<nicolagreco_>
ok I just saw it on the calendar, perfect
<noffle>
mnp: the changes should pick up. if it's a gx dep, going to go/src/gx/ipfs/QMfoobar/go-somelib to edit ought to do it
simpbrain has joined #ipfs
dedgecumbe has joined #ipfs
<dedgecumbe>
hello! how does the public gateway to ipfs work? who donates the bandwidth, and how is it possible to distinguish between something being unreachable and the gateway being slow/throttling at the moment?
overdangle has quit [Remote host closed the connection]
<noffle>
dedgecumbe: it's run / paid for by protocol labs
<mnp>
noffle, thx, i tried that, but good to know i was on the right track
<noffle>
dedgecumbe: unreachable and slow are impossible to differentiate :(
<dedgecumbe>
noffle: :(.
<noffle>
dedgecumbe: distributed networks are tricky -- the thing you want may exist, but be partitioned from the node's own network
<dedgecumbe>
noffle: i need to read about how the routing works at some point.
<mnp>
at least you know it's consistent
<dedgecumbe>
noffle: am I correct in thinking that the gateway doesn't need a direct connection to a host that has the file, just some route?
<dedgecumbe>
the whitepaper explains this very concisely. :)
<dedgecumbe>
(there is much to learn, young padawan)
oneiroscopist has joined #ipfs
<noffle>
dedgecumbe: sorry, it's not fresh enough in my mind to give a good answer -- whyrusleeping or jbenet ought to have a good answer
<dedgecumbe>
it seems clear to me from the whitepaper that the direct connection isn't necessary, but that a full explanation requires a fair amount of reading outside of the scope of ipfs itself
<drwasho>
Hey folks, can someone try and see this gif I added: QmVjyhutYZDMdYHCbrxZqoRZQRKfYQZCvRS8ZoqT9MsUh2
<dedgecumbe>
'a DSHT based on S/Kademlia and Coral'; it's like Bitcoin all over again (for me); new words! :P
Akaibu has quit [Quit: Connection closed for inactivity]
<drwasho>
Anyone?
<noffle>
drwasho: a good predicate is seeing if it shows up for you on the public gateway
computerfreak has quit [Remote host closed the connection]
<dedgecumbe>
it's not working for me (which might explain some things.. :P)
<dedgecumbe>
i'm wondering if being behind NAT is causing me issues. i've managed to access some small things using the gateway but otherwise it doesn't seem to work
<dedgecumbe>
(small files that I've added, with a daemon running)
hellertime has quit [Quit: Leaving.]
<drwasho>
dedgecumbe: strange because I'm on an open NAT I think
<drwasho>
Also I can't view the ID details of my friend Chris and he's behind a nat
<drwasho>
we're both on 0.4.0-dev
<drwasho>
I thought there was supposed to be some NAT traversal on 0.4.0
ploopkazoo has quit [Quit: ZNC - 1.6.0 - http://znc.in]
<dedgecumbe>
whyrusleeping: well i've been looking into that but I need to make sure I can actually get rid of android
<whyrusleeping>
i'm like, 90% certain you can run ubuntu on a nexus tablet
<dedgecumbe>
it looks like the nexus 7 can run ubuntu touch and someone managed arch linux arm at some point but it's probably not very supported and I'm guessing the kernel is years out of date by now
<whyrusleeping>
dedgecumbe: eh, if youre intepid i'd just grab the kernel configuration files that they had before and try it on a later kernle
<whyrusleeping>
that worked for me most of the time on chromebooks with arch
O47m341 has joined #ipfs
<dignifiedquire>
good morning whyrusleeping
<whyrusleeping>
dignifiedquire: goooood morning
ipfsrocks has quit [Ping timeout: 260 seconds]
tilgovi has joined #ipfs
<dignifiedquire>
whyrusleeping: someone is in a good mood :D
ploopkazoo has joined #ipfs
ipfsrocks has joined #ipfs
<whyrusleeping>
lol, one extra hour before forced social interaction?
