notduncansmith has quit [Read error: Connection reset by peer]
hjoest2 has quit [Ping timeout: 264 seconds]
akhavr has quit [Read error: Connection reset by peer]
akhavr has joined #ipfs
hjoest has joined #ipfs
rendar has joined #ipfs
hjoest has quit [Ping timeout: 264 seconds]
hjoest has joined #ipfs
<davidar>
jbenet: The scheme consists of a sequence of characters beginning with a letter and followed by any combination of letters, digits, plus (+), period (.), or hyphen (-).
<davidar>
x+:/ ! ☺
<jbenet>
haha
<jbenet>
x.x://
<davidar>
How about oo:/ as in infinity?
<davidar>
(jbenet)
<M-Kodo>
We need more emoji based protocols
<davidar>
haha
<jbenet>
i think it's important to drive the point: it's a unix path.
<davidar>
<jbenet>
otherwise there's no reason for it
<M-Kodo>
<M-Kodo>
I'm really surprised there is no
<M-Kodo>
They make up like 80% of the internet
hjoest has quit [Ping timeout: 264 seconds]
<davidar>
jbenet: yeah, but it also has to be sexy ☺
<fps>
I://
<fps>
or does followed by any mean "followed by at least one.."
* M-Kodo
goes off to write a binary serialization spec where the first character is
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<davidar>
webnix is descriptive, but it also looks a bit weird
<rendar>
why can't we just use ipfs:// ? its neat
<davidar>
(not in any rational way, just aesthetically)
<jbenet>
rendar: because this is broader. it will als have "nix;//dns/..." and other protocols who want to be mounted on unix..
<rendar>
jbenet: now you are speaking about the URI, right?
hjoest has joined #ipfs
<davidar>
Yeah, I like web xweb net path
<davidar>
shorter is better though imo
<ianopolous>
jbenet: the Java http api implementation is starting to look decent, still most of the second level commands to do yet though.
<jbenet>
x:// looks very cool and mysterious. sad about not being able to use it with windows
<jbenet>
ianopolous that's great to hear!
<davidar>
jbenet: you reckon iana would allow web to be registered?
<jbenet>
likely not.
<jbenet>
prob reserved
<davidar>
yeah, feared not
<davidar>
We need something common, but not too general
Gaboose has joined #ipfs
<jbenet>
p2p:// is not accurate.
<davidar>
I'm still in favor of just using a verb like "get" or something
<jbenet>
get is nice actually
<jbenet>
very nice
<rendar>
yeah, i agree, get://
<davidar>
Yay
<rendar>
would be short, cool, explicative, and concise
<ianopolous>
jbenet would you rather pull it in centrally now or when it is 100% complete?
<jbenet>
ianopolous up to you really. fine either way
<davidar>
jbenet: I'm happy to centralize the Haskell API too, not yet finished though
<jbenet>
davidar: you wanted review?
<jbenet>
cleichner may be able to provide some
<davidar>
yeah, bitemyapp reviewed it already
<jbenet>
ah ok
<ianopolous>
maybe better to do it now so people can find it easier (it's already sufficient for Peergos to use)?
<davidar>
Haven't fixed the (minor) issues he raised yet though
hjoest has quit [Read error: Connection reset by peer]
domanic has joined #ipfs
<zignig>
jbenet: is there any way to get the ipns record as json ?
<flyingkiwi>
When a resource is cached on my node - is my node seeding this resource (as long as its cached)?
<zignig>
yes, until you do a 'ipfs repo gc'
<zignig>
if if you pin it , it sticks there.
<jbenet>
ianopolous sgtm. i invited you to org. should be able to transfer ownership to org
* jbenet
davidar: let's go with "get://" then. what we need is to write a "protocol loader" tool that people install and has a list of "registered mounts" (like `/ipfs, /ipns, /dns`)
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
hjoest has joined #ipfs
<davidar>
jbenet, cool, sgtm ☺
<Gaboose>
some time ago whyrusleepin said he's redoing dht, putting crdt's like grow-only sets usable to users, anyone knows if that was done? i ask humbly
hjoest has quit [Read error: Connection reset by peer]
chiui has joined #ipfs
hjoest has joined #ipfs
zabirauf has joined #ipfs
<davidar>
jbenet, would the protocol loader be in the browser or os?
<davidar>
browser should be mostly trivial
<davidar>
Also, how would registration work? Hard coded or more extensible?
gaboose_ has joined #ipfs
Dioxy has joined #ipfs
<Dioxy>
Hi all
<davidar>
Hi dioxy
<Dioxy>
struggling to get IPFS to connect to any peers or query the DHT at work - is there any way to set a proxy?
domanic has quit [Ping timeout: 240 seconds]
zabirauf has quit [Max SendQ exceeded]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<davidar>
dioxy: firewall blocking most ports?
<davidar>
I know jbenet wants to run some nodes over the https port for that reason
<ianopolous>
jbenet: I couldn't see anything about being added to org, or do the transfer..
<Dioxy>
davidar: Yes, exactly.
<ianopolous>
ah got it
dignifiedquire has joined #ipfs
<davidar>
Dioxy: do you only have 80 and 443 open?
<davidar>
i don't think there's any nodes listening on those ports yet
<ianopolous>
jbenet: all done. The name looks out of place in ipfs. Any suggestions? lower case?
<ianopolous>
java-ipfs-api ?
<davidar>
ianopolous: that's what i'd suggest
<fps>
does the rest interface only support GET operations though [re get://]
<fps>
?
<davidar>
fps: get:// != http://
<ianopolous>
jbenet, davidar: seems like I don't have the permissions to rename it? so any one else feel free to.
<davidar>
fps: get:// would be independent of the rest interface
<davidar>
that would just be an implementation detail
<davidar>
but no, i don't think it would be useful for pushing data to ipfs, only viewing it
Encrypt has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
atomotic has joined #ipfs
<Dioxy>
davidar - It would seem the proxy is filtering traffic on ports 80 and 443
rongladney has joined #ipfs
<davidar>
Dioxy: filtering to what extent?
<davidar>
443 in particular, where deep packet inspection isn't possible
<reit>
<whyrusleeping> Vyl: its really just a pubsub implementation
<reit>
<whyrusleeping> for 'real time'
<reit>
sorta, i'm thinking along the lines that eventually people are going to want to do real time streaming of arbitrary data, and dht + bitswap isn't going to be fast enough in itself, not for real time transfers
<reit>
i'm thinking something like a tree might work, with each node receiving a packet from its parent and then directly resending it off to its children, having as many of which as its connection can manage
<reit>
the interesting part is considering how to build it so that those nodes that support the most children get to go further up the tree (closer to the root)
hjoest has quit [Read error: Connection reset by peer]
hjoest has joined #ipfs
<Dioxy>
davidar - ports are completely closed from what I can tell.
<rendar>
reit: but a DHT isn't almost like that? i mean, the parent node dispatch the request to the correct node using a DHT, while the tree should be traversed...
<davidar>
Dioxy: so you have no web access then???
<davidar>
I mean remote ports
<davidar>
as in, which ports can you connect to over the internet?
<davidar>
I wouldn't expect inbound connections to work, no
<Dioxy>
Outbound are largely unblocked. Can connect to services on port 80
<reit>
rendar: not really sure what you mean, please elaborate
chiui has quit [Ping timeout: 264 seconds]
<davidar>
Dioxy: if outbound port 4001 isn't blocked, you shouldn't be having problems connecting to ipfs
<Dioxy>
4001 is unblocked. But should IPFS be listening on my internal IP address?
<davidar>
Dioxy: yeah, it can still accept connections from inside your network, just not the rest of the world
<Dioxy>
Ain't no one else running IPFS here ;)
notduncansmith has joined #ipfs
<davidar>
so it shouldn't really matter if it's listening then?
notduncansmith has quit [Read error: Connection reset by peer]
<Dioxy>
No it doesn't matter - but I'd like to be able to connect to the swarm!
hjoest has quit [Ping timeout: 264 seconds]
<davidar>
so, 4001 is unblocked, but ipfs is unable to connect to other machines on that port?
