<Kythyria[m]>
So, you want it to only work correctly in browsers that support recursive suborigins?
<Kythyria[m]>
Under the way browsers presently work, IPFS sure as heck has an "authority".
<Kythyria[m]>
Alternatively, everything on IPFS is one website.
<Fess>
there's more than one gateway
<Kythyria[m]>
According to the current design of web browsers, the `dweb` style URLs place the whole of IPFS in one website.
<Kythyria[m]>
(more precisely, one origin)
<Fess>
ah right
<Kythyria[m]>
Also, WTF is a "nestable URI"?
<Kythyria[m]>
Also, what's the difference between `/ipfs/<hash>` and `ipfs://hash`other than an extra colon?
FimFx has joined #ipfs
<Kythyria[m]>
... would it, under the "Unix paths are the One True Syntax" thing, be reasonable to stubbornly insist on writing things like `/https/gateway.ipfs.io/ipfs/<hash>`?
<Fess>
no idea, the second format you listed doesn't resolve for me when running the daemon
<Kythyria[m]>
Well, the thread cipres linked is full of claims that the former is somehow inherently superior.
<Kythyria[m]>
Also, "an IPFS path is a relative URL" only makes sense if you assume the whole of IPFS has one fixed authority.
lupi has joined #ipfs
lupi has quit [Max SendQ exceeded]
lupi has joined #ipfs
zautomata4 has joined #ipfs
Fess has quit [Quit: Leaving]
Fess has joined #ipfs
lupi has quit [Max SendQ exceeded]
<Kythyria[m]>
They're relative to a gateway or mount point, basically
joocain2 has quit [Remote host closed the connection]
zautomata3 has quit [Ping timeout: 240 seconds]
lupi has joined #ipfs
joocain2 has joined #ipfs
<Fess>
ipfs directories aren't relative to a gateway
<Fess>
if you're actually using ipfs to access them
lupi has quit [Client Quit]
<Kythyria[m]>
Then what is `/ipfs/somehash/foo` actually relative _to_?
<Fess>
the relative url's they are talking about are relative within a directory
<Fess>
those directory hashes don't change depending on how you access them
<Fess>
gateways/ipfs daemon/other
newhouse has quit [Read error: Connection reset by peer]
<Kythyria[m]>
> If this is the objective, then the corresponding URIs should be ipfs:/ipfs/… and ipfs:/ipns/…. URI-centric tools would behave properly with both absolute and relative paths of the above canonical (scheme-less) IPFS form.
<Kythyria[m]>
No. There are multiple references to an IPFS path including the scheme and the hash as being relative.
<Kythyria[m]>
And that quote is actually wrong.
<Kythyria[m]>
It is not possible under present rules to host a website that uses root-relative paths on IPFS using that scheme.
<Fess>
I feel like I'm misunderstanding your question
<Kythyria[m]>
Why is the IPFS community so full of people who either a) think unix paths are the One True Syntax, or b) want URL syntax which doesn't work properly with existing tooling?
<Fess>
who knows, it's niche software which is heavily in development
<Kythyria[m]>
(and unix paths are strictly less expressive than URLs. You can do all the "nestable" things precisely as well. It's scheme specific how you traverse a "directory" when you're nesting like that, after all)
_ch4l45h45k4 has quit [Ping timeout: 240 seconds]
<Fess>
out of curiosity, what exactly are you trying to do with this right now
<Kythyria[m]>
I'm not. Part of why IPFS is much less interesting is all the tension between browsers that aren't Chrome and the "everything is a unix path" attitude.
<Kythyria[m]>
(also the tension between everything that assumes that unix paths refer to local things)
<Fess>
I'd guess it's probably more convention than anything else
ygrek has joined #ipfs
<Kythyria[m]>
URLs are also a convention, though.
