<Gaboose>
"This client will set up their ipfs node (note: this is moderately expensive and you normally wont just spin up an instance for a single connection)"
nskelsey___ has quit [Ping timeout: 264 seconds]
s_kunk has quit [Ping timeout: 260 seconds]
<Gaboose>
ok, not exactly the same thing as what you're saying
mildred has joined #ipfs
nskelsey___ has joined #ipfs
shyamsk has joined #ipfs
<Gaboose>
M-davidar: btw, how do you respond always so quickly here? :)
<Gaboose>
do you have a separate monitor dedicated to the ipfs channel on all the time?
echo_oddly has joined #ipfs
<M-davidar>
Gaboose (IRC): you wouldn't start the daemon just for a single request, but I think having your own daemon running for the app is OK
<M-davidar>
Gaboose (IRC): no, just seem to arrive at the right time sometimes :p
technomad has joined #ipfs
<M-davidar>
probably just a symptom that I check irc too often :p
<Gaboose>
im not complaining :) i personally can't switch contexts that quickly
akaihola has quit [Ping timeout: 245 seconds]
sugarpuff has joined #ipfs
akaihola has joined #ipfs
dignifiedquire has joined #ipfs
<Gaboose>
would a daemon per app duplicate the memory usage if they pin the same thing?
leeola has joined #ipfs
leeola has quit [Changing host]
ecloud has joined #ipfs
ecloud has quit [Changing host]
M-whyrusleeping has quit [Changing host]
M-whyrusleeping has joined #ipfs
M-oddvar has quit [Changing host]
M-oddvar has joined #ipfs
M-davidar-test has quit [Changing host]
M-davidar-test has joined #ipfs
M-nated has quit [Changing host]
M-nated has joined #ipfs
M-cryptix has quit [Changing host]
M-cryptix has joined #ipfs
JasonWoof has joined #ipfs
JasonWoof has joined #ipfs
M-prosodyContext has joined #ipfs
M-prosodyContext has quit [Changing host]
M-hash has quit [Changing host]
M-hash has joined #ipfs
M-rschulman has quit [Changing host]
M-rschulman has joined #ipfs
lachenmayer has quit [Changing host]
lachenmayer has joined #ipfs
jhiesey has quit [Changing host]
jhiesey has joined #ipfs
ogd has quit [Changing host]
ogd has joined #ipfs
anderspree has quit [Changing host]
anderspree has joined #ipfs
nrw has joined #ipfs
nrw has quit [Changing host]
jonl has quit [Changing host]
jonl has joined #ipfs
evanmccarter has quit [Changing host]
evanmccarter has joined #ipfs
bigbluehat__ has quit [Changing host]
bigbluehat__ has joined #ipfs
bret has quit [Changing host]
bret has joined #ipfs
mappum has joined #ipfs
mappum has quit [Changing host]
NightRa has quit [Changing host]
NightRa has joined #ipfs
prosody has quit [Changing host]
prosody has joined #ipfs
simonv3 has quit [Changing host]
simonv3 has joined #ipfs
nskelsey___ has quit [Changing host]
nskelsey___ has joined #ipfs
sugarpuff has joined #ipfs
sugarpuff has quit [Changing host]
<M-davidar>
Yeah, unfortunately
SoreGums has joined #ipfs
jhiesey has quit [Ping timeout: 245 seconds]
SoreGums has quit [Changing host]
SoreGums has joined #ipfs
bauruine has quit [Ping timeout: 264 seconds]
<M-davidar>
That's going to be an issue when we start having in-browser-tab nodes too I guess, so I'm not sure what the solution will be
anderspree has quit [Ping timeout: 245 seconds]
sugarpuff has quit [Ping timeout: 260 seconds]
simonv3 has quit [Ping timeout: 260 seconds]
nrw has quit [Ping timeout: 245 seconds]
ogd has quit [Ping timeout: 245 seconds]
nskelsey___ has quit [Read error: Connection reset by peer]
<Gaboose>
pinning in itself would be inefective even with a single tab, right?
nskelsey___ has joined #ipfs
sugarpuff has joined #ipfs
<M-davidar>
It would only have an effect while that tab is open, yes
anderspree has joined #ipfs
echo_oddly has quit [Quit: No Ping reply in 180 seconds.]