<nomoremoney>
Hey, the links on the bottom of the webpage, that links to IPN and filecoin seems to be broken. https://ipn.io and https://filecoin.io doesn't load
bedeho has quit [Ping timeout: 246 seconds]
<ehd>
cryptix: thanks! that's exactly the reasoning i was hoping for
rongladney has quit []
rongladney has joined #ipfs
<davidar>
nomoremoney: I thought lgierth had fixed that already
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
<cryptix>
jbenet: can we have this on all of the multi* projects? :P /ipfs/QmVWanfJkFrznbCSLDfAwUVXDAtMwf3gfVuM6cggonQAmC
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
infinity0 has joined #ipfs
Encrypt has quit [Quit: Quitte]
carstn has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
magneto1 has joined #ipfs
carstn has joined #ipfs
<davidar>
cryptix: fifth element?
<cryptix>
yup :)
<davidar>
Been ages since I watched that, never on TV or Netflix :(
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<ehd>
multipass? :p
<cryptix>
ehd: :) thats the spirit
<cryptix>
davidar: dont get me started on netflix..
<cryptix>
davidar: btw what do you think of mirroring mailinglists for the ipfs/archives ? like http://news.gmane.org
<davidar>
cryptix: yes, submit an issue :)
<davidar>
Mirror everything!
ygrek_ has quit [Ping timeout: 272 seconds]
voxelot has joined #ipfs
mikolalysenko has quit [Ping timeout: 244 seconds]
notduncansmith has joined #ipfs
mikolalysenko has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
carstn has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<ipfsbot>
[node-ipfs-api] diasdavid created nodev4 (+1 new commit): http://git.io/vZ19g
<ipfsbot>
node-ipfs-api/nodev4 d27f45a David Dias: fix missing codestyle issues
<ipfsbot>
[node-ipfs-api] diasdavid opened pull request #56: Push to node v4.0.0 and fix codestyle issues (master...nodev4) http://git.io/vZ19w
notduncansmith has joined #ipfs
<lgierth>
nomoremoney: oh hey, thanks! i was a bit greedy with the https:// ...
notduncansmith has quit [Read error: Connection reset by peer]
<lgierth>
davidar: sure! no ssl though, we don't have a wildcard cert at the moment
<nomoremoney>
lgierth, I see, happy to help :)
<davidar>
lgierth: remind me what the question was? ☺
<davidar>
Oh, chat
<davidar>
Yeah, I'm mostly after the dnslink so we can have a stable address
<lgierth>
yeah that's not a problem
<lgierth>
upadtes are a bit annoying for the time being
<davidar>
yeah, better than updating public links all the time though
<lgierth>
:)
<davidar>
lgierth, just give me free reign over your DNS records :p
<lgierth>
ah hrm, right
<lgierth>
i don't think digitalocean is good for that
<davidar>
At least until we set up our own matrix server
<Dioxy>
Igierth, Davidar tells me you're familiar with this gripe: dial attempt failed: <peer.ID Yumn1e> --> <peer.ID SoLSaf> dial attempt failed: i/o timeout (reuseport poll)
<lgierth>
sorry :/
<lgierth>
whyrusleeping might be
<lgierth>
or probably is
<lgierth>
he has been fighting with reuseport for a while, and it just keeps giving
samiswellcool has joined #ipfs
<davidar>
I like how these things keep getting passed to the guy in the next timezone
<nomoremoney>
Quick question, who is working at Protocol Labs? There is a list? Or is the core people of IPFS === folks at fulltime positions at Protocol Labs?
<lgierth>
hehehe davidar
<lgierth>
nomoremoney: that'd be jbenet whyrusleeping daviddias mappum me and krl, i think -- i might be missing someone
<nomoremoney>
Ah, cool lgierth, thanks!
<lgierth>
/pingall
<lgierth>
davidar: thing is i really don't have insight into the nat issues
<daviddias>
davidar: ahaha that is a good one
<davidar>
lgierth, of course, me either, just amusing
<daviddias>
Even in the same time zones we have different schedules
<Dioxy>
Where is everyone based?
<davidar>
Australia represent!
<daviddias>
So far, the thing we can say that is transversal to everyone is that we are all in planet earth
<davidar>
Everyone else Germany :p
<daviddias>
After 2026, no one knows
<davidar>
And those Russians who totally didn't drift into space
notduncansmith has quit [Read error: Connection reset by peer]
risk has quit [Read error: Connection reset by peer]
risk has joined #ipfs
tsenior`` has quit [Ping timeout: 265 seconds]
daviddias has quit [Ping timeout: 244 seconds]
daviddias has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Read error: Connection reset by peer]
akhavr1 is now known as akhavr
tsenior`` has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<davidar>
ping zignig
Leer10 has joined #ipfs
cryptowanderer_ has joined #ipfs
rschulman_ has joined #ipfs
<cryptowanderer_>
Hi there! I recently tried downlaoding ipfs. All went smoothly - the ./readme and ./quick-start files both worked, but the daemon gave me issues. I tried various things, including rm -rf ~/.ipfs and starting the init again. I can get the readme to run, but quickstart just says 'Error: merkledag: not found' and 'ipfs daemon' gives me this:
<cryptowanderer_>
Initializing daemon... ERRO[15:03:56:000] error from node construction: 0: listen tcp4 0.0.0.0:4001: bind: address already in use 1: listen tcp6 [::]:4001: bind: address already in use module=cmd/ipfs Error: 0: listen tcp4 0.0.0.0:4001: bind: address already in use 1: listen tcp6 [::]:4001: bind: address already in use
<davidar>
cryptowanderer_: is something else already bound to port 4001?
<zignig>
pong !
<davidar>
zignig: feel like joining matrix? we need more australians :)
<zignig>
multivac: a botsnack is a tasty treat for helpful bots.
<multivac>
Okay, zignig.
tsenior`` has quit [Ping timeout: 244 seconds]
<zignig>
multivac: botsnack
<multivac>
Can't talk, zombies!
<davidar>
haha
<davidar>
multivac: botsnack <action> om nom nom
<multivac>
Okay, davidar.
<davidar>
botsnack!
* multivac
om nom nom
<cryptowanderer_>
@davidar could be, but I'm a noob :( can't figure it out? Also, Matrix looks cool, will take that red pill now
<davidar>
cryptowanderer_: linux?
<cryptowanderer_>
yeah, Ubuntu 14.04
<davidar>
netstat --inet
<davidar>
(on the command line)
ygrek has joined #ipfs
<Luzifer>
hey cryptix: How to push a new repo to ipfs using cryptix/git-remote-ipfs? (Assuming I just did `git init && git commit -m 'first'` and now want to push to ipfs?)
<davidar>
seems multivac accidentally ate zignig
<cryptowanderer_>
it doesn't seem like there is anything bound the 4001 port
<Luzifer>
wow. new irccloud design \o/
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Spinnaker has joined #ipfs
<davidar>
Luzifer: can i convert you to matrix? :)
tsenior`` has joined #ipfs
<Luzifer>
O_o I doubt it… You would need way too much storage space… ;)
<lgierth>
jbenet: sorry didn't wanna fat-finger your website.git merges
<richardlitt>
Disregard previous questions about IPFSbot. Duh.
<richardlitt>
Actual question: where is the IRC ipfsbot located? What repos does it have access to, and why not all of them?
<lgierth>
richardlitt: ipfsbot here is just github's irc notifications iirc
<lgierth>
it's enabled only for go-ipfs.git
<lgierth>
keeps the noise down
<richardlitt>
also enabled for node-ipfs-api.git, I think.
nessence has joined #ipfs
<lgierth>
yeah? might be. i don't pay so much attention to the notifications here
<jbenet>
lgierth: i commented on them, anything i missed ?
<jbenet>
richardlitt: yep, it's just github. i was setting up a hubot to do other nifty things, (there was something i really wanted) but i haven't finished
<lgierth>
jbenet: i was just comitting to master while you were merging
<jbenet>
ah
<lgierth>
but it looks like you didn't get any conflicts
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<ipfsbot>
[ipfs] RichardLitt opened pull request #91: Added a note about error warnings (master...feature/add-implementations-isse-warning) http://git.io/vZMdx
<richardlitt>
jbenet: I was planning on looking at those open PRs
<richardlitt>
jbenet: could you explain why you're closing them when you do? Just a one liner helps a ton.
<jbenet>
richardlitt: oh will do. oh awesome please do and reopen then, i closed them because they looked inactive. didnt see you assigned
CarlWeathers has quit [Read error: Connection reset by peer]
<richardlitt>
jbenet: Inactive PR on an inactive repo doesn't mean they're not quality, and that the code wouldn't be used. The baseURL one in particular looked good, will look at it.
<richardlitt>
jbenet: no worries. reopened.