zautomata4 has quit [Ping timeout: 240 seconds]
zautomata4 has joined #ipfs
jaboja has joined #ipfs
<Fess>
ipfs is a bit of a weird case with the content addressed storage, all content must be found through its hash or through a directory whose highest level is defined by a hash. That might influence design decisions
<Fess>
or it could just be people being stubborn/lazy
<Kythyria[m]>
I think it's more stubbornness than anything :)
joocain2 has quit [Remote host closed the connection]
joocain2 has joined #ipfs
ygrek has quit [Ping timeout: 248 seconds]
Ellenor has quit [Ping timeout: 255 seconds]
kaotisk has quit [Ping timeout: 252 seconds]
bwn has quit [Quit: Quit]
kaotisk has joined #ipfs
jaboja has quit [Remote host closed the connection]
roadhills20 has quit [Ping timeout: 252 seconds]
_xm has quit [Quit: leaving]
bwn has joined #ipfs
roadhills20 has joined #ipfs
Ellenor has joined #ipfs
kaotisk has quit [Ping timeout: 252 seconds]
kaotisk has joined #ipfs
}ls{ has quit [Ping timeout: 248 seconds]
}ls{ has joined #ipfs
zautomata4 has joined #ipfs
zautomata3 has quit [Ping timeout: 268 seconds]
onikuma has quit [Ping timeout: 272 seconds]
onikuma has joined #ipfs
zautomata4 has quit [Ping timeout: 252 seconds]
zautomata4 has joined #ipfs
engdesart has joined #ipfs
zautomata has joined #ipfs
zautomata4 has quit [Ping timeout: 256 seconds]
zautomata1 has joined #ipfs
zautomata has quit [Ping timeout: 248 seconds]
witten has joined #ipfs
zautomata2 has joined #ipfs
zautomata1 has quit [Ping timeout: 240 seconds]
}ls{ has quit [Quit: real life interrupt]
fredthomsen has joined #ipfs
engdesart has quit [Ping timeout: 260 seconds]
fredthomsen has quit [Quit: Leaving.]
atrosanguineous has quit [K-Lined]
nope_23 has joined #ipfs
anders[m] has joined #ipfs
Cavedude has quit [Quit: leaving]
Cavedude has joined #ipfs
witten has quit [Ping timeout: 252 seconds]
nope_23 has quit [Ping timeout: 248 seconds]
zautomata2 has quit [Ping timeout: 264 seconds]
zautomata3 has joined #ipfs
fredthomsen has joined #ipfs
rethresh has joined #ipfs
ulrichard has joined #ipfs
zautomata4 has joined #ipfs
zautomata3 has quit [Ping timeout: 272 seconds]
nope_23 has joined #ipfs
fredthomsen has quit [Quit: Leaving.]
fredthomsen has joined #ipfs
dimitarvp has joined #ipfs
fredthomsen has quit [Quit: Leaving.]
zautomata4 has quit [Ping timeout: 252 seconds]
zautomata has joined #ipfs
nope_23 has quit [Remote host closed the connection]
Neomex has quit [Remote host closed the connection]
zautomata1 has joined #ipfs
zautomata has quit [Ping timeout: 248 seconds]
rendar has joined #ipfs
zautomata2 has joined #ipfs
zautomata1 has quit [Ping timeout: 248 seconds]
zautomata4 has joined #ipfs
zautomata2 has quit [Ping timeout: 248 seconds]
nope_23 has joined #ipfs
joocain2 has quit [Remote host closed the connection]
tombusby has quit [Remote host closed the connection]
onikuma has quit [Remote host closed the connection]
trqx has quit [Remote host closed the connection]
zautomata has joined #ipfs
zautomata4 has quit [Ping timeout: 260 seconds]
joocain2 has joined #ipfs
joocain2 has quit [Remote host closed the connection]
joocain2 has joined #ipfs
nope_23 has quit [Quit: Leaving]
nope_23 has joined #ipfs
zautomata1 has joined #ipfs
zautomata has quit [Ping timeout: 252 seconds]
tombusby has joined #ipfs
nope_23 has quit [Client Quit]
movaex has joined #ipfs
movaex has quit [Client Quit]
zautomata2 has joined #ipfs
matoro has quit [Read error: Connection reset by peer]
zautomata1 has quit [Ping timeout: 268 seconds]
movaex has joined #ipfs
trqx has joined #ipfs
zautomata3 has joined #ipfs
movaex has quit [Client Quit]
zautomata3 has quit [Client Quit]
dolphy has joined #ipfs
dolphy has quit [Client Quit]
dolphy has joined #ipfs
zautomata2 has quit [Ping timeout: 264 seconds]
matoro has joined #ipfs
reit has quit [Ping timeout: 252 seconds]
[BFG] has joined #ipfs
<JCaesar>
Kythyria[m]: Btw, can you configure an average shell/… to launch a definable program when being handed a path starting with /ipfs? because I think you can with ipfs:
<Kythyria[m]>
Lots of things understand that URI schemes should be handled by particular programs, yes.
