vijayee__ has quit [Read error: Connection reset by peer]
Matoro_ has quit [Ping timeout: 255 seconds]
stoogenmeyer has joined #ipfs
Matoro_ has joined #ipfs
shea256 has quit [Remote host closed the connection]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
shea256 has joined #ipfs
joshbuddy has quit [Quit: joshbuddy]
reit has quit [Quit: Leaving]
linagee has joined #ipfs
stoogenmeyer has quit [Remote host closed the connection]
stoogenmeyer has joined #ipfs
<linagee>
can someone please help me understand IPFS? Should I make an app that's dependant upon gateway.ipfs.io existing? what funds it? (will it go down?) should I run my own IPFS node? (but then what is the point of IPFS?)
<linagee>
or is there a good frontend that will choose through several known public IPFS gateways?
stoogenmeyer has quit [Remote host closed the connection]
<Protocol_under>
hi <linagee> im not IPFS dev or even a guru but: IMO you should not make an app that's dependant of gateway.ipfs.io fro the simple reason that you don't own it, and don't have controll over it.
<Protocol_under>
i think you should run your own IPFS node
<Protocol_under>
gateway.ipfs.io make you access IPFS content via normal webserver, If you want to actualy use IPFS you need your own node
water_resistant has joined #ipfs
Matoro_ has quit [Ping timeout: 244 seconds]
<lgierth>
if you're gonna go with the public gateway, just use ipfs.io
<lgierth>
it's the same as gateway.ipfs.io (which is deprecated)
<lgierth>
i'd make it configurable
<lgierth>
use ipfs.io default, but make it possible to use another gateway
<linagee>
Protocol_under: let's say I'm making a mobile app and I want to store some resources on IPFS. Any reason to do it that way as opposed to just HTTP serving?
<linagee>
lgierth: ah, configurable makes sense.
elima_ has joined #ipfs
<lgierth>
you might also wanna use the api instead
<Protocol_under>
mobile, idk, IPFS il like torrenting, not shure users will like the bill they will recive.
<lgierth>
:5001/api is the full api, mostly local-only, and :8080/api is a very limited api, currently readonly, but will soon allow posting objects and files
<lgierth>
e.g. ipfs.io/api/v0/version
<linagee>
Protocol_under: If I was running a node on mobile, yes. If I am just using HTTP to access what needs to be accessed...
<lgierth>
it will probably still be useful to run your own gateways somewhere
<lgierth>
so that you have control over them
<Protocol_under>
^
<lgierth>
and can e.g. pin stuff that people have posted
<lgierth>
otherwise it'll get removed on the next gc cycle
e-lima has quit [Ping timeout: 240 seconds]
<linagee>
how long does content on ipfs last? :-/
<linagee>
<-- ipfs n00b
<lgierth>
garbage collection removes everything that's not pinned
<Protocol_under>
pinned is like bookmarked
<lgierth>
yeah bookmarked is a good metaphor actually
<bcavileer>
linagee if you host a node geographically close to your clients that may improve the bandwidth and latency to your mobile clients
<bcavileer>
if that is possible
<bcavileer>
I'm a n00b too though =)
<linagee>
assuming inside such an app I knew which node to prefer. (probably using some complex DNS or geoIP magic.)
<lgierth>
the ideal case is running a very thin ipfs node mobile
<lgierth>
so that you're not *relying* on connectivity
<linagee>
lgierth: mobile = people love crappy computers
stoogenmeyer has joined #ipfs
<linagee>
:)
<lgierth>
and can still peer with nodes over wifi or bluetooth
<lgierth>
ok probably not bluetooth haha
<lgierth>
but AP wifi and tethering
<Protocol_under>
lol
<lgierth>
which is very powerful
<bcavileer>
that's my question: given that IPFS isn't baked into our applications yet... can a CDN or DNS find the "closest" gateway?
<lgierth>
you only have to go the subway or country-side in $richNation to lose connectivity
<lgierth>
amazon route 53 can
<bcavileer>
if I ran 30 gateways spread across the country, could I give my clients an http://... that would resolve to their closest gateway?
fleeky has quit [Ping timeout: 240 seconds]
<bcavileer>
ahh ok
<linagee>
lgierth: yep. amazon route 53 can probably handle DNS closest finding node magic.
<bcavileer>
sweet thanks lgierth ... come to IPFS to learn IPFS... learn about DNS =)
<lgierth>
not probably, it's *the* selling point for it -- latency based routing
<linagee>
bcavileer: DNS is the fabric that binds the world together. :-D Wait no, that's BGP I'm thinking of. Wait no, most people don't care about that level of networking. ;-)
<Protocol_under>
kek
<Protocol_under>
DNS is just a GUI, pros don't use it :p
<linagee>
lgierth: couldn't someone write a bot to request from nodes a certain hash key to keep it recent and from being garbage collected?
<lgierth>
we have pinbot which pins on ipfs.io -- it's just broken right now but about to fixed this week
<The_8472>
i think it would be nice to eventually have some sort of "cache random stuff you see on the network" behavior to create some modicum of plausible deniability.
<lgierth>
!botsnack
<pinbot>
om nom nom
<The_8472>
hiding in the noise
<M-mubot>
Thanks for the treat!