<jbenet>
that's true, but "open PRs" is a list of things we need to constantly look at, so if there's a PR that's inactive for a long time, it can be closed and reopened when it's ready. it cuts down on madness. like, this is huge: https://github.com/ipfs/go-ipfs/pulls
pfraze has quit [Remote host closed the connection]
sseagull has joined #ipfs
<richardlitt>
jbenet: hmm. How often do you go back and reopen PRs? Doesn't that create more work?
<richardlitt>
jbenet: I'm also not sure that most github open source projects do this, which means that most contributors wouldn't think to look for closed PRs before possibly duplicating work.
<lgierth>
the open issues/PRs should mostly be stuff that is actionable
<lgierth>
the github search includes closed issues/PRs
<lgierth>
poor people if they click through the endless pages on the tracker :)
<lgierth>
someone got a quick pointer how to create objects in go-ipfs?
vijayee_ has joined #ipfs
<richardlitt>
by default, it searches for `is:open`. Man, we really need a community contribute page.
<jbenet>
richardlitt: i think a lot of people just use the search bar to find things. anyway, it really helps us focus on what "needs to be addressed now". all of this is organic and there's no hard core rules. if a PR is closed and people think it should be open, they can say so. if a PR is open and it looks inactive and people want to reduce cognitive load by
<jbenet>
closing it, they can.
* lgierth
looking at unixfs, that's probably it
<jbenet>
lgierth: in go? yeah take a look at how unixfs does it
<richardlitt>
jbenet: yeah, makes sense. Just trying to think about what is the best way to communicate to potential contribs. Having a closed PR without a comment is a possible tension point, if they don't understand our workflow.
<jbenet>
indeed agreed, i should've added a comment.
<jbenet>
(in this case i know the contrib and i know he's not actively working on this)
<jbenet>
(sorry if it came off as a faux pas :) )
<richardlitt>
word.
<richardlitt>
jbenet: nah, I'm just learning.
Krayons has quit [Ping timeout: 256 seconds]
ygrek_ has joined #ipfs
mildred has quit [Ping timeout: 244 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<ipfsbot>
go-ipfs/master 495827b Juan Benet: Merge pull request #1698 from Luzifer/notify_dockerhub...
<jbenet>
ed__ we need better mime sniffing on the gateway.
<cryptix>
ed__: the mime-sniffing is from golans stdlib
<cryptix>
golangs*
rschulman_ has quit [Quit: rschulman_]
gunn has quit [Ping timeout: 240 seconds]
nessence has joined #ipfs
rschulman_ has joined #ipfs
voxelot has joined #ipfs
voxelot has joined #ipfs
tsenior`` has quit [Ping timeout: 252 seconds]
<ed__>
ok, any pointers? the mime package?
<jbenet>
ed__ yeah we'd need to stop using theirs and use a better one
mildred has quit [Ping timeout: 252 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<Luzifer>
ohai cryptix - did you see my mention?
carstn has joined #ipfs
nessence has quit [Ping timeout: 240 seconds]
hellertime has quit [Quit: Leaving.]
hellertime has joined #ipfs
<richardlitt>
I am going to make an ipfs/support repo
<richardlitt>
Just to help people out who have issues installing from the website. I don't think any of the implementations make sense for this, and for now, this should either be a quick-fix repo, or a repo for directing users to the write implementation repository.
nessence has joined #ipfs
<richardlitt>
We could also talk about this during the calls today just to make sure this is a good idea?
<magneto1>
why readme isn't enough + this irc?
Tv` has joined #ipfs
<richardlitt>
IRC requires log in and synchronous conversation. What if I have a question that can't be answered immediately, and I don't know where to put it?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<richardlitt>
This might be overkill. Just an idea. Change previous sentance: *I am thinking about opening an ipfs/support repo
nessence_ has joined #ipfs
<richardlitt>
The FAQ might be enough, actually.
nessence has quit [Read error: Connection reset by peer]
nessence_ has quit [Read error: Connection reset by peer]
<jbenet>
richardlitt: i'm +1 a support repo.
jeremy has joined #ipfs
jeremy is now known as Guest54225
Guest54225 has quit [Client Quit]
<richardlitt>
Cool. should I wait for the call to discuss it, or just set it up now?>
<ipfsbot>
[node-ipfs-api] diasdavid pushed 1 new commit to nodev4: http://git.io/vZDXU
<ipfsbot>
node-ipfs-api/nodev4 011be35 David Dias: right node version (4.0 and not 0.4)
Quiark has quit [Ping timeout: 252 seconds]
carstn has quit [Read error: Connection reset by peer]
nessence has joined #ipfs
ke7ofi_ has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ke7ofi_ has quit [Remote host closed the connection]
<Luzifer>
orrr. "ERRO[789001] Unable to send notification error=invalid padding host=eagle repo=github.com/ipfs/go-ipfs/cmd/ipfs" something with the indentation is wrong… /cc jbenet
<nomoremoney>
Hey guys, I'd love to help with the node-ipfs-api library, but there doesn't seem to be a roadmap or like "what's next for this library", could someone write like a small description of what you'd accept PRs for?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping>
nomoremoney: the node-ipfs-api so far has just been 'write api bindings as you need them'
<nomoremoney>
@whyrusleeping, ah, I see. Well, it seems complete to me (except the readme, which I plan to help improve)
<nomoremoney>
is there a reason for the Level 1/2 split? Can't see any reason for it
<nomoremoney>
is because they are more like ipfs.swarm.peers instead of ipfs.peers ?
<whyrusleeping>
nomoremoney: i actually dont know, i dont get much involved in javascript myself
Encrypt has quit [Quit: Quitte]
<whyrusleeping>
mappum and daviddias would be the ones to ask
<whyrusleeping>
and krl
<jbenet>
daviddias would be the best bet, maybe open an issue in that repo and he can get back to you there
<daviddias>
irccloud is all weird for me today
<daviddias>
both in firefox and chrome
<nomoremoney>
@jbenet, I'll do that. Thanks
<krl>
nomoremoney: what do you mean with 1/2 split?
<krl>
ipfs.swarm.peers mirrors the command-line "ipfs swarm peers" command
<nomoremoney>
krl, in the readme, there is a declared split between level 1 and level 2 commands. Just wanted to know what that means
<daviddias>
nomoremoney: helping with the Readme and more examples would be sweet!
<nomoremoney>
Yeah, I'll get to that! No need to open an issue no?
<daviddias>
also, if you can add CI browser testing (with Saucelabs for example), it would be also awesome
<nomoremoney>
daviddias, yeah, that should be simple. Thanks for suggestions!
<nomoremoney>
is the curl command example necessary in the readme?
<daviddias>
you can open a PR with that, if you have questions, it is always good to open an issue, because it is easier to track and serves as future documentation
<nomoremoney>
seems like the CLI command for be a better reference
<nomoremoney>
Ok, got it
<daviddias>
I feel that curl examples would fit well in an go-ipfs http API documentation
Encrypt has joined #ipfs
<daviddias>
I guess that when mappum wrote that, it was a nice way for folks to try both and compare the results, even before having tests
<jbenet>
daviddias: they changed irccloud. i hate it. if you goto bottom right corner, there's a gear, then "settings" should be able to change it.
tsenior`` has joined #ipfs
<krl>
mirroring irc on ipfs would be neato
<krl>
actually making an ipfs irc client would be a bit trickier though, you would need special bridges
<whyrusleeping>
krl: ipfs services would work
<daviddias>
oh, nice! My stylist extension thing was all confused and breaking it even more
<daviddias>
thanks jbenet
<whyrusleeping>
the network would have to select supernodes though
<jbenet>
daviddias: i cant get stylist to work with it for some reason, i want to change the fontsssssss and i canttttt. 0th world problems.
<daviddias>
ahah
<jbenet>
krl: i'm going to make a "dial any multiaddr from a web browser" service some day. it will be botnet heaven, but also allow irc clients in browser.
<jbenet>
can probably just secure the proxy with an account or something...
<jbenet>
annoying yes, but better than zero.