<whyrusleeping>
hell yeah
gordonb has left #ipfs [#ipfs]
<richardlitt>
:|
ugjka has quit [Ping timeout: 250 seconds]
martinkl_ has joined #ipfs
ylp has quit [Quit: Leaving.]
<dignifiedquire>
:D
M-giodamelio has quit [Quit: node-irc says goodbye]
tilgovi has quit [Ping timeout: 248 seconds]
jokoon has quit [Quit: Leaving]
<noffle>
:o
bedeho has joined #ipfs
<dignifiedquire>
whyrusleeping: go-ipfs just crashed hard on me
<dignifiedquire>
not sure why though
<dignifiedquire>
wanna get a stack trace?
<noffle>
voxelot: heya! what time did you want to meet re pubsub?
<whyrusleeping>
dignifiedquire: sweet, i'd love one
<dignifiedquire>
voxelot: did you see my commit/msg
<voxelot>
noffle: was thinking after the IPLD meeting would be a good time, if dignifiedquire or richardlitt can host, haven't heard from haad yet
<voxelot>
haad: yeah idk either haha, i had question cus im a noob but i've done a lot more research recentl yand dont have them anymore
wiedi has quit [Quit: ^Z]
<noffle>
i.e. central signalling server
<noffle>
voxelot: we can talk if you're like -- share our knowledge on the topic
<noffle>
you'd
<voxelot>
i mean that would be helpful for me, if it is for others then yes!
<voxelot>
dignifiedquire: could me repo dep be outdated? "ipfs-repo": "^0.5.0"
<haad>
but seeing the conversation above, my opinion is that pubsub should first be implemented in go-ipfs, js-ipfs is still very young. in addition, we don't have a clear solution yet so there's quite a bit of work that needs to be done before we can say "hey, this is how it should be implemented" (go or js)
<dignifiedquire>
voxelot: I have 0.5.1
<haad>
I would really wanna have whyrusleeping in that conversation though, have his thoughts and ideas out in the open
<noffle>
haad: voxelot: I'm kind of unclear on what the pressing need for pubsub in ipfs is at this juncture. Who's blocked by this and that application-level alternatives are not working for?
ugjka has quit [Ping timeout: 260 seconds]
<voxelot>
haad: agreed, i think we need the go gents in on this for a good converstation
<voxelot>
noffle: i really want a fully distributed way to signal new events over an application like a market place, adding products
<voxelot>
or even just a social media post
<richardlitt>
5 mins ya'll
<haad>
noffle: from my perspective it's absolutely crucial feature to make orbit fully p2p, that's why I'm pushing it.
<richardlitt>
No sprint updates needed; today is just a hangout.
<whyrusleeping>
haad: yeah, i'm open for a chat about pubsub generally whenever
<noffle>
voxelot: haad: you can already do this using the bittorrent swarm + dht as an overlay
<noffle>
it's already doable today
<noffle>
I wrote a lib for it
<voxelot>
ohhh?? :)
<noffle>
we don't need this in core to make pubsub apps
<noffle>
distributed pubsub isn't dependant on ipfs in any way, other than as an optimization to reuse connections/its swarm
<haad>
noffle: I agree mostly. however, I wouldn't want to use the BT swarm but IPFS swarm. maybe it's more a core libp2p feature than cor ipfs feature, you're right.
<noffle>
haad: yes, eventually we should to reuse connections better, but we are *not* blocked on pubsub today
<alu>
happy pi day
tilgovi has joined #ipfs
<voxelot>
3.14 to you too alu!
computerfreak has quit [Remote host closed the connection]
micxjo has quit [Quit: Connection closed for inactivity]
martink__ has joined #ipfs
mats_ has quit [Ping timeout: 248 seconds]
martinkl_ has quit [Ping timeout: 276 seconds]
calavera has joined #ipfs
<nicolagreco_>
leaving to the stream, I got nothing to say!