<lgierth>
The_8472: you can do that on top of ipfs
<lgierth>
get a list of nodes, query what they provide, fetch some of it
<lgierth>
ideally you'd do it cooperatively of course
<The_8472>
alternatively a way where all the content is encrypted on disk and in flight and addressed with a lower-level hash than the actual hash used by the client, that way you can store things without knowing what you're storing
<lgierth>
The_8472: i *think* that's gonna be in the 0.4.0 release (cc whyrusleeping)
<lgierth>
ah no sorry
* whyrusleeping
sees a notification
<lgierth>
i was thinking about encrypted swarms
<lgierth>
not encrypted objects
<Protocol_under>
by common ownership of private key ?
<lgierth>
yeah
* lgierth
back to work
<linagee>
The_8472: re: cache random stuff: wouldn't you have just created freenet at that point?
<The_8472>
incorporated a feature of freenet, sure
<linagee>
then it would become: "What is IPFS?" ah, it's just like Freenet, but... different somehow. because maybe the original developers never saw freenet?
<bcavileer>
The_8472: if someone graciously ran https://github.com/palkeo/ipfspy on their node, you could just download everything it output
<bcavileer>
that would obfuscate some traffic and disk
<bcavileer>
maybe
<The_8472>
linagee, I was thinking of opt-in behavior
<linagee>
I'm still trying to figure out how ipfs/freenet/etc compare to storj/maid/etc. "free decentralized storage" vs "paid"
fleeky has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to master: https://git.io/v2e6u
<The_8472>
as it is right now nodes only store what they retrieve or add yourself. so you have to incentivize other nodes to store something for you. either because they also have an interest in the data, because they're altruistic or because you pay them to do so
<linagee>
they have a very flashy webpage, but when I compared prices to the lowest tiers of Amazon's offering it's like... wait a sec... these are around the same price?
<bcavileer>
how can altruistic nodes get a list of content others want pinned?
<linagee>
(Surely decentralized storage has some advantages over Amazon, but still... they are touting "we're the cheapest" on their site.)
<linagee>
MetaDisk (spingoff using StorJ) costs $0.015/GB/mo.
<ipfsbot>
[go-ipfs] whyrusleeping pushed 2 new commits to master: https://git.io/v2e6S
<ipfsbot>
go-ipfs/master 2c4de53 Stephen Whitmore: Fixes range error by using > 0 length buffer....
<The_8472>
the gateways are not obliged to keep your things around forever
<bcavileer>
or for the life of the hdd?
<The_8472>
bcavileer, you pay per month
<bcavileer>
oh right duh
<bcavileer>
it's late and has been a long day
<linagee>
bcavileer: my latest analysis of newegg stuff is: $0.023 per GB for spinny drive, $0.23 per GB for SSD drives.
<ipfsbot>
[go-ipfs] whyrusleeping deleted docs/poll-default-false at 9c53ad7: https://git.io/v2eiT
<linagee>
bcavileer: lol, SSD is still a magnitude of cost cheaper than SSD. (surely SSD still has it's place with DB stuff. just interesting for bulk storage.)
<linagee>
ack. spinny is a magnitude of cost cheaper than SSD
<linagee>
bcavileer: at those prices: you'd break even on physical hardware with about 3-4 months of usage. (not assuming any RAID extra cost with storage, controller costs, server costs, datacenter costs, etc. etc.)
<bcavileer>
haha yeah I figured that... cloud storage is profitable
<bcavileer>
well they have other expenses, but yeah... $/GB/mo not for life
<ipfsbot>
[go-ipfs] whyrusleeping closed pull request #2363: Removed double spaces, added `an`, siwitched carrot to dollar (master...docs/resolve) https://git.io/vgj5b
<linagee>
Also, Glacier is "weird". "give us at least 4 hours for a retrieval of the data". Makes me wonder what they are doing. putting it on tape? Putting it on spun down drives?
<bcavileer>
and I thought my git commit messages were boring ;)
<ipfsbot>
[go-ipfs] whyrusleeping deleted feature/add-level-arg-desc at b14403a: https://git.io/v2eiE
stoogenmeyer has quit [Remote host closed the connection]
shea256 has quit [Remote host closed the connection]
<Evermore>
linagee: I think that's the idea of Glacier
<Evermore>
It's cheap because the drives aren't necessarily even hooked up
<Evermore>
coldest possible storage
<brimstone>
there's rumor that Glacier is backed by bluray drives
<linagee>
Evermore: hrm... no cost for servers and controllers... drives on shelves in some warehouse stacked up
Oatmeal has quit [Ping timeout: 265 seconds]
<Evermore>
blu-ray just doesn't seem that dense but I guess if the disks are cheap
<linagee>
Evermore: what if it was blu-ray and you did something that was not conducive to their business model? Throw GB out there at a time, delete objects, and keep it only for a little while.
<linagee>
Oh wait... unless they have a very intelligent caching system that keeps it on spinning disks until they determine you're probably not going to access it for a while...
<Evermore>
without ever retrieving it?
