stebalien changed the topic of #ipfs to: Heads Up: To talk, you need to register your nick! Announcements: go-ipfs 0.4.18 and js-ipfs 0.33 are out! Get them from dist.ipfs.io and npm respectively! | Also: #libp2p #ipfs-cluster #filecoin #ipfs-dev | IPFS, the InterPlanetary FileSystem: https://github.com/ipfs/ipfs | Logs: https://view.matrix.org/room/!yhqiEdqNjyPbxtUjzm:matrix.org/ | Forums: https://discuss.ipfs.io | Code of Con
thomasanderson has quit [Ping timeout: 240 seconds]
zeden has quit [Quit: WeeChat 2.3]
zeden has joined #ipfs
aldenp has joined #ipfs
<aldenp>
are there any alternative implementations to go-ipfs? aside from the javascript version and the sketchy C version?
<aldenp>
also why are there so many things like chats and streaming software for IPFS; isn't that idiotic? isn't IPFS designed for *persistent* content, not for live data?
hurikhan77 has quit [Ping timeout: 240 seconds]
<postables[m]>
aldenp: i believe there is a WIP rust implementation. as far as chats that just leverages stuff like libp2p which is the networkign component of IPFS. As far as live streaming, and this moment it's probably impractical, but realistically IPFS can be used as just general data storage, live data is difficult to work with at this time
purisame has joined #ipfs
<aldenp>
well yeah, my understanding was that IPFS wasn't designed for live data at all; my question is why anybody is even trying to make livestreaming services on IPFS, it seems like a bad idea
jhulten has joined #ipfs
hurikhan77 has joined #ipfs
mischat has quit [Remote host closed the connection]
mischat has joined #ipfs
mischat has quit [Ping timeout: 252 seconds]
<Swedneck>
the data sharing aspect would be very nice for livestreams
rcat has quit [Remote host closed the connection]
<Swedneck>
like how peertube makes video sharing not require beefy upload
purisame has quit [Ping timeout: 250 seconds]
thomasanderson has joined #ipfs
thomasanderson has quit [Remote host closed the connection]
thomasanderson has joined #ipfs
thomasanderson has quit [Remote host closed the connection]
<postables[m]>
i wouldn't say its a bad idea. At this point in time IPFS isn't very well suited for live data. One of the solutions is to use IPNS but its too slow at the moment. Furthermore, the tech that lies underneath IPFS can be extremely powerful for live-streaming
<aldenp>
is it really though?
<aldenp>
cuz there's nothing wrong with ipfs only doing one thing and doing it well, instead of throwing in a million extra features just because we can
<aldenp>
if IPFS was only capable of distributing static content, that would be great, and a different service could be created for decentralized livestreaming
<aldenp>
if the issue is that IPFS wasn't designed for it and it's only now getting lumped in, that seems like a very bad idea
<aldenp>
you mentioned libp2p; perhaps someone could go off and create an entirely different service using libp2p; not everything has to be IPFS
<aldenp>
idk, I don't know exactly how IPFS works under the hood, but at a glance it seems a little sketchy
<Swedneck>
i don't quite think it's right to say it's not designed for it
<Swedneck>
after all, once the livestream data is sent, it's static
thomasanderson has joined #ipfs
<Swedneck>
what isn't static is the current head of the content, it's like an IPNS address being updated extremely quickly
<aldenp>
the trouble is that livestream data is created on the fly and involves a lot of really small files, which isn't ideal for publishing; it would make more sense for streaming software to stream using an alternate protocol designed purely with speed in mind, and then re-encode the stream into a single video file and publish it to IPFS
<Swedneck>
i think that's just a limitation of people trying to use it for livestreaming
<Swedneck>
ipfs already splits files into small chunks, i see no reason it would be unable to handle livestream data if formatted correctly
user_51 has quit [Read error: Connection reset by peer]
user_51 has joined #ipfs
<postables[m]>
that's where IPNS comes in, publish all your data under an IPNS record and you'll be able to rebuild your stream. But IPNS is slow right now so it would probably need to wait a bit before you can reasonably implement streaming via ipfs
<Swedneck>
couldn't direct pubsub be used?