<M-davidar>
Hmm... multiple daemons probably is frowned on a little bit
<Gaboose>
well, if in-browser nodes would connect to localhost:8080 thin-client-like, per-app pinning could be one of the solutions
nrw has joined #ipfs
ogd has joined #ipfs
nskelsey___ has quit [Remote host closed the connection]
<M-davidar>
Yeah, it sounds like a reasonable feature to have
dignifiedquire has quit [Quit: dignifiedquire]
simonv3 has joined #ipfs
<Gaboose>
lots of work on ipfs to be done still :)
cryptix has quit [Ping timeout: 264 seconds]
bren2010 has quit [Ping timeout: 260 seconds]
<M-davidar>
Gaboose (IRC): you could submit a feature request to github
<M-davidar>
*on
<ipfsbot>
[go-ipfs] jbenet pushed 1 new commit to master: http://git.io/vB9Xs
<ipfsbot>
go-ipfs/master 792da9d Juan Benet: Merge pull request #1866 from ipfs/lgierth-patch-1...
cryptix has joined #ipfs
bren2010 has joined #ipfs
echo_oddly has joined #ipfs
evanmccarter has quit [Quit: Connection closed for inactivity]
nskelsey___ has joined #ipfs
HastaJun has quit [Ping timeout: 245 seconds]
kumavis has joined #ipfs
jonl has quit [Ping timeout: 264 seconds]
jbenet has joined #ipfs
jonl has joined #ipfs
<alu>
hey jbenet
<alu>
are you a fan of william gibson
nrw has quit [Ping timeout: 264 seconds]
bret has quit [Ping timeout: 264 seconds]
NightRa has quit [Read error: Connection reset by peer]
simonv3 has quit [Read error: Connection reset by peer]
M-eternaleye is now known as eternaleye
eternaleye has quit [Changing host]
eternaleye has joined #ipfs
cmeiklejohn has joined #ipfs
jonl has joined #ipfs
ogzy has joined #ipfs
ogzy has quit [Changing host]
ogzy has joined #ipfs
cryptix has joined #ipfs
oleavr has joined #ipfs
domanic has joined #ipfs
step21 has quit [Ping timeout: 264 seconds]
simonv3 has joined #ipfs
jbenet has quit [Ping timeout: 264 seconds]
echo_oddly has quit [Quit: No Ping reply in 180 seconds.]
step21_ has joined #ipfs
step21_ is now known as step21
jbenet has joined #ipfs
rombou has quit [Ping timeout: 260 seconds]
Monokles has joined #ipfs
sugarpuff has quit [Ping timeout: 264 seconds]
sugarpuff has joined #ipfs
cmeiklejohn has quit [Ping timeout: 264 seconds]
ogd has quit [Ping timeout: 264 seconds]
nrw has quit [Ping timeout: 264 seconds]
cmeiklejohn has joined #ipfs
mappum has quit [Ping timeout: 264 seconds]
nskelsey___ has quit [Ping timeout: 264 seconds]
nrw has joined #ipfs
diffalot has quit [Ping timeout: 264 seconds]
anderspree has quit [Read error: Connection reset by peer]
ogd has joined #ipfs
Monokles has quit [Ping timeout: 264 seconds]
anderspree has joined #ipfs
kerozene has quit [Ping timeout: 276 seconds]
olalonde has joined #ipfs
echo_oddly has joined #ipfs
<olalonde>
Hi all. I'd like to make all the files stored on an HTTP server available over IPFS. I have the list of all URLs and corresponding hashes. Any tips on how to accomplish this?
codehero has quit [Ping timeout: 264 seconds]
<olalonde>
I can't use the local filesystem, everything has to go through http requests
<olalonde>
I've done this for bittorrent already but I had to write a custom implementation which registers all the files I have on the DHT and when someone requests a chunk, I forward the request to the http server instead of using the local filesystem. it was a PITA
gozala______ has joined #ipfs
<olalonde>
I suppose I'll have to use the same approach for IPFS but was wondering if anyone could provide tips on where to start
gozala_____ has quit [Ping timeout: 264 seconds]
gozala______ is now known as gozala_____
jonl has quit [Ping timeout: 264 seconds]
<xelra>
How's that possible? How is it indexing the files?