<Evermore>
yeah that's probably how they do it
<Evermore>
so by the time it hits long-term storage, you've bought the discs
<linagee>
Yes, because if blu-ray costs $2/GB or something, once it's burned the space is used. If you threw stuff at them and constantly deleted it... :-D
<Evermore>
$2 / GB sounds high for blu-ray
<linagee>
er sorry, 2 cents per GB
<Evermore>
oh
<linagee>
IRC text is immutable, lol
<Evermore>
I thank the spaghetti monster every day that IRC isn't Slack
<linagee>
I'm a long time IRC-er, before 2000. Slack makes me a bit nautious.
<Evermore>
me too
<linagee>
Like, "WTF? are they trying to re-invent IRC? Why?"
<Evermore>
If it was free and not centralized I would maybe get behind it
<linagee>
true. that aspect bothers me as well.
<Evermore>
But it's part of the web-only concept where I need a big fatass computer with JavaScript, and it doesn't even run on hardware I own
<Evermore>
My local hackerspace uses it cause they're all losers
<linagee>
if you love the web aspect of Slack, there are web clients for IRC, lol....
<linagee>
(although it seems like you're a second class citizen on IRC with those.)
<Evermore>
I tried to use Slack's IRC gateway and the editing doesn't even work so it's unusable if anyone edits anything
ipfsrocks has joined #ipfs
<linagee>
Ahhh!!! the worst part of slack: invite only. If that wasn't a thing, it might be semi-tolerable.
<linagee>
(Unless you pay of course, hahaha.)
<ipfsbot>
[go-ipfs] RichardLitt pushed 1 new commit to docs/edit-ls-desc: https://git.io/v2ePy
<ipfsbot>
go-ipfs/docs/edit-ls-desc 63fbc7b Richard Littauer: Added other entries...
hellertime has joined #ipfs
<Evermore>
When I come on here after programming at work I feel like a small fish in a big pond again... There's actually a feature I want in IPFS but I have no idea how to even discuss it
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed feat/reqlog-time from 3c8207d to 4ad2b4a: https://git.io/vgd23
<ipfsbot>
go-ipfs/feat/reqlog-time 4ad2b4a Jeromy: add command to change keep time for reqlog objects...
<linagee>
interesting product development concept: Glacier is just S3 that waits. Then they implement something cheaper later.
<linagee>
Do the simplest thing possible to accomplish the goal, hah.
<Evermore>
The idea of abstracting things so much is fascinating
<linagee>
it's kind of annoying from a developer aspect.
<Evermore>
I am debating buying some online backup space to dump encrypted volumes in, in case my house burns down or something
<linagee>
will the API be around for years? how much time will it take me to learn it? is it very annoying?
<Evermore>
yeah
<Evermore>
dunno
<linagee>
Evermore: how will you store the backup key in such a way that it will survive a house burning down? do you have to keep that method itself secret? heh...
<Evermore>
linagee: good question
<Evermore>
Could just be a password
<Evermore>
All I need is to deter Amazon
<linagee>
I'd imagine you'd have to keep the method itself secret, but then it means nobody else will ever know.
<Evermore>
Good crypto works even if you know everything but the key / passphrase
<Evermore>
I would probably use LUKS
<Evermore>
It's reliable and well-supported and I've already practiced on it
<Protocol_under>
LUKS do the job
<Evermore>
only question is how to do incremental updates if at all
<Evermore>
I actually thought encfs might be nicer but I don't like it generally
<linagee>
if you have to store the key in an offsite location because it's too complex to remember by memory (as it should be so it can't be bruteforced) why not just also physically store drives at that offsite location?
<Evermore>
cause updating the drives is harder
<linagee>
true... you have a point. :-D
<Evermore>
Amazon has a good internet connection but if my drives are at my parent's house or at my bank
rtlong has joined #ipfs
<linagee>
but if they are online on Amazon, maybe Amazon can be hacked and they can be deleted or someone can steal your password and they can be deleted. (argument of offline versus online storage)
<Evermore>
true
<Evermore>
But to really delete them, they'd have to burn my house down at the same time
shea256 has joined #ipfs
<linagee>
there is no solution. it's all a bunch of effort you have to take out of your life and essentially throw away. arghhhh! lol.
<Evermore>
There's levels
<Evermore>
You just have to make it worth your while and not worth your opponent's while
<Protocol_under>
wyou will have ti implant an SSD under your skin
<linagee>
Protocol_under: if its cheap and small enough why not? lol..
<Evermore>
it would be uncomfortable
<linagee>
Protocol_under: like maybe a micro SD card
<linagee>
ah, also you'd be giving up your storage mechanism that should probably be kept secret and not logged like this channel is, hahaha.
<Protocol_under>
misroSD are small and cheap, Implant can be done by any local piercing body shop
<Evermore>
like I said, you only need to protect the key
<Protocol_under>
welcome to the futur :)
<linagee>
Protocol_under: you'd be on the same level of weird as the people who get RFID tags embedded in their skin. lol.
shea256 has quit [Ping timeout: 248 seconds]
<Protocol_under>
lol
<linagee>
Evermore: do I need to pay to incentivise people to keep backups of my encrypted data? couldn't I just store it all on IPFS and constantly ping nodes to keep fresh copies of it?
<Evermore>
linagee: I'm not pinning your stuff without a bribe
<Evermore>
If you wanna back up public stuff that I also want to have, then yeah, but then it wouldn't be encrypted
<linagee>
Evermore: you won't pin my terabytes of encrypted crap data? lol.