<postables[m]>
`idk, I don't know exactly how IPFS works under the hood, but at a glance it seems a little sketchy` i don't see how its sketchy 🤷 IPFS is a big and complex beast in some aspects, but its awesome
<postables[m]>
swedneck: actually yea it probably could be used 🤔
<aldenp>
I mean splitting things up into small pieces is space inefficient because you end up with a lot of duplicate headers & such, so unless IPFS is splitting things up into 10 byte pieces lol...
<TimMc>
AIUI, streaming is generally accomplished by streaming a text-based manifest file that points to successive chunks of the stream data. IPFS can no doubt handle the chunks, but you can't content-address the streaming manifest.
<aldenp>
streams use more bandwidth and computer power to play back than a properly encoded file
<aldenp>
idk, I don't watch a lot of livestreams, but it depends on the usual usage; if users are frequently seeking back in the video, then maybe IPFS makes sense; however, if users are almost entirely just watching it live, it would be more efficient to just directly transport the video as it comes in to viewers, and upload a re-encoded video file to IPFS after the stream for the future people
<TimMc>
You could certainly stream a list of IPFS addresses, though, and I bet there's a way to use IPNS differently to say "yes, this is the authoritative stream".
thomasanderson has quit [Remote host closed the connection]
<postables[m]>
`I mean splitting things up into small pieces is space inefficient because you end up with a lot of duplicate headers & such, so unless IPFS is splitting things up into 10 byte pieces lol...` by default its 256KB chunks, i would really recommend that you read about IPFS in depth, it's extremely easy to misjudge technology without understanding the fundamentals of it
<postables[m]>
nor can one even reach a valid conclusion about something without understanding it
<postables[m]>
*edit:* ~~nor can one even reach a valid conclusion about something without understanding it~~ -> nor can one even reach a valid conclusion about something without understanding it. You can als configure chunking parameters yourself, although it will result in different hashes
<postables[m]>
IPFS is extremely flexible, and very modular. So if hte default implementation doesn't work, there's probably a way you can make changes to various pieces of the puzzle to do what you want
cwahlers has joined #ipfs
cwahlers_ has quit [Ping timeout: 258 seconds]
aarshkshah1992 has joined #ipfs
<TimMc>
The more relevant question is what chunk size you *want* in the stream.
<TimMc>
For streaming video, 256 kB seems fine as a chunk size.
<TimMc>
Bad idea: Use dnslink with a short TTL.
Steverman has quit [Ping timeout: 250 seconds]
jhulten has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<TimMc>
Is there a standard interface that pinning services speak? Or plans for one?
aldenp has quit [Ping timeout: 250 seconds]
aldenp has joined #ipfs
thomasanderson has joined #ipfs
aldenp has quit [Read error: Connection reset by peer]
nihilarchy has joined #ipfs
<postables[m]>
what do you mean by that? im not really sure what standard interface they would need to satisfy
thomasanderson has quit [Ping timeout: 272 seconds]
purisame has joined #ipfs
silotis has quit [Ping timeout: 250 seconds]
silotis has joined #ipfs
<TimMc>
On second thought, I'm not really sure either. :-/
<postables[m]>
I mean as long as you just provide access to IPFS through a pinning service it should be fine right? I run a pinning service, and while im not necessarily opposed to the idea of a standard interface, I'm unsure as to the benefits. Although I suppose a set of standard "methods" that can be used would be nice to have for building out applications on other platforms like ethereum which would use IPFS
<postables[m]>
although for that you might just be better off having a smart contract with a standard interface (ie, `Pin` `PubSub`) which is listened to by an IPFS pinning service
thomasanderson has quit [Remote host closed the connection]
silotis has quit [Ping timeout: 250 seconds]
silotis has joined #ipfs
thomasanderson has joined #ipfs
thomasanderson has quit [Ping timeout: 250 seconds]
user_51 has quit [Ping timeout: 268 seconds]
alexgr has quit [Ping timeout: 252 seconds]
thomasanderson has joined #ipfs
user_51 has joined #ipfs
zopsi_ has joined #ipfs
zopsi has quit [Read error: Connection reset by peer]
}ls{ has quit [Ping timeout: 252 seconds]
thomasanderson has quit [Remote host closed the connection]
}ls{ has joined #ipfs
cratliff has joined #ipfs
zeden has quit [Quit: WeeChat 2.3]
cwahlers has quit [Ping timeout: 250 seconds]
cwahlers has joined #ipfs
lassulus_ has joined #ipfs
lassulus has quit [Ping timeout: 246 seconds]
lassulus_ is now known as lassulus
cratliff has quit [Ping timeout: 258 seconds]
thomasanderson has joined #ipfs
thomasanderson has quit [Ping timeout: 250 seconds]
user_51 has quit [Ping timeout: 240 seconds]
x2 has quit [Quit: Leaving]
user_51 has joined #ipfs
nonono has quit [Ping timeout: 272 seconds]
nihilarchy has quit [Remote host closed the connection]
Belkaar has quit [Ping timeout: 258 seconds]
Belkaar has joined #ipfs
Belkaar has joined #ipfs
stoopkid has quit [Quit: Connection closed for inactivity]
<TimMc>
That makes sense.