<Kythyria[m]>
The best you'll get with unix paths is to treat it like a local file, so you'll just get an error unless the FUSE module is mounted in the right places.
tombusby has quit [Ping timeout: 272 seconds]
tombusby has joined #ipfs
tombusby has quit [Remote host closed the connection]
tombusby has joined #ipfs
<[BFG]>
If all goes according to plan, Ethereum Fog will be capable of processing one million transactions per second by Q2 of 2019. This is a very ambitious goal, as no cryptocurrency in existence comes even close to that. Further down the line, the developers hope to match the business capabilities of Amazon Web Services, but in a decentralized manner. All of these things sound good on paper, but they will be very difficult to realize in the real world.
<[BFG]>
any relation between filecoin and ETH FOG?
trqx has quit [Quit: WeeChat 1.9.1]
<JCaesar>
No idea. But there is a #filecoin channel, that may be more appropriate…
trqx has joined #ipfs
reit has joined #ipfs
<lemmi>
Fess, Kythyria[m]: content addressing is the _CORE_ of ipfs is. stubbornness has nothing to with it. it is _the_ design decision.
<Fess>
I
<Fess>
I'm aware*
<Fess>
that conversation was about decisions in url syntax
<Fess>
obviously you always need to start with a content hash
<Fess>
whether it points to actual content or a directory or an IPNS address
<r0kk3rz>
a lot of things under the ipfs umbrella are about mashing all the different encoding methods into 'one true schema'
ylp has joined #ipfs
ylp has quit [Quit: Leaving.]
ylp has joined #ipfs
rgrau has joined #ipfs
Taoki has joined #ipfs
bomb-on has joined #ipfs
goiko has quit [Quit: ﴾͡๏̯͡๏﴿ O'RLY? Bye!]
dgrisham has quit [Ping timeout: 255 seconds]
_ch4l45h45k4 has joined #ipfs
dgrisham has joined #ipfs
dolphy has quit [Ping timeout: 256 seconds]
n7rc has joined #ipfs
n7rc has quit [Client Quit]
n7rc has joined #ipfs
dolphy has joined #ipfs
<lemmi>
Fess: ipfs dns --help
<lemmi>
Fess: under the hood, you need hashed content, yes. but that doesn't mean you need to see them
<Fess>
hmm interesting
_ch4l45h45k4 has quit [Quit: Leaving]
raynold has quit [Quit: Connection closed for inactivity]
AkhILman has quit [Ping timeout: 248 seconds]
jaboja has joined #ipfs
lupi has joined #ipfs
OneOneOneOne has joined #ipfs
lupi has quit [Max SendQ exceeded]
lupi has joined #ipfs
xnbya has quit [Ping timeout: 272 seconds]
Encrypt has joined #ipfs
xnbya has joined #ipfs
AkhILman has joined #ipfs
lupi has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
clemo has quit [Remote host closed the connection]
clemo1 has joined #ipfs
clemo1 is now known as clemo
deltab has joined #ipfs
AkhILman has joined #ipfs
Ronsor has quit [Ping timeout: 265 seconds]
Ronsor has joined #ipfs
Ronsor is now known as Guest68078
newhouse has joined #ipfs
chrysn has quit [Ping timeout: 255 seconds]
AkhILman has quit [Ping timeout: 260 seconds]
ulrichard has quit [Remote host closed the connection]
vivus has joined #ipfs
chriscool1 has joined #ipfs
chrysn has joined #ipfs
ralphtheninja has joined #ipfs
Seven_ has joined #ipfs
Guest68078 is now known as Ronsor
Ronsor has quit [Changing host]
Ronsor has joined #ipfs
mrBen2k2k2k has quit [Remote host closed the connection]
Jesin has joined #ipfs
mrBen2k2k2k has joined #ipfs
rethresh has quit [Remote host closed the connection]
Encrypt has quit [Quit: Quit]
n7rc has joined #ipfs
kaotisk has joined #ipfs
joocain2_ has joined #ipfs
joocain2 has quit [Ping timeout: 272 seconds]
dolphy has quit [Ping timeout: 260 seconds]
rgrau has quit [Ping timeout: 248 seconds]
muravey has joined #ipfs
chriscool1 has quit [Ping timeout: 252 seconds]
dragrope has joined #ipfs
fsiWebefb2 has joined #ipfs
fsiWebefb2 is now known as calvincs
Ronsor has quit [Ping timeout: 240 seconds]
atrapado_ has joined #ipfs
Ronsor has joined #ipfs
Ronsor is now known as Guest70229
rgrau has joined #ipfs
muravey has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
jaboja has quit [Ping timeout: 240 seconds]
onikuma has joined #ipfs
Guest70229 has quit [Changing host]
Guest70229 has joined #ipfs
Guest70229 is now known as ROnsor
ROnsor is now known as Ronsor
<calvincs>
Hello IPFS community, I had a few questions after tinkering with IPFS I was hoping someone could shed some light on. In short I want to distribute video content much the sameway webtorrents has the ability to do via WebRTC/P2P, where the more people visit the content, they seamlessly being sharing the content with each other. I believe ipfs-js can do this as well (or will eventually w/o them running a ipfs node locally), is this a
<Icefoz>
calvincs: It should be possible, though I don't know much about it.