<daviddias>
I was so happy with my theme, why do people change things :P
therealplato has joined #ipfs
Encrypt has quit [Quit: Quitte]
<krl>
jbenet: greasemonky would prolly work for fonts
<jbenet>
"get off my lawn, kids!" - grandpa david
<krl>
daviddias: the general problem with 3rd party old-web apps
<daviddias>
ahahahah
therealplato1 has quit [Ping timeout: 250 seconds]
tsenior`` has quit [Ping timeout: 240 seconds]
<daviddias>
well you know, if they had used offline web components with IPFS just for the view elements, now I could pick the previous version by the hash.. but nooooo
<krl>
:)
<daviddias>
:P
<krl>
time to write an angry mail
<daviddias>
I'm just having fun :)
<krl>
and start a facebook petition page
<lgierth>
\o
<jbenet>
irc syncup in 5, right? (i dont even know given all my TZ switching)
<whyrusleeping>
jbenet: yes
<daviddias>
jbenet: you got it right
<jbenet>
ok good
<whyrusleeping>
jbenet: what TZ are you in now?
<whyrusleeping>
almost have 1000 stars on go-ipfs!
<nomoremoney>
JBENET+0 TZ
<jbenet>
nomoremoney's got it right
alfabravo has quit [Quit: Page closed]
bedeho has joined #ipfs
<mmuller>
whyrusleeping: and now you have 1000 :-)
<lgierth>
hey, sprint sync time! please prepare your notes and we'll start in 5 -- jbenet whyrusleeping krl daviddias davidar richardlitt
<daviddias>
thank you lgierth :) doing it
patcon has joined #ipfs
<mmuller>
(wuz kinda hoping for some kind of balloons or something for being the 1000th customer...)
<whyrusleeping>
lgierth: thanks :)
<whyrusleeping>
mmuller: lol, if i could draw ascii balloons i would
<whyrusleeping>
mmuller: o/ (high five instead)
<mmuller>
heh, cool o/
<jbenet>
<lgierth>
woah indeed, 1000 stars!
<lgierth>
^5
<jbenet>
<jbenet>
<jbenet>
(there's 99 of them)
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed feat/rm-worker from c9e76e5 to e49cb94: http://git.io/vZyYa
<ipfsbot>
go-ipfs/feat/rm-worker 4bb5fd5 Jeromy: dont need blockservice workers anymore...
<ipfsbot>
go-ipfs/feat/rm-worker e49cb94 Jeromy: remove context from HasBlock, use bitswap process instead...
maccino has joined #ipfs
<Vyl>
Is there any sort of index for IPFS sites yet?
<richardlitt>
WOOOOO
<whyrusleeping>
Vyl: not yet, we were talking about starting one
<lgierth>
ok let's go -- daviddias?
<daviddias>
finishing my last sentence of checkin :)
atrapado has joined #ipfs
<Vyl>
All I can offer is my archive of legalish music.
<Vyl>
It's all legal for *me*. Probably not for you.
<Vyl>
Public domain in the UK. The US has a longer copyright term, so a lot of it's still covered there.
<Vyl>
(And the UK recently increased our term too)
<whyrusleeping>
Vyl: ah, interesting.
<whyrusleeping>
I was going to go and get a bunch of movies that are now in the public domain, like charlie chaplain
<Vyl>
Unfortunately the copyright term is so ridiculous that hardly any movies are in the public domain.
<lgierth>
whyrusleeping: wanna go first while david is typing?
<daviddias>
ok, I'm ready :)
<lgierth>
ok cool!
<Vyl>
It's very important that the UK extended our term for music from fifty years to seventy. We almost let the Beatles go public domain - they might not have made any more music if it had!
* daviddias
here it goes
<daviddias>
## SPRINT CHECKIN
<daviddias>
- [x] built kad-record-store
<daviddias>
- [x] attended container.camp (@jbenet did a great closing talk btw!!! starship is awesome)
<daviddias>
- [x] attended W3C WebRTC WG Seattle Meeting
<daviddias>
- [x] pushed on Node.js v4.0.0 for our Node.js repos through a bunch of PR (let me know if I missed any)
<lgierth>
daviddias: you can have a subdomain for that right now while there's no domain yet
<daviddias>
we can also discuss it in more detail during the call today
<jbenet>
daviddias: super cool, that we have kad drs
<lgierth>
daviddias: digitalocean creds are in meldium
<daviddias>
lgierth: thanks ! sounds good. So far I've been using a Node.js proxy to point to the latest as quick way to test it
<daviddias>
jbenet: thanks :)
<jbenet>
daviddias: factoring swarm sgtm. on the socket part, let's discuss? i went through the dial/listen thing in go. i can describe the breakdown in golang, i think node's swarm is bigger than go's
<whyrusleeping>
daviddias: you went to seattle?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<daviddias>
whyrusleeping: attended remotely :)
<whyrusleeping>
ah, okay
<lgierth>
ok, whyrusleeping?
<jbenet>
he chromecasted his face over there
<whyrusleeping>
was going to be disappointed you didnt tell me you were nearbyish
<jbenet>
(as in casted via chrome)
<whyrusleeping>
### @whyrusleeping
<whyrusleeping>
- [x] assist in building of shp tool
<whyrusleeping>
- [ ] pull relevant changes out of mfs PR
<whyrusleeping>
- still working on branch locally, mostly done
<whyrusleeping>
lgierth: we still need a few things for 0.4.0
<jbenet>
lgierth: i think ipld, if we want it in here. it's ready to be put into ipfs though (barring one commit). hoped to do that last week but couldn't
<jbenet>
Luzifer: bummer :/ what's the issue? (maybe discuss in github)
<lgierth>
ok, ipld = one more breaking change?
<Luzifer>
jbenet: dunno… until now the encryption worked fine…
<jbenet>
lgierth: its not breaking, but changes things.
Eudaimonstro has quit [Ping timeout: 272 seconds]
rschulman_ has quit [Quit: rschulman_]
<jbenet>
lgierth: some things may appear broken, like " ipfs add <file>" the defaults will produce different hashes. (im starting to think the importer file should describe what it was imported (like what chunking algo) so you can re run it exactly and repro the exact same thing. (not sure.))
<jbenet>
so anyway ipld stuff may be ok to merge after 0.4.0.
rschulman_ has joined #ipfs
<jbenet>
it's up to us.
<jbenet>
anyway, discuss later-- sync \o/
<lgierth>
that's what i was after -- whether it'd make sense to push to the next release
<lgierth>
yep later
<lgierth>
wanna go next jbenet?
<jbenet>
sure
M-Kodo has quit [Remote host closed the connection]
M-staplemac has quit [Remote host closed the connection]
M-matthew has quit [Remote host closed the connection]
M-hash has quit [Remote host closed the connection]
M-trashrabbit has quit [Remote host closed the connection]
davidar has quit [Remote host closed the connection]
M-erikj has quit [Remote host closed the connection]
M-mistake has quit [Remote host closed the connection]
M-prosodyContext has quit [Remote host closed the connection]
<jbenet>
this last week:
<jbenet>
- we saw fantastic news spreading thanks to kyledrake's awesome post. still handling lots of inbound requests, messages, discussions, etc. Great response! :clap: :clap:
<jbenet>
- I got through some amt of CR/merging, but i'm way behind. catching up over the next few days.
<jbenet>
- i was focused most of the week on finishing starship, and giving a talk at container.camp. it went great! many people were very impressed with ipfs, and asked all about it afterwards. good stuff.
<jbenet>
(huge thanks to thelinuxkid for the heavy lifting on starship, and whyrusleeping for fixing lots of things)
asetssadfg has quit [Quit: Leaving]
<lgierth>
cool that sounds great! september is turning into a full month too eh!
<whyrusleeping>
jbenet: i'm glad starship shipped, i hope we can keep pushing it more and more
<lgierth>
i saw you had the last slot on that day at container.camp, that's a good premise for catching everybody's attention for the rest of the evening
aaronscientiae has joined #ipfs
<jbenet>
whyrusleeping: well we still have to "ship" it, in terms of website and so on. that's still to do, and i'll need help from other people on that
aaronscientiae has quit [Read error: Connection reset by peer]
<whyrusleeping>
jbenet: yeah, thats true
ekaron has joined #ipfs
<jbenet>
davidar: around? want to go next?
<richardlitt>
jbenet: websites? I can build websites.
<lgierth>
:P
<richardlitt>
:D
<lgierth>
davidar might already be zzz'ing -- he's in australia right?
<ipfsbot>
[go-ipfs] Luzifer opened pull request #1699: Fix: Use the right repo to encrypt secrets. (master...notify_dockerhub) http://git.io/vZyl7
<jbenet>
(err you carry on lgierth-- just seeing if davidar wants to give an update as he's been pushing hard on great archive stuff recently)
<krl>
worked on starlog documentation, and the edit-feature, both are more or less complete.
