]
ecloud_wfh is now known as ecloud
jkilpatr has quit [Ping timeout: 246 seconds]
apiarian_mobile has joined #ipfs
subtraktion has joined #ipfs
google77 has joined #ipfs
Foxcool has joined #ipfs
subtraktion has quit [Remote host closed the connection]
subtraktion has joined #ipfs
krzysiekj has quit [Quit: WeeChat 1.5]
krzysiekj has joined #ipfs
jkilpatr has joined #ipfs
maxlath has quit [Ping timeout: 240 seconds]
apiarian_mobile has quit [Quit: boom.]
cemerick has joined #ipfs
apiarian_mobile has joined #ipfs
apiarian_mobile has quit [Quit: boom.]
benthor has quit [Ping timeout: 258 seconds]
null_radix has joined #ipfs
mogria has joined #ipfs
monkhood has quit [Remote host closed the connection]
monkhood has joined #ipfs
sabba[m] has joined #ipfs
Oatmeal has joined #ipfs
dimitarvp has joined #ipfs
dimitarvp has quit [Read error: Connection reset by peer]
maxlath has joined #ipfs
zabirauf has joined #ipfs
galois_dmz has joined #ipfs
zabirauf has quit [Ping timeout: 258 seconds]
dimitarvp has joined #ipfs
kvda has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
krzysiekj has quit [Quit: WeeChat 1.5]
krzysiekj has joined #ipfs
maxlath has quit [Ping timeout: 260 seconds]
maxlath has joined #ipfs
zombu2 has quit [Ping timeout: 240 seconds]
tmg has joined #ipfs
zombu2 has joined #ipfs
maxlath has quit [Ping timeout: 256 seconds]
Akaibu has joined #ipfs
<lgierth>
docs sprint standup in 3 hours (5p UTC)
<Kubuxu>
lgierth: it was moved
<Kubuxu>
sorry
shizy has joined #ipfs
<Kubuxu>
it wasn't.
<lgierth>
:)
<seharder>
What time is the test-lab meeting
<Kubuxu>
seharder: we have to move it as Matt messed up his flight
<Kubuxu>
we will try to do it next week
<Kubuxu>
seharder: could you post on the test lab issue about this
<seharder>
You bet!
<Kubuxu>
as I think some people are also expecting this call
benthor has joined #ipfs
warbo has joined #ipfs
ashark has joined #ipfs
google77 has quit [Quit: Lost terminal]
mogria has quit [Ping timeout: 260 seconds]
subtraktion has quit [Remote host closed the connection]
chris613 has joined #ipfs
mogria has joined #ipfs
Encrypt has joined #ipfs
jsgrant- has joined #ipfs
Boomerang has joined #ipfs
shizy has quit [Quit: WeeChat 1.7]
maxlath has joined #ipfs
zabirauf_ has joined #ipfs
jsgrant has joined #ipfs
zabirauf_ has quit [Ping timeout: 246 seconds]
maxlath has quit [Ping timeout: 240 seconds]
maxlath has joined #ipfs
Gruu has quit [Ping timeout: 260 seconds]
mogria has quit [Quit: Leaving]
JayCarpenter_ has quit [Quit: Page closed]
JayCarpenter has joined #ipfs
tmg has quit [Ping timeout: 256 seconds]
<inetic>
Hello, I'm reading the ipfs.pdf I think I understand most parts (though I did have only skimmed some parts). But the is one claim right at the beginning which kind of baffles me. It is "IPFS could be seen as a single BitTorrent swarm". Could someone please shed some light on what that means?
<inetic>
My understanding of the word swarm (from the BitTorrent's nomenclature) is a set of peers concentrated around a particular DHT key sharing same content.
<r0kk3rz>
yeah thats a major part of IPFS
<r0kk3rz>
all nodes participate in a single DHT
<inetic>
But that is what nodes do in BT as well, no?
<r0kk3rz>
ipfs and bittorrent are similar in many ways
Foxcool has quit [Ping timeout: 260 seconds]
ygrek has joined #ipfs
galois_dmz has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
galois_dmz has quit [Remote host closed the connection]
jaybee has joined #ipfs
jaybee is now known as jb9991
galois_dmz has joined #ipfs
<jb9991>
Hey. I'm sorry to ask what I'm sure is an oft-repeated question, but I found reading through the spec discussions a bit overwhelming: has a consensus been reached on whether fs:/ipfs/ will be used instead of ipfs:/ipfs/ for fully-qualified IPFS URLs?