<Icefoz>
Check out https://d.tube/ for inspiration perhaps, though as far as I can tell it's not open source.
<Icefoz>
calvincs: Also you got cut off at "is this a..."
<calvincs>
cont .. is this a correct assumption?
<Icefoz>
Ah, I don't know much about the current capabilities of js-ipfs
vflyson has joined #ipfs
muravey has joined #ipfs
xnbya has quit [Ping timeout: 268 seconds]
xnbya has joined #ipfs
muravey has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ygrek has joined #ipfs
chriscool1 has joined #ipfs
lamuguo has joined #ipfs
witten has joined #ipfs
Cavedude has quit [Ping timeout: 260 seconds]
bomb-on has quit [Quit: SO LONG, SUCKERS!]
Cavedude has joined #ipfs
Just_ has joined #ipfs
guideline has quit [Ping timeout: 272 seconds]
guideline has joined #ipfs
<calvincs>
d.tube looks like a cool project, but man do the videos take forever to load. This is exactly what I would want to avoid tbh. All that said, great concept.
gzs has joined #ipfs
<JCaesar>
calvincs: Did you manage to get it to use your local gateway?
<JCaesar>
(Also, I'm annoyed that the site itself does not seem to be served over IPFS and that there is no nice way to get the IPFS links to the videos from the site.)
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
<calvincs>
JCaesar, no, I just visited the site and clicked around.
voxelot has quit [Read error: Connection reset by peer]
sirn has quit [Ping timeout: 240 seconds]
jnagro has quit [Ping timeout: 240 seconds]
manveru has quit [Ping timeout: 240 seconds]
AphelionZ has quit [Ping timeout: 252 seconds]
mikolalysenko has quit [Ping timeout: 252 seconds]
emschwartz has quit [Ping timeout: 252 seconds]
ChrisMatthieu has quit [Ping timeout: 252 seconds]
robmyers has quit [Ping timeout: 252 seconds]
bigbluehat has quit [Ping timeout: 252 seconds]
Flyingmana_ has quit [Ping timeout: 252 seconds]
lohkey has quit [Ping timeout: 252 seconds]
bcavileer has quit [Ping timeout: 252 seconds]
tibor has quit [Ping timeout: 252 seconds]
ffmad_ has quit [Ping timeout: 252 seconds]
ehd has quit [Ping timeout: 252 seconds]
massino_ has quit [Read error: Connection reset by peer]
hsanjuan has quit [Read error: Connection reset by peer]
<miflow[m]>
gears -> api -> 127.0.0.0:XXXX
sickill has quit [Read error: Connection reset by peer]
ungoldman has quit [Ping timeout: 255 seconds]
bmpvieira has quit [Read error: Connection reset by peer]
koalalorenzo has quit [Ping timeout: 255 seconds]
wedowmaker__ has quit [Ping timeout: 255 seconds]
cloudcell has quit [Ping timeout: 265 seconds]
prettymuchbryce has quit [Ping timeout: 265 seconds]
PyHedgehog has quit [Ping timeout: 265 seconds]
cdata has quit [Read error: Connection reset by peer]
olizilla has quit [Read error: Connection reset by peer]
kumavis has quit [Ping timeout: 250 seconds]
nicknikolov has quit [Ping timeout: 240 seconds]
Cir0X has quit [Ping timeout: 240 seconds]
anderspree_ has quit [Read error: Connection reset by peer]
jesseclay has quit [Read error: Connection reset by peer]
aboodman has quit [Write error: Connection reset by peer]
karissa has quit [Read error: Connection reset by peer]
yoshuawuyts has quit [Read error: Connection reset by peer]
jimpick has quit [Read error: Connection reset by peer]