jonl has joined #ipfs
<olalonde>
I actually operate the web server so have the list of hashes / urls
<olalonde>
But I'm using s3 as a backend for storage so don't have a local filesystem with the files there
diffalot has joined #ipfs
diffalot has joined #ipfs
jhiesey_ has joined #ipfs
echo_oddly has quit [Ping timeout: 264 seconds]
jhiesey has quit [Ping timeout: 264 seconds]
anderspree has quit [Read error: Connection reset by peer]
<olalonde>
The easiest option would be to mount s3 as a file system but I've been told it would be extremely slow
jhiesey_ is now known as jhiesey
Monokles has joined #ipfs
bret has joined #ipfs
<zaggynl>
test
echo_oddly has joined #ipfs
<olalonde>
jbenet: what do you recommend? I suppose best would be to write a custom implementation / fork?
devbug has quit [Ping timeout: 244 seconds]
cryptix has quit [Ping timeout: 264 seconds]
<Gaboose>
olalonde: im not exactly an expert, but maybe a fork with a custom bitswap strategy (one that actually fetches http instead of asks other peers for blocks) would do
codehero has joined #ipfs
jhiesey has quit [Ping timeout: 264 seconds]
sugarpuff has quit [Ping timeout: 264 seconds]
<olalonde>
Gaboose: thanks, I think I'll do something like that
NightRa has quit [Quit: Connection closed for inactivity]
<Gaboose>
a bridge between ipfs hashes and torrent magnet links would be interesting to see too :)
jhiesey has joined #ipfs
sugarpuff has joined #ipfs
<Gaboose>
maybe easier in some sense, because addresses of both are hashes of their content - immutable
<kyledrake>
Might be a good idea to throw a process monitor like monit on them if there isn't one
<kyledrake>
automate the process of giving go-ipfs a swift kick when it needs one
takinbo has quit [Ping timeout: 245 seconds]
takinbo has joined #ipfs
devbug has quit [Ping timeout: 244 seconds]
joeyh has quit [Ping timeout: 260 seconds]
joeyh has joined #ipfs
tidux has joined #ipfs
<tidux>
is it normal for an "ipfs pin" operation for a 324MB file to not exit after several hours?
<tidux>
this is on a colocated box with a 1Gbps uplink so I assume it's not my hardware bottlenecking
pfraze has joined #ipfs
cleichner has quit [Ping timeout: 260 seconds]
diffalot has quit [Ping timeout: 260 seconds]
mvr_ has quit [Quit: Connection closed for inactivity]
diffalot has joined #ipfs
technomad has quit [Ping timeout: 260 seconds]
cleichner has joined #ipfs
hellertime has quit [Ping timeout: 261 seconds]
<computerfreak>
tidux: there are some bugs, its still alpha , u can report bugs at github
<tidux>
so it is a bug?
brixen has quit [Ping timeout: 259 seconds]
<computerfreak>
yes there are sometimes people reporting same problems
<computerfreak>
but dont ask me for details xD i never tried such a large file
<computerfreak>
but maybe look at cashing or something?
<tidux>
it just seems to be stalling
<tidux>
~/.ipfs/ isn't getting any bigger and RAM usage is about constant
<computerfreak>
on both nodes?
<tidux>
what do you mean "both"?
Encrypt has joined #ipfs
<computerfreak>
1 node sends data, one receives it
<tidux>
this is just a file I'm pinning frome somewhere out there on IPFS
<tidux>
I've downloaded it before with ipfs get
<tidux>
I only have one IPFS node working at the moment
<computerfreak>
so u have it already on the computer and try to pin it ? (distribute it to others) ?
<computerfreak>
did u changed the file?
<computerfreak>
is the hash u try to pin the same as the original hash ?
<tidux>
Type coherently or don't bother typing at all.