<ipfsbot>
[go-ipfs] RichardLitt created docs/fix-tagline (+1 new commit): https://git.io/v2e16
<ipfsbot>
go-ipfs/docs/fix-tagline 3da4419 Richard Littauer: Updated tagline for queryDhtCmd...
<ipfsbot>
[go-ipfs] RichardLitt opened pull request #2372: Updated tagline for queryDhtCmd (master...docs/fix-tagline) https://git.io/v2e1i
Oatmeal has joined #ipfs
computerfreak has quit [Remote host closed the connection]
linagee has quit [Quit: Lost terminal]
stoogenmeyer has joined #ipfs
devbug has quit [Ping timeout: 250 seconds]
Oatmeal has quit [Ping timeout: 244 seconds]
Matoro_ has joined #ipfs
dignifiedquire has quit [Quit: Connection closed for inactivity]
<ipfsbot>
[go-ipfs] chriscool created remove-old-unused-tests (+2 new commits): https://git.io/v2ey6
<ipfsbot>
go-ipfs/remove-old-unused-tests fa57fb7 Christian Couder: sharness: remove x0045-add-cat-iptb.sh...
<ipfsbot>
go-ipfs/remove-old-unused-tests ad71b15 Christian Couder: sharness: remove xt0130-multinode.sh...
<ipfsbot>
[go-ipfs] chriscool opened pull request #2373: Remove old unused tests (master...remove-old-unused-tests) https://git.io/v2eyi
ipfsrocks has quit [Ping timeout: 255 seconds]
voxelot has joined #ipfs
shea256 has joined #ipfs
Oatmeal has joined #ipfs
zorglub27 has quit [Quit: zorglub27]
shea256 has quit [Ping timeout: 255 seconds]
guruvan has quit [Remote host closed the connection]
BananaLotus has quit [Remote host closed the connection]
BananaLotus has joined #ipfs
guruvan has joined #ipfs
hellertime has quit [Quit: Leaving.]
shea256 has joined #ipfs
<daviddias>
voxelot: pong :)
<voxelot>
daviddias: heyo! did you follow that convo up there with dig?
<voxelot>
i have bufferImporter working exactly the same way with the same tests as fileImporter with the "/test-data/1.2MiB.txt" file
shea256 has quit [Ping timeout: 240 seconds]
<voxelot>
does that cover all of the cases we need for a buffer importer?
NDuma has joined #ipfs
<voxelot>
and i had to change the way fixedSizeChunker handles chunks of data that are bigger than the CHUNK_SIZE
joshbuddy has joined #ipfs
mkelly has joined #ipfs
rhalff has quit [Quit: Leaving]
stoogenmeyer has quit [Remote host closed the connection]
<daviddias>
are you using the same size chunks as the rest?
<daviddias>
voxelot: importing a 'big buffer' and a big buffer that has non scalar size from the chunking
<voxelot>
daviddias: haha exactly, it has something to do with the other test cases never hit fsc with the case of a chunk being bigger than the chunk size i think
<voxelot>
yeah so the big file has non-scalar size
<voxelot>
so i'me testing the worse case with that right
shea256 has quit [Ping timeout: 250 seconds]
nuun is now known as evening
pfraze has quit [Remote host closed the connection]
<daviddias>
<daviddias>
:D
<voxelot>
haha have a moment for streams now?
<voxelot>
so that should be pretty easy right? just need a function that takes a stream as the arg instead of path or buffer right
TheWhisper has quit [Read error: Connection reset by peer]
<daviddias>
voxelot: yep yep :)
<daviddias>
did you find a problem for the streams case as well?
TheWhisper has joined #ipfs
<voxelot>
nono just haven't started it yet
<voxelot>
now does that have to cover if the stream is streaming a directory?
amstocker has joined #ipfs
jaboja has quit [Ping timeout: 240 seconds]
<voxelot>
i guess it would just be like the file source
Arakela has joined #ipfs
<voxelot>
an incoming stream could be anything
<daviddias>
voxelot: that would require that we accept some kind of encoding for dirs
<daviddias>
(like multipart \o/)
<daviddias>
but not what we are trying to achieve right now (but hey.. I'm also not stopping you and that might be useful at somepoint)
<voxelot>
haha yeah that's an idea, i'll keep it open
<str1ngs>
too many links client.go:247 is there a limit to how many files can be in a directory?
cellvia has quit [Quit: Page closed]
ylp1 has joined #ipfs
<_chad_>
str1ngs: i believe this is an underlying OS thing. EMLINK error.
disgusting_wall has quit [Quit: Connection closed for inactivity]
hpk has left #ipfs ["WeeChat 1.1.1"]
s_kunk has quit [Ping timeout: 240 seconds]
shea256 has joined #ipfs
_chad_ has quit [Remote host closed the connection]
tmg has joined #ipfs
shea256 has quit [Ping timeout: 240 seconds]
zootella has joined #ipfs
reit has quit [Ping timeout: 255 seconds]
zootella has left #ipfs [#ipfs]
<M-zick>
who know why I get ipfsApi.exceptions.ipfsApiError: Could not resolve name. ?