dvim has quit [Ping timeout: 246 seconds]
voldyman has quit [Read error: Connection reset by peer]
Wizek has quit [Ping timeout: 252 seconds]
zmanian has quit [Ping timeout: 252 seconds]
zielmicha_ has quit [Ping timeout: 252 seconds]
ebryn has quit [Ping timeout: 252 seconds]
jbenet has quit [Read error: Connection reset by peer]
strauberry has quit [Read error: Connection reset by peer]
caiogondim has quit [Ping timeout: 255 seconds]
exit70 has quit [Ping timeout: 255 seconds]
alkyl has quit [Ping timeout: 240 seconds]
leeola has quit [Read error: Connection reset by peer]
rflot has quit [Ping timeout: 264 seconds]
JaabbliZ has quit [Ping timeout: 264 seconds]
cmeik has quit [Read error: Connection reset by peer]
dryajov has quit [Ping timeout: 255 seconds]
Guest63586 has quit [Ping timeout: 255 seconds]
natewalck has quit [Ping timeout: 255 seconds]
jzelinskie has quit [Read error: Connection reset by peer]
bbielsa has quit [Ping timeout: 250 seconds]
grandy has quit [Ping timeout: 250 seconds]
bitspill has quit [Ping timeout: 250 seconds]
prosody has quit [Read error: Connection reset by peer]
iinaj has quit [Read error: Connection reset by peer]
Captain_Beezay has quit [Read error: Connection reset by peer]
Just_ has quit [Ping timeout: 272 seconds]
<miflow[m]>
and gears -> gateway -> 127.0.0.0:XXXX
<JCaesar>
I also can't manage to get it to save settings…
mappum has quit [Ping timeout: 255 seconds]
risk has quit [Ping timeout: 255 seconds]
hacdias has quit [Ping timeout: 252 seconds]
chriscool has quit [Ping timeout: 252 seconds]
rodarmor has quit [Ping timeout: 250 seconds]
sneak has quit [Ping timeout: 250 seconds]
machinaut has quit [Ping timeout: 264 seconds]
pedrot has quit [Ping timeout: 264 seconds]
bergie has quit [Ping timeout: 264 seconds]
jhiesey has quit [Ping timeout: 264 seconds]
kehao95 has quit [Ping timeout: 240 seconds]
blahah has quit [Ping timeout: 272 seconds]
Muis has quit [Ping timeout: 272 seconds]
aaaaaaaaa____ has quit [Ping timeout: 255 seconds]
sugarpuff has quit [Ping timeout: 255 seconds]
Arwalk has quit [Ping timeout: 255 seconds]
dsal has quit [Ping timeout: 255 seconds]
bengl has quit [Ping timeout: 250 seconds]
bret has quit [Ping timeout: 250 seconds]
hosh has quit [Ping timeout: 250 seconds]
physpi has quit [Ping timeout: 252 seconds]
taravancil has quit [Ping timeout: 252 seconds]
kareniel has quit [Ping timeout: 252 seconds]
daviddias has quit [Ping timeout: 252 seconds]
teej has quit [Ping timeout: 240 seconds]
caseorganic has quit [Ping timeout: 252 seconds]
runde has quit [Ping timeout: 246 seconds]
zenground0 has quit [Ping timeout: 264 seconds]
kyledrake has quit [Ping timeout: 255 seconds]
wa7son has quit [Ping timeout: 255 seconds]
justache has quit [Ping timeout: 255 seconds]
feross has quit [Ping timeout: 250 seconds]
metalrain_ has quit [Ping timeout: 250 seconds]
NeoTeo has quit [Ping timeout: 250 seconds]
shibacomputer has quit [Ping timeout: 255 seconds]
ec has quit [Ping timeout: 272 seconds]
mooman219 has quit [Ping timeout: 272 seconds]
<calvincs>
Yeah, I seen the ability to join my local IPFS node, however I really want a solution that can join IPFS w/o the user needing to have a local IPFS instance running (like webtorrents), its all in the browser.
