<daviddias>
mafintosh: oh, the 17th? I'm landing in Lisbon on the 16th at noon, I might actually be able to make it! Are people doing the full weekend?
<daviddias>
and thank you for inviting me 😃)
bsm117532 has joined #ipfs
s_kunk has quit [Ping timeout: 260 seconds]
bsm117532 has quit [Ping timeout: 265 seconds]
gabsbo has joined #ipfs
<ianopolous>
whyrusleeping: great work on the memory usage! Very nice to hear.
<whyrusleeping>
ianopolous: don't thank me yet! theres still a long road ahead of us
<whyrusleeping>
but ipfs at least runs on a 1GB vps for more than an hour
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
<ianopolous>
10MiB is great
<whyrusleeping>
would be great
<ianopolous>
oh, did I read wrong?
<whyrusleeping>
probably, lol
<ianopolous>
ah, damn. sorry :-)
<whyrusleeping>
haha
<whyrusleeping>
i've got tweaks that get things down to ~50MB, but its pretty hacky
<whyrusleeping>
and require more updates to be pushed out all over
<ianopolous>
right
<whyrusleeping>
past that, there will be a lot of refining and adding config options like "don't keep a cache here"
<ianopolous>
that will be very useful for us :-)
<whyrusleeping>
yeah, its coming, right now the bottleneck is testing and code review
<ianopolous>
or keep a cache, but only serve it to localhost, not the network
anewuser has joined #ipfs
pawal has quit [Ping timeout: 248 seconds]
gabsbo has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
maxlath has quit [Quit: maxlath]
gabsbo has joined #ipfs
gabsbo has quit [Client Quit]
fleeky_ has joined #ipfs
bsm117532 has joined #ipfs
pawal_ has joined #ipfs
fleeky__ has quit [Ping timeout: 250 seconds]
anonymuse has joined #ipfs
Kane` has quit [Ping timeout: 258 seconds]
<fleeky_>
so im curious is it a best practice to install an ipfs daemon per user or systemwide?
<fleeky_>
seems like its geared more towards installing stuff in /home but im a bit confused on this
Kane` has joined #ipfs
<lgierth>
you can do either
<lgierth>
totally depends on what you wanna do :)
DiCE1904 has quit [Ping timeout: 244 seconds]
Kane` has quit [Remote host closed the connection]
DiCE1904 has joined #ipfs
Kane` has joined #ipfs
<fleeky_>
k just checking
bwerthmann has joined #ipfs
Kane` has quit [Ping timeout: 260 seconds]
geemili has quit [Ping timeout: 260 seconds]
theseb has quit [Ping timeout: 248 seconds]
<fleeky_>
on master branch, running go 1.7.3 i am getting import cycle not allowed , any idea ?
<lgierth>
did you run make deps? (just to make sure)
<mib_kd743naq>
whyrusleeping: submit a PR to change required to option of MIMEType in the metadata .pn, which means I need to regenerate the go source that goes with it
Aranjedeath has quit [Quit: Three sheets to the wind]
sametsisartenep has joined #ipfs
rugu has joined #ipfs
<rugu>
haad: daviddias: Hey, thans for accepting access to the repo. Not sure if you had the time to check it out, but wanted to have a 5-10min discussion with you. Do let me know when you have 10min to spare. Thanks
herzmeister has quit [Ping timeout: 250 seconds]
<rugu>
haad: The repo is a bit of a mess, trying to structure it
herzmeister has joined #ipfs
ribasushi has quit [Ping timeout: 245 seconds]
muvlon_ has quit [Ping timeout: 260 seconds]
ribasushi has joined #ipfs
ralphthe1inja is now known as ralphtheninja
ygrek has quit [Ping timeout: 240 seconds]
dignifiedquire has joined #ipfs
jchevalay has joined #ipfs
silwol[m] is now known as silwol[m]1
silwol[m]1 is now known as silwol[m]2
silwol[m]2 is now known as silwol[m]
silwol[m] is now known as silwol
reit has joined #ipfs
ygrek has joined #ipfs
<kenCode[m]1>
making lots of headway on c-ipfs now..