Akaibu has quit [Quit: Connection closed for inactivity]
amstocker_ has joined #ipfs
amstocker has quit [Ping timeout: 240 seconds]
pfraze has joined #ipfs
warner` has joined #ipfs
dignifiedquire has joined #ipfs
warner has quit [Ping timeout: 240 seconds]
pfraze has quit [Ping timeout: 248 seconds]
warner` is now known as warner
devbug has joined #ipfs
shea256 has joined #ipfs
ipfsrocks has joined #ipfs
NDuma has quit [Ping timeout: 252 seconds]
shea256 has quit [Ping timeout: 240 seconds]
ipfsrocks has quit [Ping timeout: 244 seconds]
amstocker_ has quit [Ping timeout: 240 seconds]
Tv` has quit [Quit: Connection closed for inactivity]
amstocker_ has joined #ipfs
<daviddias>
M-zick: in what context?
<M-zick>
daviddias (IRC) I want to resolve address in python-ipfs-api
<M-zick>
Traceback (most recent call last):
<M-zick>
File "./get_list.py", line 89, in <module>
<M-zick>
main()
<M-zick>
File "./get_list.py", line 86, in main
<daviddias>
M-zick: ah, the python client. are you just trying to connect? (the python api lib might be out of date)
<M-zick>
print add_to_list(api, args.bulletin)
<M-zick>
File "./get_list.py", line 39, in add_to_list
<M-zick>
path = api.name_resolve(peer_id)['Path']
<M-zick>
File "/d/evox/env/dd7df51d477e0e97340f42c8609b49da45775ca8.160217.224709/local/lib/python2.7/site-packages/ipfsApi/client.py", line 306, in name_resolve
<M-zick>
File "/d/evox/env/dd7df51d477e0e97340f42c8609b49da45775ca8.160217.224709/local/lib/python2.7/site-packages/ipfsApi/commands.py", line 31, in request
<M-zick>
File "/d/evox/env/dd7df51d477e0e97340f42c8609b49da45775ca8.160217.224709/local/lib/python2.7/site-packages/ipfsApi/http.py", line 24, in wrapper
<M-zick>
return f(self, *args, **merged)
<M-zick>
File "/d/evox/env/dd7df51d477e0e97340f42c8609b49da45775ca8.160217.224709/local/lib/python2.7/site-packages/ipfsApi/http.py", line 91, in request
<M-zick>
raise ipfsApiError(ret['Message'])
<M-zick>
ipfsApi.exceptions.ipfsApiError: Could not resolve name.
<okket>
M-zick: first day in IRC?
<M-zick>
daviddias (IRC): I want to get hash from ipns
<M-zick>
okket (IRC): no
<daviddias>
M-zick: mind opening an issue on the repo? the lib needs to be updated anyway
<okket>
M-zick: then you should know that is a very bad habit to paste more then 1-2 lines into IRC, please use some kind of pastebin for multi line stuff
<M-zick>
okket (IRC): ok:) thanks)
<M-zick>
daviddias (IRC): mm, this working on one PC, but on production does not works:(
yellowsir has joined #ipfs
ugjka has joined #ipfs
s_kunk has joined #ipfs
martinkl_ has joined #ipfs
amstocker_ has quit [Ping timeout: 240 seconds]
chad has joined #ipfs
IlanGodik has joined #ipfs
joshbuddy has joined #ipfs
shea256 has joined #ipfs
voxelot has quit [Ping timeout: 250 seconds]
shea256 has quit [Ping timeout: 240 seconds]
reit has joined #ipfs
C-Keen has quit [Changing host]
C-Keen has joined #ipfs
chad has quit [Remote host closed the connection]
pfraze has joined #ipfs
pfraze has quit [Ping timeout: 244 seconds]
shea256 has joined #ipfs
shea256 has quit [Ping timeout: 240 seconds]
ipfsrocks has joined #ipfs
ipfsrocks has quit [Ping timeout: 240 seconds]
devbug has quit [Ping timeout: 240 seconds]
kalmi has joined #ipfs
kalmi has quit [Ping timeout: 240 seconds]
dansmith_btc is now known as dansmith__
dansmith__ is now known as grandmaster
computerfreak has joined #ipfs
_chad_ has joined #ipfs
_chad_ is now known as chad__
<ipfsbot>
[js-ipfs-api] dignifiedquire created greenkeeper-karma-mocha-reporter-1.2.0 (+1 new commit): https://git.io/v2ffX
<ipfsbot>
js-ipfs-api/greenkeeper-karma-mocha-reporter-1.2.0 9374705 greenkeeperio-bot: chore(package): update karma-mocha-reporter to version 1.2.0...
mildred1 has joined #ipfs
<ipfsbot>
[js-ipfs-api] dignifiedquire deleted greenkeeper-karma-mocha-reporter-1.2.0 at 9374705: https://git.io/v2fJC
<haad>
re. twtxt, we'll have fully ipfs-powered microblogging as soon as pubsub is in ipfs (plus a couple of weeks)
<haad>
:)
<NeoTeo>
haad: Cool, thx. So the ability to create multple ipfs keys and associate them with an ipns hash is there so I guess/hope integration is coming soon.