<tidux>
I'm pinning the hash to make sure it remains available. I didn't add the file locally, I'm just pinning the same /ipfs URL I used with ipfs get
<computerfreak>
whois tidux
<tidux>
forgot your /
RJ2 has quit [Remote host closed the connection]
SWingedSeraph has quit [Ping timeout: 261 seconds]
anderspree has quit [Ping timeout: 257 seconds]
rollick has quit [Ping timeout: 257 seconds]
anderspree has joined #ipfs
RJ2 has joined #ipfs
SWingedSeraph has joined #ipfs
lidel has quit [Ping timeout: 262 seconds]
Guest56061 is now known as jero
cryptix has quit [Ping timeout: 257 seconds]
cryptix has joined #ipfs
brixen has joined #ipfs
brixen has quit [Ping timeout: 264 seconds]
SWingedSeraph has quit [*.net *.split]
mildred has quit [Read error: Connection reset by peer]
nekomune has quit [Ping timeout: 245 seconds]
pfraze has quit [Remote host closed the connection]
rombou has quit [Ping timeout: 260 seconds]
Rylee has quit [Ping timeout: 260 seconds]
seductivebarry has quit [Ping timeout: 240 seconds]
rendar has quit [Ping timeout: 260 seconds]
joeyh has quit [Ping timeout: 260 seconds]
dignifiedquire has quit [Ping timeout: 260 seconds]
Rylee has joined #ipfs
dignifiedquire has joined #ipfs
s_kunk has quit [Ping timeout: 260 seconds]
SWingedSeraph has joined #ipfs
takinbo has quit [Ping timeout: 260 seconds]
simonv3 has quit [Read error: Connection reset by peer]
simonv3 has joined #ipfs
HastaJun has quit [Ping timeout: 260 seconds]
HastaJun has joined #ipfs
evanmccarter has joined #ipfs
shyamsk has quit [Ping timeout: 260 seconds]
akaihola has quit [Ping timeout: 260 seconds]
akaihola has joined #ipfs
<Confiks>
Does corenet.Dial currently guarantee that the node you're dialing is in possesion of the private key connected to its peerID?
wiedi has quit [*.net *.split]
kerozene has quit [*.net *.split]
The_8472 has quit [*.net *.split]
Guest28376 has quit [*.net *.split]
<Confiks>
nodeID*
<Confiks>
Err, peerID* :D
NightRa has joined #ipfs
wiedi has joined #ipfs
The_8472 has joined #ipfs
svetter has joined #ipfs
kerozene has joined #ipfs
s_kunk has joined #ipfs
ogd_ has joined #ipfs
ogd has quit [Ping timeout: 264 seconds]
ogd_ is now known as ogd
SWingedSeraph has quit [*.net *.split]
<tidux>
OK, so question answered
<tidux>
my problem was basically forcing the daemon to drink out of a firehose by doing ipfs add -r on a 24GB directory
<tidux>
now it's starting to spit out file-added messages
<tidux>
and ~/.ipfs is growing accordingly
SWingedSeraph has joined #ipfs
Confiks_ has joined #ipfs
wiedi has quit [*.net *.split]
The_8472 has quit [*.net *.split]
<zignig>
simonv3 has quit [Ping timeout: 264 seconds]
simonv3 has joined #ipfs
<zignig>
jero has left #ipfs ["Leaving"]
akaihola has quit [Ping timeout: 260 seconds]
<tidux>
yes?
dignifiedquire_ has joined #ipfs
akaihola has joined #ipfs
nskelsey___ has quit []
nskelsey has joined #ipfs
rombou has joined #ipfs
takinbo- has joined #ipfs
Gaboose has quit [Ping timeout: 260 seconds]
joeyh has joined #ipfs
technomad has joined #ipfs
brixen has joined #ipfs
ashark has joined #ipfs
hjoest has quit [Ping timeout: 246 seconds]
HostFat has joined #ipfs
hellertime has joined #ipfs
chriscool has joined #ipfs
hjoest has joined #ipfs
rendar has joined #ipfs
computerfreak has quit [Ping timeout: 244 seconds]
seductivebarry has joined #ipfs
wiedi has joined #ipfs
The_8472 has joined #ipfs
rollick has joined #ipfs
SWingedSeraph has quit [*.net *.split]
pfraze has joined #ipfs
mildred has joined #ipfs
bigbluehat__ is now known as bigbluehat
bigbluehat is now known as Guest48697
elima has joined #ipfs
computerfreak has joined #ipfs
<cryptix>
Confiks: re privkey posession: corenet.Dial in ipfs needs the pubkey hash at the end of the multiaddr, iirc that's exactly what it is used fore
SWingedSeraph has joined #ipfs
<ansuz>
if I haven't updated in a while, what's the best way to do so?