<inetic>
r0kk3z: True, it's just that in BT terminology this claim doesn't doesn't add up to me. If all the nodes are in a single swarm, then they'd be at only one DHT key, which would render the DHT useless. What am I missing?
galois_dmz has quit [Ping timeout: 246 seconds]
<r0kk3rz>
huh? why should they only be using one key?
<jb9991>
inetic: There's a single DHT network, which is used to find connections to separate per-torrent swarms.
<jb9991>
node = DHT participant, peer = swarm participant, so your use of "nodes in a single swarm" doesn't make sense in BT vocab
<r0kk3rz>
ah i see, yes the terms dont quite map cleanly
<inetic>
r0kk3rz: in bt's terminology, a swarm is a set of nodes sharing the same content, a content is uniquely identified by it's info_hash, thus all peers are at one dht key.
<inetic>
s/all peers/all peers in one swarm/
<r0kk3rz>
yes that makes sense
<inetic>
jb9991: yes, sorry, it should have been "peers in a single swarm". I often mix it up when trying to type fast.
lindybrits has joined #ipfs
<lindybrits>
Hi all! How does one use the ipfs.bootstrap.list() in js-ipfs?
<inetic>
r0kk3rz: thanks, that clear my itch :). But what does the statement mean then?
<r0kk3rz>
inetic: perhaps its not entirely accurate :) another way to look at it is that each IPFS hash is a torrent swarm
<r0kk3rz>
since that is one set of data
<inetic>
r0kk3rz: aha, yes, that would make sense. Would it also make sense to say that one IPFS hash could represent a merkle dag node which then points to other IPFS hashes.
<inetic>
?
ylp has quit [Quit: Leaving.]
<r0kk3rz>
yeah basically
arpu has quit [Remote host closed the connection]
<inetic>
thanks. that clears some things up for me.
mildred1 has quit [Ping timeout: 240 seconds]
arpu has joined #ipfs
maxlath has quit [Ping timeout: 264 seconds]
inetic has quit [Ping timeout: 240 seconds]
synth[m] has joined #ipfs
tilgovi has joined #ipfs
maxlath has joined #ipfs
Boomerang has quit [Quit: Lost terminal]
<lidel>
Hi all, if anyone is interested in getting dnslink to be resolved natively by Firefox, I've added a suggestion to include TXT lookup in proposed DNS WebExtension API at https://bugzilla.mozilla.org/show_bug.cgi?id=1343849#c2
<lindybrits>
@dignifiedquire, can you please check my question above? thanks :)
<dignifiedquire>
lindybrits: js-ipfs or js-ipfs-api ?
<lindybrits>
@dignifiedquire they should, I am implementing them wrongly most likely... can I do the following ipfs.bootstrap.list((result) => { console.log(result); }) giving me the list of peers added to the bootstrap list?
<sprint-helper>
The next event "docs standup" is in 15 minutes.
<dignifiedquire>
I didn't know tht was in multiaddr
<lgierth>
oh yeah right it needs to be added to the the table and all that :D
<dignifiedquire>
and there are also simply many things I don't know yet :)
<lindybrits>
@lgierth, will replacing /tcp with /ipcidr solve my problem then?
<lindybrits>
I want to create a private swarm
<lgierth>
ah!
<lgierth>
check out private networking then
<lindybrits>
@dignifiedquire -> you don't think I can do that yet? no chance of kicking creating a private ipfs swarm here, or another way of doing it possible?
<lgierth>
0.4.7 has been released yesterday and private networks are in there ^
<lindybrits>
ahh thanks, I'll have a look! perfect timing I would say, for me :)
<dignifiedquire>
yeah on go-ipfs that's your best bet
maxlath has quit [Ping timeout: 260 seconds]
<dignifiedquire>
on js it is possible but lots of hand wiring required atm
tilgovi has joined #ipfs
arkimedes has joined #ipfs
<lindybrits>
@dignifiedquire I am figuring out what's what on go-ipfs, ultimately though I need this private network in js-ipfs as well? where can I look to start hand wiring?