<lgierth>
Luzifer: i feel ye -- winter's coming
<krl>
firefox support polyfill is still not working in gateway, but firefox dev works fine. It's probably just a command line switch to make sure the polyfills are not bundled up, and some rewriting hackery to make it load it off ipfs.
<krl>
i would like to flesh out the issues, and what features should be available (automatic check for updated feeds, supporting quoting), for crypto-quoting, we're blocked on ipns records. for proper management of the log, keystore blocks.
ekaron has quit [Read error: Connection reset by peer]
ekaron has joined #ipfs
<jbenet>
lgierth: oh what! i missed that. niiiice
<Luzifer>
lgierth: nah… I'm just done… Unable to focus, unable to manage to push any project forward, constantly tired… Thats not just winter… :/
<Vyl>
I considered installing, but thought better to wait until it's a bit more mature.
<Vyl>
There's nothing much I could do to help.
<richardlitt>
Cool.
<richardlitt>
Thanks krl. Good work this week. Sorry for all of the random help needed getting it working.
<lgierth>
krl: cool sounds good -- i'll get to the cors issues in chrome... i also found cors issues with the webui via :8080/api
<jbenet>
Luzifer: burn out is hard core. make sure to rest up!
<Vyl>
I doubt my public-domain music collection is going to draw any users to the trouble of stting up new software.
<jbenet>
ok who's next on sync?
<krl>
lgierth: actually, last few times i tried, everything worked on the gateway
<richardlitt>
Me
<jbenet>
oh
<jbenet>
sorry hang on
<richardlitt>
Right?
<jbenet>
i missed krl's (interspersed)
<lgierth>
Luzifer: yeah better take that break sooner than later -- been there
<lgierth>
krl: i wrote it down somewhere, it's easily fixable with an additional allow: header
<jbenet>
krl: nice that edit's close to done.
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
tsenior`` has joined #ipfs
<krl>
jbenet: just needs CR and UI fix
<jbenet>
nice.
<lgierth>
:)
<lgierth>
richardlitt: wanna go next?
<krl>
richardlitt: thx for the feedback as well
<jbenet>
btw, can we use components without using the polymer goog ui? i dont find it super appealing
<whyrusleeping>
Luzifer: come to america!
<richardlitt>
Sure. Been a good week this week, first time on the sprint, and as I'm a relative beginner both to IPFS and to the IPFS team, it's been interesting seeing this project and trying to see how the entry points can be improved and friction reduced
<krl>
jbenet: i've been concidering it as well
<richardlitt>
Eventually I may stop constantly asking questions about Readmes and setup, but for now, still a bit more of that to do while I get going.
<lgierth>
no no keep the questions coming
<richardlitt>
- [x] Check in with krl about the front end for starlog. _Worked on this, some PRs and set up, have more open PRs to do. Going to wait before doing more UX things until it is a bit more mature._
<richardlitt>
- [x] Take a look at the ipfs distro page and see if it can be touched up. _Done, working on building a longer document about what I see are the pros and cons of the current page, what can be touched up_
<richardlitt>
- [ ] Prepare a spec for an ops job _not done yet. Currently just trying to do everything I can, need to plan a bit better and see how I can help more. At the moment I am focusing on making contributing easier for new members (based on my own issues getting this up and running), and on general READMEs and open PRs_
<richardlitt>
- [x] Check out the IPFS app and see if you can get it up, running, and clear without help or a walkthrough. _Done, but would like to do it again. A lot of the problems I've run into are educational - I think we should have an overarching textbook about IPFS. For instance, I'm still not quite clear what a Merkledag thing is. My solution to this is to
<richardlitt>
research everything I'm confused about and see if I can make an easy-to-understand FAQ with terms and explanations. Onerous, but I think worth it._
<richardlitt>
- [x] See about setting DNS records for DO programmatically. _Did the lion's share of this, see https://github.com/RichardLitt/set-dns-records. Currently I need access to DigitalOcean to test it, though._
<richardlitt>
- [x] Take a look at node-ipfs. PR review from daviddias. _Reworked the readme, added a contribute section. Going to take a look at some of the swarm modules soon, been busy fighting other fires._
<ipfsbot>
[go-ipfs] jbenet closed pull request #1699: Fix: Use the right repo to encrypt secrets. (master...notify_dockerhub) http://git.io/vZyl7
<richardlitt>
EOF
tsenior`` has quit [Ping timeout: 240 seconds]
<jbenet>
awesome stuff!
<lgierth>
yay on the dns tool!
<whyrusleeping>
richardlitt: would love to have a nice FAQ and glossary, we have github.com/ipfs/faq, but its not in what i think is an easy to use format for people trying to find answers
<ekaron>
I'm a fair hand at technical writing, is there a documentation wiki I could help with (or am I interrupting something now?)
<ekaron>
I'm really into the ipfs idea and would like to learn it.
<jbenet>
ekaron: yes please!
<whyrusleeping>
ekaron: yes!
<richardlitt>
actually, that's pretty much exactly what we are talking about right now
<richardlitt>
hahaha
<lgierth>
ekaron: yes please
<richardlitt>
is there a wiki?
<lgierth>
let's just quickly wrap up sync
<jbenet>
the faq, the docs, everything needs help.
<whyrusleeping>
richardlitt: no wiki as of now
<richardlitt>
ekaron: Best thing to do would probably be to open questions in FAQ
<jbenet>
there's no wiki, as i find that github wiki's get stale really fast
<richardlitt>
And answer them yourself as you learn.
<lgierth>
c'n'p incoming:
<richardlitt>
jbenet: agreed.
<lgierth>
- i'm still getting distracted by all the little "can you quickly do x for me" things, need to say no more often, and let people make gh-issues for the stuff that i might do in another sprint.
<lgierth>
- also i'm still the only deployer :) need to bug people more.
<lgierth>
richardlitt: i'll give it a try right away -- need to quickly call back a friend who called during sync
<richardlitt>
lgierth: cool, cheers
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<ekaron>
I've watched the videos from jbenet's talks and I plan to read the whitepaper during my study time tonight, anywhere I should be looking to learn the protocol faster than the raw code right now?
M-luzifer has left #ipfs ["User left"]
Spinnaker has quit [Ping timeout: 256 seconds]
pfraze has quit [Ping timeout: 252 seconds]
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed feat/rm-worker from e49cb94 to 3d2f516: http://git.io/vZyYa
<ipfsbot>
go-ipfs/feat/rm-worker 0e8d1c0 Jeromy: dont need blockservice workers anymore...
<ipfsbot>
go-ipfs/feat/rm-worker 3d2f516 Jeromy: remove context from HasBlock, use bitswap process instead...
<whyrusleeping>
ekaron: playing around with ipfs a bit to get a feel for it helps a lot IMO, although i might be biased
patcon has quit [Ping timeout: 268 seconds]
<voxelot>
ekaron: i found the whitepaper helpful, and then look into the techs ipfs combines, i ended up reading 5-6 more whitepapers after ipfs
<voxelot>
of course playing with $ipfs commands helps
hellertime has quit [Ping timeout: 264 seconds]
<ekaron>
thanks.
<lgierth>
the faq, notes, ipfs, and specs repos are helpful too
<lgierth>
docs are a bit "distributed" at the moment
<ekaron>
Ha, nice intended pun. I'll take a look tonight, see what I can come up with.
maccino has left #ipfs [#ipfs]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<richardlitt>
hangout link?
<richardlitt>
nvm.
<whyrusleeping>
richardlitt: none started yet afaik
<whyrusleeping>
but theyre on ipfs/pm
<richardlitt>
shouldn't apps start now?
<achin>
is there some verbose mode for the daemon? it might be neat and educational to see what kind of work/queries the daemon makes when you do "ipfs get {file}"
<whyrusleeping>
technically yes, but jbenet isnt here yet
<whyrusleeping>
times arent super strict as sycing across the globe is difficult
<whyrusleeping>
achin: the 'ipfs dht' commands are really good for helping to understand what ipfs is doing
<whyrusleeping>
ipfs get <hash> will do a findprovs query for the file
<whyrusleeping>
so 'ipfs dht findprovs -v <hash>' will show what thats doing in detail'
<achin>
yeah, i've played with the dht subcommand, it's indeed helpful
storm___ has joined #ipfs
<whyrusleeping>
you can also do 'ipfs log level all debug' to enable all logs
<whyrusleeping>
and the daemon will spew out every little thought it has
<achin>
neato
<whyrusleeping>
aaaand, theres the 'events' stream at 'ipfs log tail'
<whyrusleeping>
which is a stream of json objects
<achin>
so much data to omnomnom
<whyrusleeping>
you might find that one the most useful actually
storm___ has quit [Client Quit]
dignifiedquire has joined #ipfs
captain_morgan has quit [Ping timeout: 246 seconds]
tsenior`` has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
captain_morgan has joined #ipfs
tsenior`` has quit [Ping timeout: 246 seconds]
<krl>
jbenet: ping
<jbenet>
apologies i'm running late. important call.