jondashkyle has quit [Ping timeout: 272 seconds]
victorbjelkholm has quit [Ping timeout: 272 seconds]
jryans has quit [Ping timeout: 272 seconds]
<calvincs>
I dont want to expect that level of end user capability
omnigoat has quit [Ping timeout: 265 seconds]
<miflow[m]>
then ask heimindanger to add js-ipfs as an option :)
kaotisk has quit [Ping timeout: 260 seconds]
dignifiedquire has quit [Ping timeout: 246 seconds]
gzs has quit [Remote host closed the connection]
Mateon1 has quit [Ping timeout: 248 seconds]
dexter0 has quit [Ping timeout: 240 seconds]
Mateon1 has joined #ipfs
kaotisk has joined #ipfs
dexter0 has joined #ipfs
<calvincs>
could be wrong here, but it looks like its only accessing IPFS gateways to server up video vs using a js client (ipfs-js) to access the IPFS network to fetch the video content, Unless you have an IPFS node running locally.
<miflow[m]>
yes, but it is also kinda needed because else, unwatched videos would go down if the authors node went offline
<miflow[m]>
the site needs some filecoin incentive to keep uknown videos online
<calvincs>
absolutely, that totally makes sense. I personally want to setup a private ipfs cluster, seed some media content, and when users visit the content they initially pull from my gateways the data, however then want others who visit during that same timeframe to pull from other users, not just me.
<miflow[m]>
but i guess dtube design discussions are offtopic here
voxelot has joined #ipfs
dignifiedquire has joined #ipfs
n7rc has quit [Quit: n7rc]
<miflow[m]>
surely doable
<calvincs>
* weekend project :-)
onikuma has quit [Remote host closed the connection]
sugarpuff has joined #ipfs
sickill has joined #ipfs
JaabbliZ has joined #ipfs
tibor has joined #ipfs
olizilla has joined #ipfs
massino_ has joined #ipfs
ungoldman has joined #ipfs
nicknikolov has joined #ipfs
jhiesey has joined #ipfs
taravancil has joined #ipfs
caiogondim has joined #ipfs
yoshuawuyts has joined #ipfs
kehao95 has joined #ipfs
dryajov has joined #ipfs
pedrot has joined #ipfs
strauberry has joined #ipfs
mikolalysenko has joined #ipfs
kareniel has joined #ipfs
ebryn has joined #ipfs
emschwartz has joined #ipfs
PyHedgehog has joined #ipfs
jesseclay has joined #ipfs
aboodman has joined #ipfs
anderspree_ has joined #ipfs
bergie has joined #ipfs
jimpick has joined #ipfs
wa7son has joined #ipfs
hosh has joined #ipfs
karissa has joined #ipfs
blahah has joined #ipfs
kyledrake has joined #ipfs
AphelionZ has joined #ipfs
jbenet has joined #ipfs
kumavis has joined #ipfs
prosody has joined #ipfs
lohkey has joined #ipfs
ffmad_ has joined #ipfs
bengl has joined #ipfs
jryans has joined #ipfs
dsal has joined #ipfs
mappum has joined #ipfs
bbielsa has joined #ipfs
Flyingmana_ has joined #ipfs
metalrain_ has joined #ipfs
prettymuchbryce has joined #ipfs
bigbluehat has joined #ipfs
cmeik has joined #ipfs
leeola has joined #ipfs
exit70 has joined #ipfs
chriscool has joined #ipfs
<lgierth>
Kythyria[m]: tbh i haven't read the whole discussion from last night yet, have put it in a textfile though :) -- we don't think unix paths are the One True Way of doing things, that's why there's also URL schemes: ipfs:// and ipns://
* Kythyria[m]
nods
<lgierth>
ideally in the long term everything would be dweb: URIs with unix-ish paths in them, but we're not willing to gamble all the rest on that
<Kythyria[m]>
Those schemes would fit nicely
cloudcell has joined #ipfs
risk has joined #ipfs
<lgierth>
URLs are what we have right now so that's what we'll integrate with
NeoTeo has joined #ipfs
koalalorenzo has joined #ipfs
bitspill has joined #ipfs
<Kythyria[m]>
(for chaining URIs, I'm partial to doing so with bangs)
<lgierth>
ipfs-companion already does it nicely
sirn has joined #ipfs
<lgierth>
unix paths are also something we already have though!