<NeoTeo>
whyrusleeping: and it's pretty much the same for ipfs-key
<whyrusleeping>
NeoTeo: what platform?
<noffle>
richardlitt: o/
<NeoTeo>
os x
Aeon has quit [Read error: Connection reset by peer]
<noffle>
jbenet: I don't see a strong conclusion on where 'size' from protobuf ipfs objects fits into ipld (just a bit of mention on https://github.com/ipfs/go-ipld/issues/14#issuecomment-151091648). some examples mention size as a link property, but is this going to be the canonical way?
<whyrusleeping>
NeoTeo: and youre sure that you have go1.5.2?
<richardlitt>
noffle: I listened to 'Wheat Kings' like 40 times last night
<NeoTeo>
That's what my go version says
<noffle>
jbenet: we need to figure out what this means re e.g. progress bars on 'ipfs get'
<NeoTeo>
whyrusleeping: go version go1.5.2 darwin/amd64
ipfsrocks has quit [Ping timeout: 240 seconds]
<kseistrup>
i just installed "ipfs version 0.4.0-dev" and now ipfs can't seem to find any peers.
<whyrusleeping>
NeoTeo: can you delete $GOPATH/src/gx/ipfs/Qme1box... and rerun make install?
<NeoTeo>
whyrusleeping: one mo
<noffle>
richardlitt: rock on garth
<richardlitt>
I bring it up because you are Canadian
<richardlitt>
And therefore love all things Canada
<richardlitt>
Also, Rush are awesome.
<whyrusleeping>
why why zed?
<NeoTeo>
whyrusleeping: MAGIC \o/
Aeon has joined #ipfs
Aeon has joined #ipfs
<whyrusleeping>
NeoTeo: woo!
<NeoTeo>
So I take it it got corrupted?
<whyrusleeping>
yeah, i need to add some checks for that i guess :)
<whyrusleeping>
thanks for reporting it
Proto_type has quit [Ping timeout: 276 seconds]
<NeoTeo>
whyrusleeping: thanks for fixing it!
<whyrusleeping>
NeoTeo: not a problem
computerfreak has joined #ipfs
<NeoTeo>
time to play with ipns-pub. Is it going into ipfs proper "real soon now" or...?
<whyrusleeping>
NeoTeo: not *real* soon, but maybe within a few week
<whyrusleeping>
s
MrChrisJ has joined #ipfs
<NeoTeo>
whyrusleeping: Cool. I've just been in the twtxt channel and (possibly) sold the guys on using/integrating ipfs. That and pubsub would be really good for that.
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #249: Update chai-enzyme to version 0.4.1
<whyrusleeping>
MrChrisJ: Probably, i'm not sure how to build electron on arm though, have to ask dignifiedquire
<MrChrisJ>
Many thanks whyrusleeping, I am about to launch a new version of the Fullnode OS with the new version of Jessie and Bitcoin 0.12 so it would be nice to get the latest IPFS has to offer too
<whyrusleeping>
MrChrisJ: cool :) timeline on that launch?
<MrChrisJ>
I committed to end of March whyrusleeping
joshbuddy has joined #ipfs
<MrChrisJ>
I didn’t promise this though so I think the people can be flexible
<MrChrisJ>
I am also working on bash scripts that install stuff automatically
<whyrusleeping>
MrChrisJ: okay, cool. version 0.4.0 is merged into master, but not officially shipped quite yet
<MrChrisJ>
so we can do remote updates
calavera has joined #ipfs
<MrChrisJ>
ok cool, what’s the timeline on 0.4? Is it ‘soon’TM?
<whyrusleeping>
'soon'TM
<MrChrisJ>
:)
<MrChrisJ>
ok cool
<whyrusleeping>
its essentially in validation right now
<MrChrisJ>
but now there is an update command in 3.9 anyway right?
<whyrusleeping>
remaining barriers to shipping it are just lots of testing
<whyrusleeping>
MrChrisJ: yeap!
<MrChrisJ>
wicked, thanks
<noffle>
whyrusleeping: enjoying your make-an-issue-for-every-travis-failure crusade; I think it's helping
<whyrusleeping>
noffle: :D
<Kubuxu>
Previously there was one big issue but nobody knew what got fixed and what not.
patcon_ has joined #ipfs
<whyrusleeping>
yeah, i think this is a better way to get everything fixed
<whyrusleeping>
mega-issues suck
<Kubuxu>
They do.
<daviddias>
hi MrChrisJ ! :)
<MrChrisJ>
hey there daviddias
<MrChrisJ>
:)
<noffle>
many issues + label aggregation works pretty well
<noffle>
as long as the # of labels doesn't get out of control
<noffle>
and everyone is diligent in labeling
patcon has quit [Ping timeout: 248 seconds]
<daviddias>
yeah, as whyrusleeping said (which you also met at CCC, your helper during the live demo), we are 'close' to launch 0.4.0, it should be on time for your launch
<MrChrisJ>
nice work!
<MrChrisJ>
Do you know how I can get Station working on the RPi?