<lgierth>
(i didn't know private networking existed in js-ipfs already)
<lgierth>
sorry zoom crashed on me
<dignifiedquire>
there are no private networks in js like the ones shipped in js
<dignifiedquire>
but you can set up your nodes such that they only connect to the one you want them to through various api calls
<dignifiedquire>
I have to run now, cc. daviddias who should also know how to help you
maxlath has joined #ipfs
<daviddias>
hey o/
<daviddias>
catching up
<daviddias>
let me see if I don't miss any question
<daviddias>
- private networks doesn't exist on js-ipfs currently
<daviddias>
- swarm addrs filters also doesn't exist in js-ipfs too. First time someone requests it actually, so it was never a priority, should not be hard to add :)
<lindybrits>
@daviddias... that is very bad news for me. is there a workaround? can I get a private network somehow? I very much need one for our pilot! it will be going live toward November and my PoC for my IPFS piece needs to be done in about a month and a half
<daviddias>
- bootstrap functions are not yet documented, wanna help us? It is straight forward, just need to follow any of the other documented API and do the same (https://github.com/ipfs/interface-ipfs-core/issues/97)
<daviddias>
- private networks exist in go-ipfs though. Which implementation do you want to use?
<lindybrits>
@daviddias, I would need it in js-ipfs. is my understanding correct that if I will be able to remove the unwanted nodes from my bootstrap list as well as add certain peers in as filters, that would complete the private network piece?
<daviddias>
lindybrits: it would get you a 'disjoint network'
<daviddias>
when we mean 'private', we mean, only nodes with a certain cryptographic key can be part of it
<lgierth>
yeah it'll likely work for that, just note it's not "secure"
<daviddias>
but yeah, you can run your 'disjoint network' today, just not connect to any of the public nodes
<lindybrits>
meaning?
Encrypt has quit [Quit: Quit]
<daviddias>
that if you have a 'disjoint network', public nodes won't be able to find files that are in your network
<lindybrits>
I don't quite get what the disjoint network means, and why it will be created when I do as I specified
<lindybrits>
public nodes t
<lindybrits>
public node like all of you guys etc etc?
<lindybrits>
my typing has now been terrible, sorry!
shizy has joined #ipfs
<daviddias>
no problem :)
<lindybrits>
jikes, I am not familiar...
<warbo>
how can I extract/backup the key for an IPNS name, and use that key from another machine?
espadrine_ has quit [Ping timeout: 260 seconds]
<warbo>
I've tried copying the key from my config to a file, and using `ipfs name publish -k filename some-ipfs-hash` and it tells me `Error: key names may not contain slashes`
parazyd has joined #ipfs
<warbo>
seems the keys must be 'installed' to a 'name' somehow?
<lindybrits>
@daviddias -> ultimately I want a couple of servers to talk to each other, securely, as ipfs nodes. I don't want outsiders in my ipfs swarm, it has to be highly secure
<warbo>
(ipns-pub looks useful too, but a) I can't get it to build and b) a github issue about failing builds recommends using 'ipfs name publish -k ...`)
mbags has joined #ipfs
<daviddias>
lindybrits: you can today with js-ipfs:
<daviddias>
- have a couple of nodes talking with each other
<daviddias>
- using secio (this is default) to guarantee that all connections are encrypted
<daviddias>
- not connect to any nodes that you don't control (so no outsiders)
<daviddias>
The only caveat, is that if I happen to discover the address of your super secret nodes if they are reachable 'from the internet', then they will accept my dial to them. With "Private Networks", they would reject the call
<lindybrits>
@daviddias -> hmm. I guess I can just make the servers accept incoming traffic from the ip addresses I trust?
<daviddias>
yep
<daviddias>
that works too
palkeo has quit [Ping timeout: 260 seconds]
<lindybrits>
then we are all good? I can use swarm peer connect
<lindybrits>
even if some random guy tries to connect it will kick him out anyways, because my servers are set up as such?
<lindybrits>
@daviddias, am I on the right track here? :)
<daviddias>
you are :)
s_kunk has quit [Ping timeout: 256 seconds]
<lindybrits>
all is gooood then! :) thanks @daviddias
<daviddias>
you bet :)
<parazyd>
hi
<parazyd>
trying to do a `make install`: /bin/sh: /home/parazyd/go/src/github.com/ipfs/go-ipfs/bin/gx: No such file or directory
<parazyd>
what is the reason for this? gx is in my $PATH ($GOPATH/bin)
<parazyd>
not sure why it's looking for it in this place
<lgierth>
Kubuxu: about gateway and /ws, i'm just wondering if we could end up in a situation where the listener created at startup wouldn't be available anymore, and replaced by a new listener
<Kubuxu>
I don't think so
<lgierth>
that would be insane to debug
<lgierth>
let me pass a WsTptOption to the gateway
<lgierth>
which returns a wstpt.Listener
<Kubuxu>
for sure I don't like the solution of exposing transports, then type casting it to find right listener and so on.