<lgierth>
URLs were the deviation from that standard way back in the day
dvim has joined #ipfs
teej has joined #ipfs
justache has joined #ipfs
<Kythyria[m]>
Trying to jam chaining into unix path syntax makes it unobvious at what point in the path you need to invoke the next handler.
physpi has joined #ipfs
Cir0X has joined #ipfs
hsanjuan has joined #ipfs
jnagro has joined #ipfs
bmpvieira has joined #ipfs
<Kythyria[m]>
In particular, if there are any handlers which take a directory as their context (ie, something that can be in the middle of a path), you can create ambiguity.
ec has joined #ipfs
grandy has joined #ipfs
victorbjelkholm has joined #ipfs
Wizek has joined #ipfs
machinaut has joined #ipfs
<lgierth>
inventing a new path syntax for that is a high price to pay though
hacdias has joined #ipfs
iinaj has joined #ipfs
<Kythyria[m]>
Or if you're using HTTP, where truncating the path might yield another path you can use as a file.
Captain_Beezay has joined #ipfs
<lgierth>
making the chain a syntax element also takes a way quite a bit of flexibilityu
muravey has joined #ipfs
<Kythyria[m]>
It is a syntax element though.
<lgierth>
(sorry about teh sloppy typing, i'm watching 34c3 streams :))
Ecran has joined #ipfs
<Kythyria[m]>
Otherwise you have stupid results like the names of schemes are reserved filenames.
<lgierth>
separate schemes yield incompatible applications though
omnigoat has joined #ipfs
<Kythyria[m]>
eg, suppose I register `git` as a protocol that can read bare git repositories.
<lgierth>
the idea at the heart of these unix-ish paths is that the filesystem is the interface
zielmicha_ has joined #ipfs
<lgierth>
ok
voldyman has joined #ipfs
<Kythyria[m]>
I now cannot use `git` as a filename.
daviddias has joined #ipfs
Just_ has joined #ipfs
<lgierth>
that means you cannot freely create files or other things in any arbitrary directory
shibacomputer has joined #ipfs
bcavileer has joined #ipfs
<lgierth>
that's a limitation that's already there though
<Kythyria[m]>
Really any directory.
Muis has joined #ipfs
sneak has joined #ipfs
<Kythyria[m]>
No it's not.
robmyers has joined #ipfs
<Kythyria[m]>
Either things which really are filesystem-like MUST only be the last thing in the chain, or you get ambiguity.
jzelinskie has joined #ipfs
caseorganic has joined #ipfs
manveru has joined #ipfs
<Kythyria[m]>
Also, the attitude that the filesystem is the interface implies that the POSIX filesystem API actually overlaps every other API in the world in a reasonable manner.
<Kythyria[m]>
And take that hypothetical git handler: installing it would prevent having any file in the filesystem called `git`. Have fun trying to run git itself.
<lgierth>
you can totally have specialized APIs -- we already these, e.g. an API that's transported over a unix socket
cdata has joined #ipfs
<lgierth>
or classic APIs that take paths as their subjects
<Kythyria[m]>
But then you're not really using the filesystem as more than a place to stuff the handle.
<lgierth>
git handler would be e.g. at /git, not ./git
<Kythyria[m]>
Incidentally, how do you expose a HTTP client as a filesystem? Oh right, you can't.
<Kythyria[m]>
So how do I chain things onto the end of that?
<Kythyria[m]>
So it would be like doing git:file:/home/kythyria/whatever ?
ehd has joined #ipfs
<lgierth>
a unix socket at e.g. /dns/foo.com/tcp/443/https/foo.com/some/path
<Kythyria[m]>
And the protocol over that socket?
rflot has joined #ipfs
<Kythyria[m]>
It would have to be an entire HTTP request minus the first line and host header.
<Kythyria[m]>
And the entire response.
<lgierth>
(looking for that plan9 http client)
<Kythyria[m]>
That plan 9 HTTP client is using the filesystem as an RPC interface. A bad one.
<Kythyria[m]>
You'll still need to generate a client library for the http client!
<lgierth>
yeah who knows if really *everything* needs to be addressed by paths -- that'd be way out of scope for the ipfs project
<lgierth>
what we're talking about is having a small handful of specific things addressed that way
<Kythyria[m]>
It's an assumption people make in the context of the project though.
<lgierth>
content-addressed data, and networked processes (=multiaddr)
<lgierth>
yep that's true
<lgierth>
i have this spec in the works and it's a good point that it should have a "scope of this" section
<Kythyria[m]>
multiaddr has a very specific path syntax though, it's just that / is the only official special character.