erratic has quit [Ping timeout: 246 seconds]
Fessus has quit [Remote host closed the connection]
Fessus has joined #ipfs
dimitarvp has joined #ipfs
b7219264_ has quit [Read error: Connection reset by peer]
b7219264 has joined #ipfs
<stebalien>
Dear bikeshed: What would you name a flag that means "please auto-upgrade CIDv0 CIDs to CIDv1 CIDs in output so that they can be properly formatted using multibase". Current contenders are things like `--force-multibase-output`, `--force-multibase-cids`, `--upgrade-cids`, `--upgrade-cids-in-output`, `--force-upgrade-v0-cids-to-v1-in-output`, `--output-cidv1`, `--force-upgrade-cidv0-in-output`, ...
<stebalien>
Sticking points: This only applies to output, this only applies to CIDs.
Onesiphorus has quit [Quit: leaving]
jamiew has quit [Quit: ZNC 1.6.6+deb1ubuntu0.1 - http://znc.in]
jamiew has joined #ipfs
xcm has quit [Remote host closed the connection]
xcm has joined #ipfs
cwahlers_ has joined #ipfs
cwahlers has quit [Ping timeout: 245 seconds]
aarshkshah1992 has quit [Ping timeout: 250 seconds]
aarshkshah1992 has joined #ipfs
JohnWick has left #ipfs [#ipfs]
reapposition has joined #ipfs
aarshkshah1992 has quit [Ping timeout: 245 seconds]
spinza has quit [Quit: Coyote finally caught up with me...]
aarshkshah1992 has joined #ipfs
FlyingJanitor[m] has joined #ipfs
spinza has joined #ipfs
wking has joined #ipfs
appa_ has quit [Ping timeout: 272 seconds]
}ls{ has quit [Quit: real life interrupt]
<postables[m]>
has there been any investigation into embedding a tensorflow graph into an IPLD object?
Pulse2496 has joined #ipfs
erratic has joined #ipfs
Ai9zO5AP has quit [Read error: Connection reset by peer]
i9zO5AP has joined #ipfs
Ai9zO5AP has joined #ipfs
can has quit [Quit: ZNC 1.6.6+deb1ubuntu0.1 - http://znc.in]
can has joined #ipfs
i9zO5AP has quit [Ping timeout: 272 seconds]
hsn has joined #ipfs
ikari` has joined #ipfs
lidel` has joined #ipfs
lidel has quit [Ping timeout: 246 seconds]
lidel` is now known as lidel
achingbrain has quit [Read error: Connection reset by peer]
tperson has quit [Read error: Connection reset by peer]
achingbrain has joined #ipfs
tperson has joined #ipfs
johnhmay has quit [Read error: Connection reset by peer]
johnhmay has joined #ipfs
erratic has quit [Ping timeout: 272 seconds]
thomasanderson has joined #ipfs
cwahlers has joined #ipfs
cwahlers_ has quit [Ping timeout: 250 seconds]
thomasanderson has quit [Ping timeout: 250 seconds]
ylp has joined #ipfs
Mateon3 has joined #ipfs
Mateon1 has quit [Ping timeout: 252 seconds]
Mateon3 is now known as Mateon1
cwahlers has quit [Ping timeout: 250 seconds]
erratic has joined #ipfs
cwahlers has joined #ipfs
BeerHall has joined #ipfs
erratic has quit [Ping timeout: 250 seconds]
BeerHall has quit [Client Quit]
BeerHall has joined #ipfs
hsn has quit [Quit: hsn]
florianH_ has joined #ipfs
<ylp>
does any one tried to run ipfs on a Exynos5422 cpu ?