<lgierth>
yeeah that part is hacky
<Kubuxu>
it is the way you get spaghetti from penne
<lgierth>
:D
datz_ has quit [Quit: Leaving]
<lgierth>
well i guess i can detect whether the listener is still usable
DiCE1904 has joined #ipfs
<Kubuxu>
whyrusleeping: now it lost pinroot. qq
<whyrusleeping>
>.>
<Kubuxu>
Ok, I will just move this whole repo to an archive once I get home.
<kevina>
whyrusleeping: would you mind if I take a pass on the pinning code and try to make pinning operations more more robust in general?
<kevina>
I will do it in small steps and explain what I did to make reviewing easier.
<kevina>
The GC should fail if there are pinned objects can't be retrieved, but other operations don't need to fail...
<Kubuxu>
migrating my deidi, we will see how it goes
Encrypt has joined #ipfs
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
rendar has quit [Ping timeout: 264 seconds]
<whyrusleeping>
kevina: hrm... what changes are you thinking?
<warbo>
how can I get the keys (used for `ipfs name publish`) from one machine to another? How can I back them up in a file somewhere?
<Kubuxu>
they are in the config
<Kubuxu>
you can copy whole config over
<kevina>
like when we scan for an indirect pin, don't abort if a pinned object is unavailable, instead issue a warning and continue
<warbo>
Can I "insert" a keypair, for use with `ipfs name publish -k`?
<warbo>
I tried giving -k a file, but it complained :(
<kevina>
and possible if necessary add a "--force" option to pin rm
<kevina>
whyrusleeping: ^
<whyrusleeping>
warbo: you have to use it through the keystore
<whyrusleeping>
warbo: 'ipfs key gen' will create keys and place them in files in $IPFS_PATH/keystore
<warbo>
ah
<warbo>
whyrusleeping: are the original keys (created through init?) stored there?
<warbo>
since my keystore dir is empty
<whyrusleeping>
warbo: unfortunately, we still have the identity keys in the config file
<whyrusleeping>
we're going to be changing this soon
<warbo>
although I did init with 0.4.4 which AFAIK predates the keystore stuff, and today upgraded to 0.4.6
<warbo>
ah ok
<warbo>
I saw the base64 encoded key in there
<warbo>
is there a simple way I can add that, rather than regenerating?
<warbo>
e.g. `base64 -d < key_from_config > $IPFS_PATH/keystore/self`?
ygrek has quit [Ping timeout: 240 seconds]
* warbo
goes to peruse the go-ipfs source :)
<Kubuxu>
warbo: self name for sure won't work
<Kubuxu>
it is reserverd
cemerick has quit [Ping timeout: 246 seconds]
cwahlers_ has quit [Ping timeout: 240 seconds]
cwahlers has joined #ipfs
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
<warbo>
hm, should these key files be owned by my ipfs user or root?
<warbo>
the keystore directory is currently `drwx------ 2 root root 4.0K Mar 17 20:30 keystore`
<warbo>
that doesn't seem right, if the ipfs daemon's meant to have access
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
IRCFrEAK has joined #ipfs
IRCFrEAK has left #ipfs [#ipfs]
antiantonym has joined #ipfs
robattila256 has joined #ipfs
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
realisation has joined #ipfs
<Kubuxu>
you might have initialized it wrong
<Kubuxu>
or run ipfs as root
<Kubuxu>
drwx------ 2 kubuxu kubuxu 6 Dec 14 20:42 keystore
<Kubuxu>
warbo: ^^
IRCFrEAK has joined #ipfs
<A124>
Are there any resoruces for people wanting to use IPFS/libp2p for turn based games?