<dignifiedquire>
MrChrisJ: Pushed a fix (hopefully) for npm@2 to station#master
patcon_ has quit [Ping timeout: 244 seconds]
martinkl_ has joined #ipfs
kanzure has joined #ipfs
kanzure has quit [Changing host]
Matoro has quit [Ping timeout: 255 seconds]
patcon_ has joined #ipfs
<Evermore>
Is there a recommended way to organize or index files in IPFS? Say I have a bunch of movies and I want to delete them to clear space, is there a program that can remember which hashes are movies and how big they are?
joshbuddy has quit [Quit: joshbuddy]
shea256 has joined #ipfs
<lovelaced>
i would also be interested in the answer to this ^
<Evermore>
Particularly if I want to have several client programs sharing an IPFS instance, and some of them need to unpin something but maybe another program wants to keep it, like reference-counting
patcon__ has quit [Ping timeout: 244 seconds]
<Evermore>
I am yet to hear of an overwhelmingly good solution for local file search at all, but it may be easier to develop on IPFS where everything is already hashed and files don't 'move'
patcon__ has joined #ipfs
shea256 has quit [Ping timeout: 255 seconds]
shea256 has joined #ipfs
joshbuddy has joined #ipfs
joshbuddy has quit [Client Quit]
stoogenmeyer has quit [Remote host closed the connection]
patcon_ has quit [Ping timeout: 250 seconds]
martinkl_ has quit [Ping timeout: 240 seconds]
nicolagreco_ has quit [Quit: nicolagreco_]
stoogenmeyer has joined #ipfs
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #250: Update eslint-plugin-react to version 4.0.0
<Evermore>
So far IPFS works really great if you don't need to delete files
shea256 has quit [Ping timeout: 252 seconds]
patcon_ has joined #ipfs
nicolagreco_ has joined #ipfs
martinkl_ has joined #ipfs
Matoro has joined #ipfs
patcon__ has quit [Ping timeout: 240 seconds]
Encrypt has joined #ipfs
<NeoTeo>
Evermore: Not suggesting this as a final feature but I have written a wrapper around my ipfs add that logs the hash with the file name. In the spirit of UNIX the next step would be to write another small tool that looked in that file for the file name you were after and returned its hash.
felixn_ has joined #ipfs
<Evermore>
NeoTeo: That sounds like a good start. To make the searching faster maybe it could be dumped into a sqlite DB or something. From there arbitrary stuff could be added
shea256 has joined #ipfs
<Evermore>
Ideally there would be a way to run it in an "update" mode where it could scan IPFS for unindexed files so even if something got added externally it would be picked up
<Evermore>
More ideally the database should also exist in IPFS but I'm not sure how to do that
felixn_ has quit [Read error: Connection reset by peer]
<lovelaced>
I think they want to do some metadata thing
<lovelaced>
not sure on the status of that though
felixn has quit [Ping timeout: 252 seconds]
martinkl_ has quit [Ping timeout: 250 seconds]
<Evermore>
I like that the core design has no metadata
<Evermore>
So when I upload to ipfs.pics the picture doesn't have my IP address or filename in it (unless ipfs.pics added that which I doubt)
<lovelaced>
right
<lovelaced>
there's a git-annex thing about adding metadata but I'm not sure I understand it
patcon__ has joined #ipfs
<Evermore>
I would like to have a music player where the songs, playlists, and music database are all IPFS objects but so far I've been all talk on that front
<Evermore>
bbl
martinkl_ has joined #ipfs
martinkl_ has quit [Max SendQ exceeded]
martinkl_ has joined #ipfs
felixn has joined #ipfs
patcon_ has quit [Ping timeout: 240 seconds]
shea256_ has joined #ipfs
shea256_ has quit [Read error: Connection reset by peer]
shea256_ has joined #ipfs
patcon_ has joined #ipfs
shea256 has quit [Ping timeout: 248 seconds]
nicolagreco_ has quit [Quit: nicolagreco_]
patcon__ has quit [Ping timeout: 252 seconds]
shea256_ has quit [Remote host closed the connection]
patcon has joined #ipfs
trock has quit [Remote host closed the connection]
<ipfsbot>
[js-ipfs] diasdavid created feature/swarm (+1 new commit): https://git.io/v2JMR
<ipfsbot>
js-ipfs/feature/swarm b8b20c6 David Dias: start
patcon_ has quit [Ping timeout: 276 seconds]
yellowsir has quit [Quit: Leaving.]
martinkl_ has quit [Ping timeout: 252 seconds]
patcon_ has joined #ipfs
Senj has quit [Ping timeout: 255 seconds]
rendar has quit [Ping timeout: 276 seconds]
<noffle>
Evermore: you could do this today /w regular players by storing your music on ipfs, mounting /ipfs, and giving the player /ipfs/Qmfoobar/mysong.mp3 paths
felixn has quit [Ping timeout: 252 seconds]
patcon has quit [Ping timeout: 255 seconds]
patcon__ has joined #ipfs
rendar has joined #ipfs
shea256 has joined #ipfs
shea256 has quit [Remote host closed the connection]
Encrypt has quit [Quit: Quitte]
patcon_ has quit [Ping timeout: 255 seconds]
shea256 has joined #ipfs
patcon has joined #ipfs
se3000 has joined #ipfs
Guest25 has joined #ipfs
se3000 has quit [Client Quit]
Guest25 has quit [Client Quit]
patcon__ has quit [Ping timeout: 244 seconds]
martinkl_ has joined #ipfs
felixn has joined #ipfs
Matoro has quit [Ping timeout: 255 seconds]
martinkl_ has quit [Ping timeout: 244 seconds]
palkeo has joined #ipfs
palkeo has joined #ipfs
<Kubuxu>
It would be awesome because you wouldn't need even playlists, just join files in directories and add numbers in begging, all this without storage overhead.