<ylp>
I guess it should work since it's an ARM but if someone has a first hand experience let me know
<r0kk3rz>
ive run it on snapdragon arm processors
<r0kk3rz>
worth a shot on exynos
Mottengrotte has joined #ipfs
<lidel>
stebalien, I'd go with `--upgrade-cids-in-output` or `--normalize-cids-in-output`
spinza has quit [Quit: Coyote finally caught up with me...]
N30N has joined #ipfs
Tranquil has quit [Ping timeout: 250 seconds]
aarshkshah1992 has quit [Remote host closed the connection]
Caterpillar has joined #ipfs
Steverman has joined #ipfs
spinza has joined #ipfs
rendar has joined #ipfs
aarshkshah1992 has joined #ipfs
MCFX2 has joined #ipfs
MCFX2_ has quit [Ping timeout: 250 seconds]
cwahlers_ has joined #ipfs
cwahlers has quit [Ping timeout: 246 seconds]
Tranquil has joined #ipfs
spinza has quit [Quit: Coyote finally caught up with me...]
N30N has quit [Ping timeout: 246 seconds]
Tranquil has quit [Read error: Connection reset by peer]
Tranquil has joined #ipfs
Mateon3 has joined #ipfs
Mateon1 has quit [Ping timeout: 250 seconds]
Mateon3 is now known as Mateon1
spinza has joined #ipfs
clemo has joined #ipfs
Tranquil has quit [Read error: Connection reset by peer]
Tranquil has joined #ipfs
<aarshkshah1992>
Hi guys, I see we have a js API for using IPLD as seen in the examples here
<aarshkshah1992>
I see what you mean. So the go-ipfs implementation depends on `go-ipld-git,eth etc etc`.
<aarshkshah1992>
If I want to write a piece of code to put some 'Git' blocks into IPFS, can I do that by using `go-ipfs` as a library
<aarshkshah1992>
Or do I 'have' to go via the go-ipfs-api that you shared ?
<aarshkshah1992>
If you know what I mean
aarshkshah1992 has quit [Remote host closed the connection]
<Magik6k>
You don't have to use go-ipfs-api, there are a few different ways to do this (although, I'd say using go-ipfs-api is the recommended one for now). If you have access to 'raw' git blocks you should use `ipfs block put`
<Magik6k>
Using go-ipfs as a library is fairly hard now, so I wouldn't go that route
florianH_ has quit [Quit: Connection closed for inactivity]
Papa_Alpaka has joined #ipfs
detran has quit [Ping timeout: 250 seconds]
cratliff has quit [Ping timeout: 272 seconds]
spinza has quit [Quit: Coyote finally caught up with me...]
zane has joined #ipfs
spinza has joined #ipfs
jhulten has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
cwahlers_ has joined #ipfs
jhulten has joined #ipfs
cwahlers has quit [Ping timeout: 246 seconds]
MDude has quit [Ping timeout: 250 seconds]
zane has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
jhulten has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
jhulten has joined #ipfs
cwahlers has joined #ipfs
cwahlers_ has quit [Ping timeout: 272 seconds]
<aldenp>
when I use `ipfs pin add` to pin a directory, does it try to download everything in that dir before the command exits? (pinning a 200GB dir right now and I don't really want to download 200GB on my crappy internet connection atm)
mischat has quit [Remote host closed the connection]
mischat has joined #ipfs
MDude has joined #ipfs
ikari` has quit [Quit: This computer has gone to sleep]
jhulten has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
jhulten has joined #ipfs
cratliff has joined #ipfs
<pedrohlc>
I'm in a network where DHT is blocked (doesn't work at least for torrents), should ipfs work?