pokeball99 has quit [Quit: Connection closed for inactivity]
<ansuz>
there was a lot of talk about an update command not too long ago
Taek42 is now known as Taek
pfraze has quit [Remote host closed the connection]
<daviddias>
- [x] travel logistics for Node.js Interactive
<daviddias>
- [~]() prepare talk for Node.js Interactive (Got slides ready, need to practice and make sure it only takes 20 minutes.. which is soo hard!)
<daviddias>
ahahahahahah
<dignifiedquire>
<whyrusleeping>
how is ipscend stuff going?
<daviddias>
the thing is pretty sweet
<whyrusleeping>
i havent heard much on that front
<daviddias>
the preview feature now has a very good experience
<whyrusleeping>
what is surge?
<richardlitt>
Anything I can help with, david?
<daviddias>
thanks from Andres (which I had the chance to meet at the IPFS Hack Fest Barcelona organized by Victor and Pau)
<richardlitt>
Talk prep, maybe?
pfraze has joined #ipfs
<daviddias>
richardlitt: getting feedback would be certainly great! Plus, getting registry-mirror really good, making sure experience is nice, that the present README explains the magic well enough, that are no set up bugs and so on, would be awesome help
pfraze has quit [Remote host closed the connection]
<richardlitt>
Might meet up with Crosscloud founder soon to talL: http://crosscloud.org/
<richardlitt>
Which sounds like the w3 implementation of POST
<richardlitt>
Yeah, I know, I want to use POST
<richardlitt>
*talk
<daviddias>
but like, right now, I would be happy if we have a thing where we can write some pages and there is a RSS feed and a list of links on ipfs.io/blog that people can check
<daviddias>
oh, can we tag the CrossCloud Founder on our POST discussions
<richardlitt>
Yeah
<richardlitt>
probably
<daviddias>
we would be happy if it became an actual spec
<richardlitt>
aye
<richardlitt>
Ok, all good?
<daviddias>
or at least partially, when it comes to interface and things, which is kind of the Domain for W3C
<richardlitt>
whyrusleeping: Please add your stuff to the sprint
<whyrusleeping>
i did!
<whyrusleeping>
issue #60?
Oatmeal has joined #ipfs
Oatmeal has quit [Max SendQ exceeded]
CounterPillow has quit [Remote host closed the connection]
<richardlitt>
whyrusleeping: ahhh
<whyrusleeping>
hrm... right. we add it to the previous place dont we
<richardlitt>
I just added a note to the Readme about this
<Confiks>
cryptix: Yeah, I thought so to. I was doing RTFS, but I guess the peerID verification on corenet.Dial and listener.Accept is burried away so deep I can't find it. I don't yet know the source too well though.
<richardlitt>
### Officially this week:
<richardlitt>
- [x] Sprint management.
<richardlitt>
- [x] Remove video links from the readme. Issue: ipfs/pm/issues/57. PR: ipfs/pm/pull/59.
<richardlitt>
- [x] Add a PR to ipfs.io/docs/api about how the commands there are not all-inclusive. PR: ipfs/website/pull/81.
<richardlitt>
- [x] Suggest community infrastructure for meetups. Comment: ipfs/community/issues/64#issuecomment-160335959
<richardlitt>
- [x] Make the documentation around Commit Sign-Offs more friendly. PR ipfs/community/pull/82
<richardlitt>
#### Extra
<richardlitt>
- [x] Discussed the Readme doever at length, ending with jbenet's ipfs/ipfs/pull/128.
<richardlitt>
Ok, that's me. In other news I also launched antinomadic.com, if ya'll are digital nomads, and closed the user is drunk with a long blog post. But that is boring and off topic.
<richardlitt>
whyrusleeping: you're up next
<whyrusleeping>
- [ ] ipfs-update :ship:
<whyrusleeping>
- [ ] fs-repo-migrations (needs CR to merge)
<whyrusleeping>
- [ ] commands lib path argument type
<daviddias>
A sprint issue should be the plan and the snapshot of what happened that week
<whyrusleeping>
- [ ] write layout and chunking algo to root block of created dags.