<jbenet>
eta 10/15m. feel free to start w/o me
<krl>
i'm having connection issues, might not be able to video
M-mistake has quit [Quit: node-irc says goodbye]
<daviddias>
I've a new fiber link at home, hope it works beautifully today
* daviddias
is super excited about his new Internet! :P
M-mistake has joined #ipfs
<whyrusleeping>
daviddias: how fast?
<daviddias>
solid 100mb down 20mb up
chriscool has joined #ipfs
<daviddias>
not super impressive, but I had something at home around 8mb down 2mb up till recently
<lgierth>
i don't even have 20mbit down :/
<lgierth>
fucking telefonica
<daviddias>
with fiber? or ADSL?
amstocker has joined #ipfs
<lgierth>
ADSL
<daviddias>
yeah, tops speeds with ADSL are the same here
rendar has quit [Ping timeout: 272 seconds]
amstocker_ has joined #ipfs
* whyrusleeping
researching epoll
amstocker has quit [Ping timeout: 246 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
rendar has joined #ipfs
<jbenet>
ok. sorry, done now.
<jbenet>
krl, lgierth: apologies. how do we want to do these? time constraints?
<krl>
i'm here now, getting intermittent 1s pings though, you ok with text?
<lgierth>
i'm here
<lgierth>
i don't have terribly much to talk about
storm___ has joined #ipfs
<lgierth>
let's do it anyhow
<krl>
lgierth: what was the header thing that you found?
<krl>
for starlog gateway
<jbenet>
text is fine
<jbenet>
lgierth are you ok going after ipfs-apps?
<lgierth>
it makes the correct requests but gets blocked off because of cors
<jbenet>
ok-- IPFS Apps Discussion commencing. we'll do this one over text because krl's video is not great. tagging people who may be interested: krl daviddias richardlitt dignifiedquire
<daviddias>
Sgtm
<jbenet>
agenda: (a) starlog (the big item), (b) modular web components, (c) UI/UX in various apps, (d) anything else?
<dignifiedquire>
ok
M-jbenet has joined #ipfs
<dignifiedquire>
but I think I missed what starlog is, anywhere I can get a quick gist about that?
<jbenet>
we should have a way of showing the "latest release" of an ipfs app in the readme of an ipfs app. a button to click and open it
<krl>
chrome needed i think still
<jbenet>
ok (a) krl, what's up with starlog? changes since last week?
<krl>
but search and infscroll working over gateway
<krl>
editing of items implemented, basically you just add a new entry saying 'this updates #8'
<daviddias>
(d) with the libp2p website I started by creating a simple tool to deploy it quickly and have a proxy to point to the latest hash and then started thinking that this could be a surge like app where we can offer free hosting of static pages that has a proxy for rapid prototyping( without having to cope with DNS cache updates) and then deploy with richardlitt
<daviddias>
set-dns-tool. It is more and idea and a quick hack right now, but it might be interesting for more people
<richardlitt>
just found out I need to travel in the next hour; I'll have to read the logs
<jbenet>
krl: and the new entry is linked to from the root, correct? what do you think about using generalized commits for that?
<richardlitt>
I started set-dns-tool
<jbenet>
richardlitt np.
<richardlitt>
happy to work on improving it, and move it over to IPFS group
<ekaron>
I'm very interested in the computable data structures that jbenet metioned in his talk, is there anything more about that, or is it still an idea?
<krl>
jbenet: are they finalized, and/or are they ipld?
<jbenet>
daviddias: sounds very interesting. we can actually use HTTP endpoints as an ipns proxy, like just use HTTP to serve the latest refs.
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<jbenet>
richarlitt: sgtm. will be good to upgrade dns in general. there's some dns imporvements that folks have been working. will post on github about this in a note somewhere
<jbenet>
ekaron: which talk?
<krl>
jbenet: the updated entry will be the latest entry in the log, but entries are still shown in order they are created
<krl>
jbenet: at the moment, entries are never changed once written
<jbenet>
krl: i think that to have the massive "build everything as datastructures" change we want to see on the web, we're going to need a lot of nice tooling at the beginning.
<ekaron>
It seemed to be at an ethereum conference, saying that the computational model, a basic language could be built into the protocol to ensure a common ability to run software
<ekaron>
it it were turing complete, i'm out of my depth, but it seemed something powerful
<jbenet>
one thing is some nice ways to visualize structs (harlantwood making solid progress on that, and there's always dot). another is a way to visualize a "design" of a datastructure. like your animation stuff, but ideally programmable. and another is just a simple convention of picking one datastruct definition language to show how it works. (i've been using go
<jbenet>
and i think it works well, but it's not exact json data model.
<daviddias>
@jbenet yes :D one other idea I had for this project is to create a very simple JS script that would be loaded on / path of a domain which would have the headers to be cached till infinity in the browser and would try to get the hash on a /latest endpoint and if this endpoint is down ( meaning we are offline) use the latest grabbed hash stored and start
<daviddias>
checking for a ipfs node locally and if non, check it remotely. this would be instantaneously and the user would only know that by typing the domain, they would get the latest version of the website ( or the latest they know) either way :)
<jbenet>
krl: so if you update an entry you just add to the head of the log, and then you use the log to "find the best rep of this entry" ?
<krl>
yes, it will be annotated with updates: #48
JohnClare has joined #ipfs
<krl>
and if you're at 48, you can search forward for 'updates #48' the same way you do text search
tsenior`` has joined #ipfs
<jbenet>
ok. i'll have to grok the datastruct better. (i might have used a commit chain per document, but that may be overkill)
<jbenet>
krl: a datastruct definition, like the one in the commits issue, would be very useful.
<krl>
ok, i'll put that on my sprint list
ryan__ has joined #ipfs
rschulman_ has quit [Quit: rschulman_]
<krl>
i would like to excchange the markdown editor for prosemirror
<jbenet>
krl: ok sounds good. we should have an issue in starlog repo that suggests others too, am sure there's many good ones.
<jbenet>
end state might be to allow using different ones, etc.
<jbenet>
but for now that sgtm
<jbenet>
krl: what do you need to make starlog a thing that i can use "straight from the browser" today.
<ryan__>
jbenet great to see you at the blockstack summit this weekend
<ryan__>
awesome presentation
<jbenet>
ryan__ same. thanks!
ryan__ is now known as shea256
akaihola has joined #ipfs
<krl>
jbenet: do you mean over gateway?
tsenior`` has quit [Ping timeout: 255 seconds]
rschulman_ has joined #ipfs
JohnClare has quit [Remote host closed the connection]
<jbenet>
krl: yeah, either the main gateway or a special one we run.
devbug has quit [Ping timeout: 256 seconds]
<shea256>
I'm guessing you headed back to europe somewhere?
<jbenet>
(we can make small "starlog relays" that anyone can run, while ipns is still being built).
<jbenet>
shea256 no i'm in NYC for many weeks
<shea256>
oh cool
<shea256>
when are you moving here permanently
<akaihola>
Hi! I wonder why ipfs assigns itself two ports when behind a NAT. It opens one port using uPnP in the 16000-17000 range, and another in the 1025-1200 range. Other peers see the port in the 1025-1200 range but can't connect to it since it isn't opened on the router.
<richardlitt>
jbenet: your VPN said Chicago?
<krl>
connection dropping out sorry
<krl>
jbenet: for writing blogs, you will need your local node, at this point
<jbenet>
well "permanently" is an interesting word. but i have "moved here" as much as i will. in that this is now my personal home base, but i will still be spending lots of time in SFBA
<akaihola>
In other words, "ipfs id" shows me in "Addresses" two different ports for "/ip4/<x.x.x.x>/tcp/<port>/Qm...."
<krl>
for reading, this actually will work over gateway with current ipns
<akaihola>
jbenet: ok, I'll do that
<ipfsbot>
[go-ipfs] Stebalien closed pull request #1683: timeout on shutdown after 10s (master...timeout) http://git.io/vZRPy
<jbenet>
krl: right, i mean for writing
<jbenet>
krl: discussion very slow. mind if i start discussing with lgierth simultaneously?