martink__ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<ipfsbot>
[js-ipfs] xicombd force-pushed feature/object from b6c07a3 to b06415d: https://git.io/vzdPv
<ipfsbot>
js-ipfs/feature/object b06415d Francisco Baio Dias: Add object endpoints and cli commands
dvn has joined #ipfs
zeroish has joined #ipfs
nonaTure has joined #ipfs
<ipfsbot>
[js-ipfs] xicombd force-pushed feature/object from b06415d to 69e16f0: https://git.io/vzdPv
<ipfsbot>
js-ipfs/feature/object 69e16f0 Francisco Baio Dias: Add object endpoints and cli commands
<ipfsbot>
[go-ipfs] dignifiedquire opened pull request #2463: feat: Update the webui to work with the latest changes in 0.4 (master...webui-04) https://git.io/va8Ny
<shadoxx>
voxelot: are you just having trouble with the development environment? i'm surprised there isn't a Vagrant for IPFS already setup for development. i could probably do something about that
atrapado has joined #ipfs
<voxelot>
shadoxx: that would be awesome, as far as i know we dont hav ea standard env
<voxelot>
but it's also good to try and track down bugs early on and not just assume it will outside of our dev env
<shadoxx>
voxelot: Vagrant will completely change the workflow for development in a really good way. Most of you are developing on Ubuntu 14.04? I'll have to get familiar with your toolchain.
<voxelot>
shadoxx: cool, interested in checking out, but i'd say 14.0.4 is probably a rare OS around here
<shadoxx>
what do most people develop on?
<voxelot>
pass a straw poll around?
<voxelot>
hmm probably too many variations for that
<dignifiedquire>
shadoxx: nope not a lot of people on Ubuntu here
<dignifiedquire>
at least 3 on OS X
<dignifiedquire>
and 1 on arch
<nicolagreco_>
jbenet: do we have any hangout url?
<dignifiedquire>
tbh I think it's good that a lot of people develop on different OSes without a vm, as that means things are tested on a much broader variety
<voxelot>
we need more windows users around here
atrapado has quit [Remote host closed the connection]
atrapado has joined #ipfs
<nicolagreco_>
jbenet: we are all in the hangout ^
calavera has joined #ipfs
calavera has quit [Client Quit]
calavera has joined #ipfs
calavera has quit [Client Quit]
calavera has joined #ipfs
martinkl_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
mildred has quit [Read error: Connection reset by peer]
mildred has joined #ipfs
rendar has quit [Ping timeout: 244 seconds]
ipfsrock1 has joined #ipfs
ipfsrocks has quit [Ping timeout: 240 seconds]
Senji has joined #ipfs
devbug has joined #ipfs
wiedi has joined #ipfs
slothbag has joined #ipfs
rendar has joined #ipfs
nicolagreco_ has joined #ipfs
devbug has quit [Ping timeout: 260 seconds]
martinkl_ has quit [Read error: Connection reset by peer]
martink__ has joined #ipfs
<whyrusleeping>
dignifiedquire: i think that transfer encoding is a bug in go
ashark has quit [Ping timeout: 244 seconds]
<dignifiedquire>
whyrusleeping: Oo
martink__ has quit [Client Quit]
<whyrusleeping>
in the method that they write all the headers out, they define this 'setHeaders' thing, and set the transfer encoding on it: https://golang.org/src/net/http/server.go#L1071
<whyrusleeping>
they then write the headers we set, followed by the headers from that object
<dignifiedquire>
looks like they are adding chunked themselves
<dignifiedquire>
are we setting it ourselves as well?
<whyrusleeping>
yeah, we're supposed to
<whyrusleeping>
and it checks that we set it up higher
<dignifiedquire>
Oo but why do they add it themselves then :/
<whyrusleeping>
hrm... i think i have an idea
<whyrusleeping>
(might still be our bug)
<whyrusleeping>
yeah, i'm more sure its not our bug
<whyrusleeping>
in the code i linked, i'm 90% sure they should call delHeader
<whyrusleeping>
on the transfer encoding
<dignifiedquire>
or not add it if it already exists
<whyrusleeping>
right
mildred has quit [Ping timeout: 248 seconds]
devbug has joined #ipfs
qbg has joined #ipfs
calavera has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
calavera has joined #ipfs
O47m341 has quit [Ping timeout: 246 seconds]
ipfsrock1 has quit [Ping timeout: 244 seconds]
<whyrusleeping>
dignifiedquire: so it didnt like that we were doing chunked encoding and specifying content length
<dignifiedquire>
so?