<lgierth>
yep multiaddr basically is an attempt at mapping every possible network stack on a filesystem path
<Kythyria[m]>
Have the people that thought this was a good idea met the people who think that /https/io/ipfs/dist/go-ipfs/v0.3.13 is a good URL syntax?
<lgierth>
not sure i understand the question
<lgierth>
but URLs are a less than ideal syntax for representing network connections
rgrau has quit [Ping timeout: 265 seconds]
<lgierth>
ipfs already works with a couple of network constructions that simply can't be represented in a URL
<Kythyria[m]>
That doesn't mean unix paths are a signifcantly more reasonable design. Certainly a hypothetical dnsfs used that way could not contain any mountpoints without significant changes to the concept of a mountpoint.
Neomex has joined #ipfs
<cipres>
hi everyone
<Kythyria[m]>
(ie, it's the job of dnsfs to know what `tcp` means)
tombusby has quit [Remote host closed the connection]
tombusby has joined #ipfs
<cipres>
still discussing URLs syntax ?
tombusby has quit [Remote host closed the connection]
tombusby has joined #ipfs
calvincs has quit [Quit: Page closed]
fsiWebefb2 has joined #ipfs
fsiWebefb2 is now known as calvincs
fredthomsen has joined #ipfs
movaex_ has joined #ipfs
fredthomsen has quit [Quit: Leaving.]
movaex_ has quit [Quit: Leaving]
espadrine has joined #ipfs
witten has quit [Ping timeout: 265 seconds]
lupi has joined #ipfs
rgrau has joined #ipfs
lupi has quit [Client Quit]
fredthomsen has joined #ipfs
lupi has joined #ipfs
Encrypt has joined #ipfs
jondashkyle has joined #ipfs
atrapado_ has quit [Quit: Leaving]
ChrisMatthieu has joined #ipfs
natewalck has joined #ipfs
muravey has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
zenground0 has joined #ipfs
feross has joined #ipfs
Arwalk has joined #ipfs
wedowmaker__ has joined #ipfs
runde has joined #ipfs
fredthomsen has quit [Quit: Leaving.]
alkyl has joined #ipfs
aaaaaaaaa____ has joined #ipfs
Just_ has quit [Quit: Connection closed for inactivity]
muravey has joined #ipfs
Guest63586 has joined #ipfs
rodarmor has joined #ipfs
Encrypt has quit [Quit: Quit]
Just_ has joined #ipfs
lupi has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
muravey has quit [Ping timeout: 240 seconds]
lupi has joined #ipfs
ONI_Ghost has joined #ipfs
nibble has quit [Ping timeout: 248 seconds]
l2d has quit [Read error: Connection reset by peer]
l2d has joined #ipfs
lupi has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
calvincs has quit [Quit: Page closed]
droman has joined #ipfs
rafajafar has joined #ipfs
<rafajafar>
hey guys, I'm trying to create a CIDv0 in-browser based on an arbitrary buffer. I found the js-cid library but nothing i put in seems to work.
<rafajafar>
anyone think they can help?
<rafajafar>
I get an error "Error: multihash length inconsistent:" when I try to put a protobuff into the CID()
rafajafar has quit [Quit: Leaving]
PyHedgehog has quit [Quit: Connection closed for inactivity]
chriscool1 has quit [Quit: Leaving.]
robattil1 has joined #ipfs
Ellenor is now known as Eamon
robattila256 has quit [Ping timeout: 240 seconds]
Jesin has quit [Quit: Leaving]
chriscool1 has joined #ipfs
Adbray has quit [Quit: Quit]
chriscool1 has quit [Ping timeout: 240 seconds]
discopatrick has joined #ipfs
Adbray has joined #ipfs
witten has joined #ipfs
Fess has joined #ipfs
droman has quit [Quit: WeeChat 2.0.1]
jaboja has joined #ipfs
chriscool1 has joined #ipfs
rgrau has quit [Ping timeout: 248 seconds]
witten has quit [Ping timeout: 272 seconds]
LiuWeiHua[m] has joined #ipfs
Eamon is now known as Ellenor
dragrope has quit [Read error: Connection reset by peer]
newhouse has quit [Read error: Connection reset by peer]
Adbray has quit [Remote host closed the connection]