<jchevalay>
rugu with ids.map( () => {//your works}) and declare an nodes where push your peerNode i think
<jchevalay>
rugu i think will be more clear =)
Kane` has joined #ipfs
<rugu>
jchevalay: ya will check it out. for now just used async.parallel and left it at that. Thanks
<jchevalay>
after that i think the principal topic to check if to create an indepent node
<jchevalay>
for exemple your node need keep an hash table all node which know
maxlath has quit [Ping timeout: 268 seconds]
<jchevalay>
an share that each node where is connect
<rugu>
yes as of now, I have hardcoded it to just ensure communication etc. is working. Then I will work on additional features and making it scalable.
domanic has joined #ipfs
<jchevalay>
rugu ok good luck for your adventure :p
jcmj has joined #ipfs
ulrichard has joined #ipfs
slothbag has quit [Quit: Leaving.]
rugu has quit [Quit: ChatZilla 0.9.93 [Firefox 50.0/20161114144739]]
kaotisk has quit [Quit: Leaving]
elico has quit [Quit: Leaving.]
frmendes has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
nunofmn has joined #ipfs
teaspoon_ has joined #ipfs
anonymuse has joined #ipfs
elico has joined #ipfs
maxlath has joined #ipfs
anonymuse has quit [Remote host closed the connection]
<seharder>
daviddias: Good morning David. I have a question for you if you have time.
brianhoffman_ has joined #ipfs
domanic has quit [Ping timeout: 268 seconds]
<jchevalay>
dignifiedquire
<jchevalay>
dignifiedquire do you see that ?
<jchevalay>
dignifiedquire
<dignifiedquire>
ohh interesting
<dignifiedquire>
thanks
<jchevalay>
yw
<jchevalay>
maybe my resolution ??
<jchevalay>
1680 * 1050
brianhoffman has quit [Ping timeout: 265 seconds]
brianhoffman_ is now known as brianhoffman
<haad>
dignifiedquire: I also like the latter design more for the docs :+1:
<dignifiedquire>
jchevalay: yeah I developed on 1280 resolution that's why haven't seen it yet
<dignifiedquire>
haad: cool :)
<haad>
dignifiedquire: although, the code higlight colors are a little aggressive, perhaps try once more if you can find a more fitting higlight colors.
shizy has quit [Read error: Connection reset by peer]
<dignifiedquire>
aggggreessssivve
<dignifiedquire>
:D
frmendes has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
shizy has joined #ipfs
frmendes has joined #ipfs
bsm117532 has quit [Read error: Connection reset by peer]
bsm117532 has joined #ipfs
<jchevalay>
i like ref to code line on github it's very pretty =)
Qwertie has quit [Read error: error:1408F119:SSL routines:SSL3_GET_RECORD:decryption failed or bad record mac]
Qwertie has joined #ipfs
shizy has quit [Read error: Connection reset by peer]
shizy has joined #ipfs
m0ns00n_ has joined #ipfs
nunofmn_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
m0ns00n_ has quit [Quit: quit]
<dignifiedquire>
haad: jchevalay pushed some changes
taaem has joined #ipfs
<Kubuxu>
collecting coverage in go-ipfs is harder than I thought
<jchevalay>
haad dignifiedquire yes it's good
<haad>
dignifiedquire: that look fanastatic!
<dignifiedquire>
:)
* haad
claps
dcallagh has quit [Remote host closed the connection]
dcallagh has joined #ipfs
<haad>
dignifiedquire: oh, one more, the navigation links on the left, they need a little padding/margin so that they're not as closely together vertically, just a tiny bit
<pinbot>
now pinning /ipfs/QmZseVpMKTaJ2xikMn6zN9oA4umzBkV9PTrdU3StN9xzLw
<pinbot>
[host 2] failed to pin /ipfs/QmZseVpMKTaJ2xikMn6zN9oA4umzBkV9PTrdU3StN9xzLw: cannot store pin state: write /data/ipfs/datastore/369121.log: no space left on device
nunofmn has quit [Quit: Connection closed for inactivity]
maxlath has joined #ipfs
<victorbjelkholm>
\o/
<nicolagreco>
There in a sec
chris613 has joined #ipfs
<richardlitt>
For this call, so we don't flood the channel, we've opened a channel for #ipld
chris_ has joined #ipfs
ebel has quit [Ping timeout: 250 seconds]
<daviddias>
victorbjelkholm: what are you planning to do with the 2 CCC tickets you got? Are you going anyways? //cc wa7son
<daviddias>
seharder: I'm around :)
<daviddias>
what's up?
<chris_>
Hi there IPFS community - good work on everything so far! I have a question re: the security of files stored in IPFS. I understand that in order to access content stored on the network, I need to have the hash (e.g. https://ipfs.io/ipfs/QmWSwsdgzZVJJzyPJRp6JEtU9NqrG3NSi65Q2x75nddqSx), but my question is: is it possible to "sniff" these hashes over the network, and then access content directly (for free), instead of accessing it via a se
<Kubuxu>
yes it is possible to "sniff" them
<daviddias>
chris_: we will have primitives for encrypting files inside IPFS
<daviddias>
right now, files are stored in plaintext
<daviddias>
and by default nodes publish the hashes they have to the DHT
<daviddias>
you can disable the DHT
<daviddias>
and encrypt the content before adding it to IPFS
<whyrusleeping>
though that machine is pretty angry right now
galois_dmz has joined #ipfs
<victorbjelkholm>
daviddias: I'll probably go, haven't decided yet, some other things happening around that needs to be locked down before...