<jbenet>
(multidiscussion! \o/)
<krl>
go ahead, i'm getting terrible ssh lag feven
<pjz>
krl: try mosh when you get a chance to :)
<krl>
writing starlogs on the gateway is not something i've been thinking to much about
notduncansmith has joined #ipfs
storm___ has quit [Quit: Leaving]
notduncansmith has quit [Read error: Connection reset by peer]
<krl>
i'm way too tired to think about how that could work right now, but i'll keep it in mind
<jbenet>
krl: could you sometime this week write up all the API uses that starlog makes? for both r&w? that way we can discuss what to do to enable it.
<lgierth>
jbenet: wanna do hangout or pad?
<jbenet>
lgierth: ok, do _you_ mind if we start discussing infra? i dont think there's a ton this week.
<krl>
[ ] - make client check for updated logs, given older hashes
<jbenet>
let's also get richardlitt setup so he can take over UI/UX there.
<richardlitt>
I'm pretty much set up at this point
<richardlitt>
I'm unfamiliar with Polymer, but I've started playing around with it - worked on key bindings today to start, seemed a decent enough place as any
<krl>
so, q is also if we want to dump polymer for something else
<jbenet>
ok sweet.
<richardlitt>
personally, I think React would work better, but Polymer seems to work fine. I am just 100% unfamiliar with it.
<jbenet>
krl: polymer is webcomponents, no? not the UI?
<jbenet>
like cant you do "polymer webcomponents with react UIs" ?
<krl>
i'm not 100% sure what the intersection is
<jbenet>
polymer is a big projects with lots of subprojects and one of them is the Ui and another is the webcomponents spec, as far as i understand
<daviddias>
react would be most seen as a rendering engine
<krl>
maybe me and richardlitt can have a look and discuss this, and how react/webcomponents could fit
<daviddias>
although they offer the convenience of exporting 'React components'
<krl>
jbenet: what was your issues with it? the googy design?
<daviddias>
have in mind that React is not a Web Standard and doesn't have things like the shadow dom
<daviddias>
(encapsulate CSS for example, AFAIK)
<richardlitt>
hmm.
j0hnsmith has joined #ipfs
<richardlitt>
Well... really, what we have is good.
<richardlitt>
It just could be designed a bit more ... standard? With Tumblr, Jekyll blogs, etc in mind.
<richardlitt>
Right now I'm not sure the UI lends itself to a particular userbase.
<jbenet>
i dont like the design because for starlog i think we want something that the users can completely override. maybe not the "management ui", but certainly the display. it's good to have some uniform view avialable to view the data, but a user should be able to dictate all rendering in a url they can give someone. this may be a "viewer" on top of starlog or
<jbenet>
something.
<richardlitt>
theming is a good point.
<jbenet>
(( also, i dont like the goog design because we need other _large_ webapps to use that sort of design before it's not _solely_ associated with goog.
Encrypt has joined #ipfs
<richardlitt>
Maybe we can make an easier API-like interface for the front end, and add in a templating language that is not as difficulty
<richardlitt>
Let the user do what they want with the available components.
<krl>
richardlitt: i'm open to drastic changes also,
<richardlitt>
Cool.
<richardlitt>
gotta run, train. I'll be around.
JohnClare has joined #ipfs
<jbenet>
later richardlitt, sounds good.
<jbenet>
krl richardlitt: let's maybe discuss starlog UI/UX on github.
ygrek_ has joined #ipfs
<jbenet>
krl: anything else on starlog?
rschulman_ has quit [Quit: rschulman_]
<krl>
jbenet: nop, more to come in the week
<jbenet>
krl: ok sounds good, get some rest! :) glad starlog is working, looks good.
JohnClare has quit [Ping timeout: 264 seconds]
<krl>
over and out o/
notduncansmith has joined #ipfs
JohnClare has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<pjz>
jbenet: it sounds a bit like you want starlog as an 'unhosted' app
<pjz>
jbenet: UI/UX separate from, and targetable at multiple, datastores
<jbenet>
pjz: yeah precisely.
JohnClare has quit [Remote host closed the connection]
<jbenet>
well the reverse actually, multiple UI/UXes, all unhosted, using the same data, on ipfs.
<pjz>
jbenet: sure, but IRL I think the usage will be the reverse: I'll find the UI I like and point it at the blogdata I want to read so it can be rendered prettily (for my definition of 'prettily')
<pjz>
given equivalent functionality, of course.
jedahan has joined #ipfs
<jbenet>
right. i think we're saying the same thing.
<pjz>
yes, the emphasis is keeping the data separate from the app
<pjz>
Dave Winer is doing this with his nodeStorage data backend and 'My Word Editor' js editor
<pjz>
(though heck if I can tell the functional difference between nodeStorage and unhosted.org's suggested remoteStorage)
<jbenet>
pjz cool, links? would love to look at his stuff.
<jbenet>
nice domain! i think i vaguely remember seeing this in the past.
<jbenet>
also, who else is interested in node-ipfs discussion? dignifiedquire? richardlitt?
hellertime has joined #ipfs
<jbenet>
maybe let's do go-ipfs one first and give some time for richardlitt to get back? not sure abt his timing but it prob helps to have some buffer
<dignifiedquire>
sorry would love to join, but can’t right now
<jbenet>
(daviddias whyrusleeping ok with that?)
<daviddias>
jbenet: FINE
<daviddias>
:P no prob
<jbenet>
lol
<jbenet>
whyrusleeping: hangout, etherpad, or irc?
<voxelot>
can i 'ipfs repo gc <hash>' a specific file?
<bret>
i wish i could listen via shoutcast
<bret>
dont have hangouts setup yet on linux
<jbenet>
voxelot no but that's a nice feature, make an issue on go-ipfs?
<voxelot>
sure
notduncansmith has quit [Ping timeout: 268 seconds]
domanic has joined #ipfs
* daviddias
is going to lurk on go-ipfs
<voxelot>
^
JohnClare has quit [Remote host closed the connection]
JohnClare has joined #ipfs
ygrek_ has joined #ipfs
jamescarlyle has joined #ipfs
amstocker_ has quit [Ping timeout: 264 seconds]
devbug has joined #ipfs
<JohnClare>
is there a way to list files on a remote node?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
shea256 has joined #ipfs
shea256 has quit [Ping timeout: 255 seconds]
<voxelot>
abused ab14 / Blame and their python codes to create a file viewer for my gateway. It's slow and buggy o.O (takes about 60 seconds to scan 1400 refs and display)
<voxelot>
anyone else kinda creeped out that there are 5 pics of that from different angles
<whyrusleeping>
JohnClar_: what are you 'ls' ing?
<blame>
ha, I cached that
<voxelot>
haha, nice. i'll cache some paint drying :p
patcon has joined #ipfs
<blame>
A guy made an issue on the ipfs repo saying he was having issues getting it to show up on gateway.ipfs.io, and I cachewarmed it to try and help
tsenior`` has quit [Ping timeout: 244 seconds]
<JohnClar_>
whyrusleeping: Peers I randomly picked from the swarm
<whyrusleeping>
JohnClar_: youre ls'ing peer IDs?
<JohnClar_>
whyrusleeping: Yep
<whyrusleeping>
yeah, peer ID's arent directories
<JohnClar_>
Hm. Okay. Is it possible to ls remote nodes?
<whyrusleeping>
at best, you would be able to cat a peerID to get that nodes public key, but we dont currently add and announce public keys that way
<whyrusleeping>
nope, you cant 'query' a node you dont own
<JohnClar_>
Gottit. So is there any provision for content discovery?
<whyrusleeping>
once ipns is working a little better you will be able to ls /ipns/<peer ID> though
<JohnClar_>
ah
<whyrusleeping>
and that will be more in line with what youre expecting
<JohnClar_>
Sounds like it :)
<revolve>
whyrusleeping: what're the scaling issues with using tcp-based overlay routing?
<whyrusleeping>
revolve: in what sense?
devbug has quit [Ping timeout: 244 seconds]
<whyrusleeping>
like, scaling issues with a tcp dht?
<revolve>
you mentioned you were thinking of using QUIC instead of the current method due to scaling
<revolve>
yeah
JohnClare has joined #ipfs
<whyrusleeping>
so the one that annoys me the most is how you have to have a new socket for each connection
<whyrusleeping>
with a udp dht, i can have thousands of connections using a single fd
<revolve>
yeah
<whyrusleeping>
with tcp, i can reasonably have somewhere around 200-300
<whyrusleeping>
more than that and you run into issues during bootstrap and dialing
<revolve>
dialing?