buckket has joined #ipfs
Senji has joined #ipfs
zoobab has quit [Ping timeout: 276 seconds]
felixn has quit [Ping timeout: 252 seconds]
felixn has joined #ipfs
ipfsrocks has quit [Ping timeout: 240 seconds]
zoobab has joined #ipfs
calavera has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
Matoro has joined #ipfs
stoogenmeyer has quit [Remote host closed the connection]
patcon_ has joined #ipfs
patcon has quit [Ping timeout: 240 seconds]
patcon__ has joined #ipfs
patcon_ has quit [Ping timeout: 252 seconds]
patcon_ has joined #ipfs
calavera has joined #ipfs
nicolagreco_ has joined #ipfs
patcon__ has quit [Ping timeout: 250 seconds]
stoogenmeyer has joined #ipfs
elima_ has quit [Ping timeout: 240 seconds]
patcon__ has joined #ipfs
computerfreak has quit [Remote host closed the connection]
patcon has joined #ipfs
patcon_ has quit [Ping timeout: 244 seconds]
hellertime has joined #ipfs
patcon_ has joined #ipfs
devbug has joined #ipfs
patcon__ has quit [Ping timeout: 250 seconds]
sobr has joined #ipfs
patcon__ has joined #ipfs
patcon has quit [Ping timeout: 276 seconds]
patcon_ has quit [Ping timeout: 255 seconds]
patcon_ has joined #ipfs
elima_ has joined #ipfs
hellertime has quit [Read error: Connection timed out]
stoogenmeyer has quit []
hellertime has joined #ipfs
patcon__ has quit [Ping timeout: 244 seconds]
nicolagreco_ has quit [Quit: nicolagreco_]
<Evermore>
noffle: Yeah but I figured if I target the HTTP server then I could also have it work on computers with no IPFS or no FUSE, possibly even through a gateway
<Evermore>
the directory-as-playlist sounds interesting
patcon__ has joined #ipfs
ylp has quit [Quit: Leaving.]
patcon_ has quit [Ping timeout: 255 seconds]
patcon_ has joined #ipfs
patcon__ has quit [Ping timeout: 255 seconds]
amstocker has joined #ipfs
zoobab has quit [Ping timeout: 276 seconds]
screensaver has joined #ipfs
patcon__ has joined #ipfs
zoobab has joined #ipfs
patcon_ has quit [Ping timeout: 248 seconds]
cemerick has quit [Ping timeout: 255 seconds]
patcon has joined #ipfs
patcon_ has joined #ipfs
patcon__ has quit [Ping timeout: 250 seconds]
patcon__ has joined #ipfs
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #251: Update eslint to version 2.2.0
<ipfsrocks>
Hi looking for some advice about distributing an ipfs web app
<ipfsrocks>
It uses the http api so I'm not sure how to do this best
<ipfsrocks>
I can't use https because the apit is http only
patcon_ has joined #ipfs
<Evermore>
IPFS web app?
<ipfsrocks>
Yeah it's a normal js html webapp but it uses the ipfs http api
<Evermore>
For what?
<ipfsrocks>
Adding files, publishing and resolving ipns
<Evermore>
Does the IPFS server run on your computer or the users'
<Evermore>
?
<ipfsrocks>
users need a local ipfs node
<Evermore>
Doesn't that mean the JS also must be loaded from their computer?
patcon has quit [Ping timeout: 276 seconds]
<ipfsrocks>
Yes but the problem is the ipfs api access control
patcon__ has joined #ipfs
<ipfsrocks>
I can allow access to my http domain but then I'm vulnerable to mitm attack on my ipfs node
<ipfsrocks>
I can allow access to the local http gateway but then I'm vulnerable to all sites accessed through it
<ipfsrocks>
Seems like there are not good options, unless I am missing something obvious..
zoobab has quit [Ping timeout: 276 seconds]
<Evermore>
I don't understand what you are trying to do
patcon_ has quit [Ping timeout: 276 seconds]
patcon_ has joined #ipfs
<yellowsir>
do you only want to access the files you added on ipfs in your api/gateway?
<ipfsrocks>
No I need full admin control of the node
patcon__ has quit [Ping timeout: 255 seconds]
<ipfsrocks>
js-ipfs-api
amstocker has quit [Ping timeout: 240 seconds]
<ipfsrocks>
The readme recommends allowing my domain access but only http is supported
patcon__ has joined #ipfs
<yellowsir>
you could put it behind nginx for https i guess
<ipfsrocks>
If I put it behind https then I can't make http requests to the api
zoobab has joined #ipfs
nicolagreco_ has joined #ipfs
bigbluehat has quit [Ping timeout: 240 seconds]
<ipfsrocks>
Seems like there are no good ways to distribute js-ipfs-api webapps.. we are either running untrusted code or allowing ipfs api access to untrusted code.