<victorbjelkholm>
I got one spare ticket though!
maxlath has quit [Ping timeout: 248 seconds]
<seharder>
daviddias: I was working on issue 304 in js-ipfs-api and just want to make sure I did things correctly. I cloned js-ipfs-api and then created a branch, made some changes, then pushed the branch back up. I then created a pull request.
<seharder>
daviddias: Is this what you were expecting?
<daviddias>
does github load for you the comments on the thread?
<daviddias>
or are you just seeing the PR comment
<whyrusleeping>
i mean over the http api
<whyrusleeping>
does wb.put(...) happen locally?
<whyrusleeping>
or does it make a request?
nunofmn has joined #ipfs
<daviddias>
The API should support that mode and then we can have smarter clients that save requests by keeping that state themselves
<whyrusleeping>
seems... complicated
<whyrusleeping>
what would the API supporting that mode look like?
<daviddias>
at a first step, we would just need the API to support it
<whyrusleeping>
how would you reference the workbench?
<daviddias>
with a token/id thing
<whyrusleeping>
this is starting to feel like my original mfs proposal, lol
<whyrusleeping>
"tokens are bad, paths are good"
maxlath has joined #ipfs
<whyrusleeping>
although if we let the daemon generate the token it should result in happier things
<daviddias>
yeah, so it is really connected to mfs right? It is the same problem: flush true/false/when?
<whyrusleeping>
yeah
<whyrusleeping>
shits hard, yo
<daviddias>
ahah
<daviddias>
seharder: in order to make that 'version checking' really nice, it might be possible that the best is to finally start using the version at the API in go-ipfs. Because note that even go-ipfs changes the http-api almost with every release. See this thread: https://github.com/ipfs/http-api-spec/issues/116 <- this long has a long beard
<daviddias>
seharder: anyways, I know you have some time and will to contribute and I really should have had sent more tasks in your direction earlier this week. Here is a couple I would love help to get done
<daviddias>
seharder: https://github.com/ipfs/js-ipfs/issues/92 missing cli command. Also checking if every command supports the same options as go-ipfs would be stellar
<daviddias>
seharder: getting the levelDB part of the IPFS Repo done, so that we can access the records and pinset like go-ipfs https://github.com/ipfs/js-ipfs/issues/51 (this will unblock the work done by AdamStone on pinning)
<seharder>
daviddias: So I should abandon issue 304 and start on some of these others?
<daviddias>
seharder: I'm afraid that 304 might be a thing that will take its time. I'll remove the help wanted from that one
<seharder>
daviddias: I will take a look at these other issues and pick one. Thanks for your help.
<daviddias>
whyrusleeping: so, wanna brain dump all the magic tricks you pulled to make mfs happen? I'm liking the `bench` idea to keep threads of mutation
<daviddias>
seharder: thank you for the help! :)
<whyrusleeping>
yeah, the bench idea is pretty much the dageditor thing that ipfs object patch uses
<whyrusleeping>
except we don't allow multiple commands between 'commits'
<whyrusleeping>
i was thinking it would be cool to let object patch take multiple operations
<daviddias>
I got the commits and reverts from wanderer
<whyrusleeping>
but, that reminds me of the first problem we need to look at solving
<whyrusleeping>
what does a 'patch' operation look like in ipld?
<daviddias>
I'm leaning towards "sets the value and bubbles up the changes"
<whyrusleeping>
bubbles how?
<whyrusleeping>
and where does it create intermediate nodes?
<daviddias>
my definition of bubble up (or at least when I say it) is: "I change a value within some hashchain and then I go and update all of the hashes, returning the new root"
<whyrusleeping>
okay
<whyrusleeping>
and when creating new paths, how do you make the decision on where node boundaries go?
<daviddias>
I don't think that patch should be able to create new nodes
<daviddias>
if patch(cid, /a/b/c, 10)
Guest88994 has quit [Ping timeout: 260 seconds]
<daviddias>
if the path doesn't exist as a node or as a path within any of the nodes resolved (as in, in order to set c to 10, I would have to create nodes), patch should fail, simply because that would add another question: "what should be the type of the nodes patch creates"
<whyrusleeping>
hrm...
<whyrusleeping>
how do you create new nodes then?
<daviddias>
ipfs.dag.put(<some instance of node you created>)
<whyrusleeping>
but how do you put that node as a child at some path within an existing dag?
<whyrusleeping>
daviddias: and what if i want it at /a/b/c/d ?
<whyrusleeping>
and maybe c doesnt exist yet?