<whyrusleeping>
yeah, when dialing we dial a few (up to 10) potential addresses for a node concurrently
<whyrusleeping>
this is an issue with NAT traversal
<revolve>
I just dropped the udp component of my project this afternoon after realising it'll be a lot simpler using upnp/nat-pmp and using the httpd instead of a udp socket
JohnClar_ has quit [Ping timeout: 244 seconds]
<whyrusleeping>
yeah, we use upnp/nat-pmp and still only get like 90% connectivitiy
<revolve>
it was also pushing it with gevent; which doesn't seem to play well within a subprocess without some hacks
<revolve>
ok
<revolve>
well when this thing is more stable it'll probably be ported to c or go
dignifiedquire has quit [Quit: dignifiedquire]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
JohnClare has quit [Remote host closed the connection]
<lgierth>
richardlitt: how do i run set-dns-records? npm run bin/index.js doesn't seem to actually run it, although i get a 0 exit-code
<lgierth>
to be fair, my nodejs is v0.10.25 and i got warnings about that during npm install
<richardlitt>
lgierth: currently $ node bin/index.js -t is the only command
<richardlitt>
$ node bin/index.js -h 2o3uoruo204 is what we will aim for
<lgierth>
aha yep! that works :P
<lgierth>
and the env var for the token
<richardlitt>
what needs to be done is setting the domainID and the record ID in lib/index.js - that's what I don't have access to.
<richardlitt>
might need to set up promise better, too.
<richardlitt>
In the call right now
<lgierth>
their backend is probably a rails app :P
<richardlitt>
Yeah
<lgierth>
as id the domain name wasn't a unique identifier...
<lgierth>
*as if
j0hnsmith has quit [Quit: j0hnsmith]
JohnClare has joined #ipfs
f[x] has joined #ipfs
Not_ has quit [Remote host closed the connection]
Not_ has joined #ipfs
ygrek_ has quit [Ping timeout: 246 seconds]
<ianopolous>
whyrusleeping: Any chance I could I get you to rename the IPFS-API-Java repo to java-ipfs-api to fit with the other ipfs names? I don't have the permission to rename it.
<jbenet>
ianopolous: done !
<whyrusleeping>
ianopolous: sure thing
<whyrusleeping>
JBENET DAMN YOU
<jbenet>
whyrusleeping: too slow
<ianopolous>
fantastic thank you :-)
therealplato has quit [Read error: Connection reset by peer]
<bret>
jbenet: sorry to miss the node-ipfs discussion was in meetings all afternoon
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<jbenet>
bret: no worries, still going. join us if desired
<bret>
backed up on work stuff atm... just got back from a 2 week vaca
<richardlitt>
jbenet: want me to sit in on any other hangouts today?
<jbenet>
richardlitt: nope all set
<jbenet>
nessence yep!
<jbenet>
Thanks though
<richardlitt>
jbenet: sweet. Catch you later then! Plenty of stuff to do.
rendar has quit []
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to feat/unwant: http://git.io/vZSN2
<ipfsbot>
go-ipfs/feat/unwant 6c02259 Jeromy: add a basic test for unwant...
<lgierth>
richardlitt: works nicely
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<M-davidar>
lgierth: your sprint meetings are at 3am here :/
<lgierth>
it would be cool to fetch the recordId from the api, so that you give it domainName and recordName, and it figures out the rest
<lgierth>
M-davidar: yeah :(
<lgierth>
richardlitt: ^ had to dig through their html to find the id :)
M-davidar is now known as davidar
<lgierth>
richardlitt: you can call the param that we use to build the TXT data a path, e.g. /ipfs/Qmfoo or /ipns/Qmbar -- the data is dnslink=<path> then
jamescarlyle has quit [Remote host closed the connection]
<davidar>
Luzifer: just replied to you on #matrix btw
<Luzifer>
davidar: that's the difference: IRC Cloud tells my phone you mentioned me... Matrix doesn't.
<davidar>
Luzifer: you have the mobile client installed?
<richardlitt>
Should I move it to ipfs/set-dns-record?
<richardlitt>
Was the extra code any difficulty?
<Luzifer>
davidar: now I have... :)
<davidar>
Luzifer :)
<Luzifer>
I think it indeed was a transport error...
<davidar>
IPFS version working now
<davidar>
?
<davidar>
chat.ipfs.io
notduncansmith has joined #ipfs
JohnClare has quit [Ping timeout: 252 seconds]
notduncansmith has quit [Read error: Connection reset by peer]
atrapado has quit [Quit: Leaving]
<Luzifer>
Don't know and I'm not gonna get out of my bed the next 8 hours... ;) (I'll test that later)
<davidar>
Ah, sorry, thanks
<Luzifer>
No problem ;)
<lgierth>
richardlitt: maybe something like dnslink-deploy for the name?
<davidar>
What timezone are you?
<richardlitt>
sure. We've got more plans for this, right?
<Luzifer>
Just was reading a bit Imgur and chatting with a colleague...
<richardlitt>
As in, it should be integrated into the `make` for a deploy
<Luzifer>
UTC+0200 at moment...
<richardlitt>
I'll rename it and transfer it over, maybe we can add issues to it with future plans, etc.
<lgierth>
richardlitt: it's more of a workaround -- until IPNS is complete and stable-ish
<richardlitt>
lgierth: genau.
<richardlitt>
lgierth: renaming and mving it now
<lgierth>
once IPNS is good, we'll just point the TXT to dnslink=/ipns/<hash>/whatever
<lgierth>
cool
<Luzifer>
Hrhr another german :D
<lgierth>
not so sure!
<lgierth>
richardlitt: oh hrm it worked for h.ipfs.io but didn't for ipfs.io
<richardlitt>
Luzifer: yep. Mailed in my retainer to the german immigration lawyer today, hopefully this time next year I'll have a Deutsch passport :D
<jbenet>
ekaron: i've to step out for a while. i can discuss computable datastructures late tonight, or tomorrow. sorry, lots on my plate today
f[x] has quit [Ping timeout: 272 seconds]
<lgierth>
a german passport is useful
* jbenet
afk for a while.
therealplato has quit [Ping timeout: 265 seconds]
<richardlitt>
lgierth: added in a readme, changed package.json. Maybe add in your code changes and then we can publish to npm? No rush!˜
andars has joined #ipfs
<ipfsbot>
[ipfs] RichardLitt opened pull request #93: Accidentally a word (master...patch-1) http://git.io/vZ9v5
<richardlitt>
jbenet: Can't merge PRs to ipfs/ipfs. +auth please?
<whyrusleeping>
richardlitt: just wait for jbenet to merge
<richardlitt>
There are only three current participants in the Sept 14th sprint. :/ Who is missing?
<richardlitt>
whyrusleeping: can do.
<davidar>
jbenet: computable data structures? I'd be interested too (later :)
dokee has joined #ipfs
<lgierth>
richardlitt: usually the pad fills up until i materialize it in ipfs/pm/issues tomorrow
<dokee>
hello
<richardlitt>
lgierth: cool. I'd be happy to be in charge of adding it to ipfs/pm/issues, that seems like something I could do to make your job easier.
<voxelot>
hello dokee
<dokee>
sorry I posted on the google group but it looks like the activity is here :)
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<dokee>
I ll connect tomorrow for my questions. Thanks !
dokee has quit [Client Quit]
devbug has quit [Read error: Connection reset by peer]
shea256 has quit [Remote host closed the connection]
devbug has joined #ipfs
<lgierth>
richardlitt: it's a whole role :) sprint sync role call and moderation, and taking care of the pad and issue
<lgierth>
i'll see to write a little checklist this week
<lgierth>
the monday needs to happen more or less reliably because it's important for the collaboration
<lgierth>
i'll take care of that issue tomorrow and we'll see what the others think of a rotating sprint manager ok?
JohnClare has joined #ipfs
devbug has quit [Read error: Connection reset by peer]
JohnClare has quit [Ping timeout: 268 seconds]
pfraze has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
nessence has quit [Ping timeout: 260 seconds]
tsenior`` has joined #ipfs
therealplato has joined #ipfs
akhavr has quit [Ping timeout: 268 seconds]
shea256 has joined #ipfs
simonv3 has joined #ipfs
nessence has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]