<dignifiedquire>
nvm I can't read
<whyrusleeping>
removing our setting of content length seems to have done the trick
<dignifiedquire>
is the length still correct?
<whyrusleeping>
length?
ipfsrocks has joined #ipfs
<dignifiedquire>
the content lentgth that go sets vs our own
<whyrusleeping>
it doesnt set content length
<whyrusleeping>
content length and chunked encoding are separate things
<dignifiedquire>
but you just said we were setting it
<dignifiedquire>
don't we need that actually?
<whyrusleeping>
we were
<whyrusleeping>
we don't need it
<dignifiedquire>
ah okay nice
<whyrusleeping>
i changed it to be X-Content-Length
<whyrusleeping>
because i think its used by the cli for progress bars sometimes
<lgierth>
while Access-Control-Allow-Origin = "Access-Control-Allow-Origin" ":" origin-list-or-null | "*"
<lgierth>
(note the "*")
ipfsrocks has quit [Ping timeout: 250 seconds]
<dignifiedquire>
jbenet: made some progress on starting defining ipld in terms of graph theory, I'm pretty sure it will be possible to define a combinator like structure for mapping ipld graphs to arbitray dags. I even found a nice way to encode seperatora in the graph I think
<dignifiedquire>
which allows them to be part of the mapping
<dignifiedquire>
lgierth: yep :/
M-amstocker has quit [Quit: node-irc says goodbye]
pfraze has quit [Remote host closed the connection]
atrapado has quit [Quit: Leaving]
devbug has quit [Ping timeout: 276 seconds]
devbug has joined #ipfs
O47m341 has joined #ipfs
nicolagreco_ has quit [Quit: nicolagreco_]
<ipfsbot>
[go-ipfs] whyrusleeping created fix/double-encoding-header from master (+0 new commits): https://git.io/va4MU
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to fix/double-encoding-header: https://git.io/va4M3
<ipfsbot>
go-ipfs/fix/double-encoding-header 7fe0092 Jeromy Johnson: fix double transfer encoding head problem...
<ipfsbot>
[go-ipfs] whyrusleeping opened pull request #2465: fix double transfer encoding head problem (master...fix/double-encoding-header) https://git.io/va4M4
<whyrusleeping>
dignifiedquire: ping
<dignifiedquire>
whyrusleeping: pong
<whyrusleeping>
take a look at that new PR
nicolagreco_ has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping deleted t0300-improve-debug-message at 94083ae: https://git.io/va4Dl
<dignifiedquire>
whyrusleeping: looks good, only on phone though will test in the morning
<dignifiedquire>
thanks :)
<whyrusleeping>
dignifiedquire: you're welcome :)
<whyrusleeping>
that was an obnoxious bus
<whyrusleeping>
bug*
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed fix/double-encoding-header from 7fe0092 to a1f037c: https://git.io/va4MU
<ipfsbot>
go-ipfs/fix/double-encoding-header a1f037c Jeromy Johnson: fix double transfer encoding head problem...
jaboja has quit [Remote host closed the connection]
wiedi has quit [Read error: Connection reset by peer]
devbug has quit [Read error: Connection reset by peer]
tmg has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed fix/double-encoding-header from a1f037c to 8506023: https://git.io/va4MU
<ipfsbot>
go-ipfs/fix/double-encoding-header 8506023 Jeromy Johnson: fix double transfer encoding head problem...
wiedi has joined #ipfs
devbug has joined #ipfs
<jbenet>
dignifiedquire: that's great! :)
devbug has quit [Ping timeout: 244 seconds]
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to fix/double-encoding-header: https://git.io/va49K