rendar has joined #ipfs
rendar has joined #ipfs
rendar has quit [Changing host]
IRCFrEAK has quit [K-Lined]
<warbo>
Kubuxu: ah, I did run the datastore migration as root
<warbo>
thanks
zabirauf has joined #ipfs
tmg has joined #ipfs
IRCFrEAK has joined #ipfs
<warbo>
looks like it's worked (using a different name than "self") :)
IRCFrEAK has quit [K-Lined]
zabirauf has quit [Ping timeout: 240 seconds]
antiantonym has quit [Ping timeout: 240 seconds]
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [Excess Flood]
wallacoloo____ has joined #ipfs
atrapado_ has quit [Quit: Leaving]
<Kubuxu>
migrated 700000 keys in just over an hour
<Kubuxu>
lgierth: I have 47 GiB of data one one server (IDK what it is there, it just grew). Do you think we could move it somewhere?
<Kubuxu>
in IPFS
jkilpatr has quit [Ping timeout: 264 seconds]
givemeparttt2000 has joined #ipfs
flyingzumwalt has left #ipfs ["wikidata"]
givemeparttt2000 has quit [K-Lined]
<lgierth>
pollux || biham
Encrypt has quit [Quit: Quit]
givemeparttt2000 has joined #ipfs
givemeparttt2000 has quit [K-Lined]
<Kubuxu>
lgierth: do we plan to ever run gc on those hosts?
<Kubuxu>
aka, do I have to pin it somehow?
<kevina>
whyrusleeping: you never did get back to me, re "pinning operations more more robust"
givemeparttt2000 has joined #ipfs
<lgierth>
Kubuxu: at least they'r exempt from automatic gc
givemeparttt2000 has quit [K-Lined]
kevina has quit [Ping timeout: 264 seconds]
<Kubuxu>
lgierth: ok, transfer is running
lamer14897856317 has joined #ipfs
lamer14897856317 has left #ipfs [#ipfs]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
kevina has joined #ipfs
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
ianopolous has joined #ipfs
tilgovi has joined #ipfs
krzysiekj has quit [Quit: WeeChat 1.5]
krzysiekj has joined #ipfs
MrControll has joined #ipfs
ashark has quit [Ping timeout: 260 seconds]
ashark has joined #ipfs
Caterpillar has quit [Quit: You were not made to live as brutes, but to follow virtue and knowledge.]
ygrek has joined #ipfs
ashark has quit [Ping timeout: 240 seconds]
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
ygrek has quit [Ping timeout: 268 seconds]
ianopolous has quit [Read error: Connection reset by peer]
IRCFrEAK has joined #ipfs
IRCFrEAK has quit [K-Lined]
zabirauf has joined #ipfs
zabirauf has quit [Ping timeout: 264 seconds]
shizy has quit [Ping timeout: 260 seconds]
IRCFrEAK has joined #ipfs
Foxcool has joined #ipfs
IRCFrEAK has quit [Excess Flood]
ianopolous has joined #ipfs
ianopolous has quit [Ping timeout: 240 seconds]
kvda has joined #ipfs
<Ellenor>
what's IPNS
<Ellenor>
nevermind
melvster has joined #ipfs
<melvster>
hi all, just read the ipfs white paper, and watched the talks, seems quite interesting ... im thinking about 'webizing' this system ... one question : can IPFS handle search?
* melvster
waves to mappum :)
<melvster>
typically in a DHT you will partition the key space, can this be done, say, for keyword search ...
<melvster>
or tags ..
<lgierth>
you'd build the search index as an immutable data structure
<lgierth>
then it's completely on ipfs
<lgierth>
or, currently more feasible, attach a traditional indexer like elasticsearch
<melvster>
lgierth: that's an interesting idea, i like it ... do you know of anyone that has worked on search to date?
<lgierth>
this does the latter
<melvster>
ah thanks!
<melvster>
oh yay, thats very cool!
kulelu88 has joined #ipfs
Foxcool has quit [Remote host closed the connection]
<melvster>
so gateway.ipfs.io seems to take a long time from that page, is it because it's overloaded or maybe the file is not found?
<lgierth>
yeah might just not get easily found in the network
Foxcool has joined #ipfs
<lgierth>
who knows, maybe there's simply no seed online right now
<melvster>
k, thanks
<melvster>
fascinating ...
<melvster>
lgierth: do you know is there some place I can find stats about the ipfs network, I'd like to understand how well it is scaling right now ...