<daviddias>
(note, I'm giving examples based on what I've built so far, feel free to say "nono, there is a simpler and more nice way, let's do it like X)
<daviddias>
whyrusleeping: I would say, that if you want to do a lot of stiching
<daviddias>
it would seem to me that a new dag is being created
<daviddias>
and if that is the case
<daviddias>
just better create node d, then node c, then node b, then node a and put them all
<whyrusleeping>
hrm, okay
s_kunk has joined #ipfs
<whyrusleeping>
the problem i see is that writing a series of these to accomplish a certain goal gets complicated
<whyrusleeping>
because now we have to add in conditional checks
<whyrusleeping>
because if i'm not sure if 'c' exists in my source dag, but i want to patch something in at /a/b/c/d
<whyrusleeping>
i have to check first if 'c' exists
<whyrusleeping>
then if it doesnt, create it, then go about inserting 'd'
<daviddias>
true
<daviddias>
is there a nice language to do that?
<whyrusleeping>
there is when you have types :P
<whyrusleeping>
'mkdir -P /a/b/c/d'
<whyrusleeping>
it knows that 'c' needs to be a directory
s_kunk has quit [Remote host closed the connection]
<whyrusleeping>
whereas with ipld, 'c' could either be a separate node, or an extra subpath within the parent
Encrypt has quit [Quit: Dinner time!]
s_kunk has joined #ipfs
ylp has quit [Ping timeout: 260 seconds]
<daviddias>
I believe people will be able to create those domain specific interfaces on top of this dag API so that they don't have to repeat themselves over and overf
<daviddias>
but creating a generic foundation without going 10 pages long of documentation might not be possible :D
ygrek has joined #ipfs
<whyrusleeping>
yeah...
<whyrusleeping>
but the thing is, people are going to have to write this for themselves every time they want to use patch
<pinbot>
now pinning /ipfs/QmZV6eoqNv4fkwYZEDCnw12fPuiajHc3JLwMDYta4PYrCY
anonymuse has joined #ipfs
gully-foyle has joined #ipfs
gully-foyle has quit [Client Quit]
Guest88994 has joined #ipfs
teaspoon_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<pinbot>
[host 2] failed to pin /ipfs/QmZV6eoqNv4fkwYZEDCnw12fPuiajHc3JLwMDYta4PYrCY: cannot store pin state: write /data/ipfs/datastore/369121.log: no space left on device
brianhoffman has quit [Quit: brianhoffman]
MadcapJake has quit [Ping timeout: 250 seconds]
Guest88994 has quit [Remote host closed the connection]
<pinbot>
now pinning /ipfs/QmZV6eoqNv4fkwYZEDCnw12fPuiajHc3JLwMDYta4PYrCY
<haad>
again!
<pinbot>
[host 2] failed to pin /ipfs/QmZV6eoqNv4fkwYZEDCnw12fPuiajHc3JLwMDYta4PYrCY: cannot store pin state: write /data/ipfs/datastore/369121.log: no space left on device
<haad>
fine
bwerthmann has quit [Quit: Lost terminal]
<whyrusleeping>
lol
lkcl has quit [Ping timeout: 250 seconds]
lkcl has joined #ipfs
bastianilso_____ has joined #ipfs
<haad>
anyways, that's binaries for early orbit-with-0.4.5. along with fresh go-ipfs, there's some long waited new features like a config screen. see the board: https://waffle.io/haadcode/orbit.
<haad>
I'm out o/ -->
rendar has joined #ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
PrinceOfPeeves has quit [Quit: Leaving]
cemerick has joined #ipfs
maxlath has quit [Remote host closed the connection]
maxlath has joined #ipfs
nunofmn has joined #ipfs
teaspoon_ has joined #ipfs
<fiatjaf>
can I run IPFS without DHT yet?
<Kubuxu>
there is client dht mode
<Kubuxu>
that is experimental
zanadar has quit [Ping timeout: 268 seconds]
prettymuchbryce has quit [Remote host closed the connection]
ebarch has quit [Remote host closed the connection]
lohkey has quit [Remote host closed the connection]
jclay has quit [Remote host closed the connection]
bitspill has quit [Remote host closed the connection]
mappum has quit [Remote host closed the connection]
risk has quit [Remote host closed the connection]
shaunooo has quit [Remote host closed the connection]
bengl has quit [Remote host closed the connection]
bcavileer has quit [Remote host closed the connection]
zielmicha_ has quit [Remote host closed the connection]
jbenet has quit [Remote host closed the connection]
karissa has quit [Remote host closed the connection]
voxelot has quit [Remote host closed the connection]
mbrock has quit [Remote host closed the connection]
manveru has quit [Remote host closed the connection]
Encrypt has joined #ipfs
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
espadrine has quit [Ping timeout: 256 seconds]
<whyrusleeping>
who runs ipfs-search?
<whyrusleeping>
i would love it if they would update their ipfs version, lol