IRCFrEAK has joined #ipfs
lothar_m has quit [Quit: WeeChat 1.7-dev]
<melvster>
oh i just found the reddit sub, i can do some searching there ...
damongant has quit [Ping timeout: 258 seconds]
<melvster>
final question: i understand the hashing for IPFS is based on kadmelia ... is there some place I can understand the hashing algorithm better?
damongant has joined #ipfs
<frood>
melvster: routing is based on kademlia. hashing uses multihash, which is just a hash with a prefix denoting the hash algorithm
<frood>
sloppy kademlia, to be specific
<frood>
multihash can be found on github. right now all ipfs hashes start with "Qm" which is the b58 encoded prefix for sha256
<melvster>
oic thanks
<melvster>
base58, i assume with the bitcoin alphabet
<frood>
yeah, non-ambiguous alphanumeric
<frood>
using sha is a departure from Kademlia, which typically has a 160 bit keyspace
<frood>
routing table gets flatter, but that shouldn't really matter.
<melvster>
frood: thanks, so base58( prefix + sha256(content) ) ?
<frood>
pretty much. where the prefix is a short code deisgnating the algorithm and digest length.
tilgovi has quit [Ping timeout: 246 seconds]
<melvster>
ah
<frood>
1 byte for function, 1 byte for digest length.
<frood>
reasoning afaict: ipfs guys are strongly opinionated about data formats.
mildred1 has joined #ipfs
<frood>
less charitable but more funny description: ipfs guys have a bunch of bikes in need of sheds
<MrControll>
I'm a bit new at this, how do I get myself on to IPFD? I'm either too stupid to figure out the guide or it's written for someone WAY more tech savvy then me.
subtraktion has quit [Client Quit]
<frood>
MrControll: what are you trying to do? view webpages on ipfs? add data to ipfs? host a site on ipfs?
<MrControll>
For now let's go with the first one.
<frood>
you can use a gateway, like ipfs.io to view webpages without installing anything.
<MrControll>
Oh.
<melvster>
frood: ty, excellent explanation! :P so one question ... is there any kind of thought surrounding file meta data? lets say I have a picture I might want to tag it with "sunset" to make is searchable ... then I have two files one the picture, and one the meta ... how do i tie them together? does that make sense?
<MrControll>
Ok then :)
<frood>
or you can install the local ipfs daemon, and a browser extension.
Foxcool has quit [Remote host closed the connection]
<frood>
melvster: it's all about the hashlinks. put the multihash of the file into the metadata.
<frood>
or, make a new object that links to both the file and its metadata.
<MrControll>
ty
<melvster>
got it
<melvster>
i have a more novel way of doing this
<frood>
search term for more reading "merkle-dag"
<frood>
and/or "IPLD"
<melvster>
lets say i have a sunset that hashes to Qmsunsetabc
<frood>
which means when information is distributed, search has low utility. or, put another way, centralized search is useful search.
<melvster>
yes that's true
<melvster>
but just something to get the ball rolling ...
<frood>
whyrusleeping and I talked a while back about using HAMTs with well-known entry points to map queries to content lists.
<frood>
but if you rely on a decentralized index like that you'd make a lot of iterative lookups to get where you're going.
<frood>
and you're still relying on whoever made the index to not censor things.
<melvster>
frood: oh excellent, yes I love this idea, I sent an email to the IETF about it last week, but didnt get a reply ... do you have any more thoughts? right now I could only work out a simple system for keyword or tag lookup ...
<melvster>
ie lookup/keyword -> array of hashes
<melvster>
or tag
<melvster>
i guess better SERPs that are ranked
arkimedes has quit [Ping timeout: 260 seconds]
<frood>
yeah, that's the basic construction we were talking about. keyword -> crawl data structure -> list of hashes that match that keyword.
<melvster>
very nice
<frood>
the question in my mind, is if someone has to assemble and maintain these indexes anyway, putting them into a merkle-dag lowers their utility compared to offering a centralized search service.
<frood>
what's the benefit?
<frood>
also, given that ipfs relies on popularity to maintain availability, unpopular indexes may just vanish
<MrControll>
I've actually been wondering about that. Is there a way to have a sort of "permanent node" to keep a site from going down entirely?
<frood>
run your own node and use `ipfs pin Qm...aaa` to maintain your own copy of it
<MrControll>
Ah. good.
<MrControll>
:)
<melvster>
centralized search has weaknesses, if you could combine decentralized indexes, you might be able to get new kind of results