<whyrusleeping>
- [x] say "pub sub" out loud three times
<whyrusleeping>
- [x] eat turkey
<whyrusleeping>
- [x] provide-many mechanism, yay bandwidth savings!
* whyrusleeping
hasnt had coffee yet
<richardlitt>
TurkeY!
<daviddias>
Turrkey
<whyrusleeping>
yeah, i pretty much took last week off
<richardlitt>
<richardlitt>
Me too. Thanksgiving.
<richardlitt>
No worries. We all did.
<richardlitt>
Late november sucks in the northern hemisphere.
<daviddias>
in America
<whyrusleeping>
did a little bit of coding, but aside from the provide many (which i'm super duper psyched about), nothing to report
<daviddias>
no one gave me a Turkey
<daviddias>
not saying thanks to anyone
<whyrusleeping>
daviddias: you have to go get your turkey
<daviddias>
:P
<whyrusleeping>
kill it yourself
<richardlitt>
daviddias: to be fair, turkeys aren't even native in portugal
<richardlitt>
my sister and her brother in law enjoyed their week in Portugal last week, so, thanks for helping make your country fun for American tourists?
<dignifiedquire>
I don’t like turkey, so I’m out
<whyrusleeping>
lol
<richardlitt>
Also, I got drunk with a lot of portuguese translators the weekend before last, they had good things to say about portugal, too. I blame your influence.
<daviddias>
you know, funny fact, so in english Turkey is an animal and a Country
<daviddias>
and if translate Turkey, the animal, to Portuguese, you get Peru, which is again an animal and a Country
<dignifiedquire>
oh yeah
<richardlitt>
And a meat. And a word for kidding children
<daviddias>
coincidence? I think not!
<richardlitt>
"You turkey! you stole my mug"
<richardlitt>
My mom used to call me that all of the time.
<daviddias>
ahah
<dignifiedquire>
there is a us quizshow on youtube where they tell a rather less intelligent lady that Turkey is a country and she refuses to believe it :D
<richardlitt>
Yeah. It's obviously called Anatolia
<richardlitt>
Anyway! Good work, whyrusleeping
<richardlitt>
Anyone else need to go today?
<dignifiedquire>
not sure
<dignifiedquire>
anyone up for doing a quick hangout on air test with me?
<daviddias>
whyrusleeping: we are planning to do a TV Show
<daviddias>
called "Jeromy talks about IPFS"
<daviddias>
we set up a topic like 'bitswap'
<dignifiedquire>
yes
<daviddias>
and then it is just you talking about it
<dignifiedquire>
and we post it on youtube :D
<daviddias>
so many subscribers!
<dignifiedquire>
and live stream it
<dignifiedquire>
to the woooorld
<daviddias>
going to get one of those Platinum Play buttons Youtube is sending to everyone
<richardlitt>
hahha
<richardlitt>
dignifiedquire: sure
<whyrusleeping>
oh?
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #109: Update font-awesome to version 4.5.0
sbruce_ has quit [Ping timeout: 245 seconds]
<whyrusleeping>
hrm, i dont know how interesting that would be
<whyrusleeping>
especially right now
<daviddias>
whyrusleeping: dignifiedquire will convert that into specs and documentation :D
<whyrusleeping>
it would just be "whyrusleeping angry about things and not having coffee"
<jbenet>
would probably be good to trim out the recorded videos a bit. the infra video is like 7min of banter. (it's good for us to have because monday sprints are the only recurring face-to-face time we get regularly with everyone, but not good for people attempting to get what's going on by watching)
<multivac>
jbenet: 2015-11-30 - 10:32:18 <dignifiedquire> tell jbenet take a look at this: https://github.com/ipfs/pm/issues/61 and ping me please
<dignifiedquire>
jbenet: you will get acces to youtube so you can cut to all your hearts content
<dignifiedquire>
daviddias: I’m out now, so you have to be the hangout master for the rest of the night ;)
wiedi has quit [Quit: ^Z]
<daviddias>
oh my
<daviddias>
we really need that shared account
HostFat has quit [Read error: Connection reset by peer]
<jbenet>
dignifiedquire daviddias: i wont be at the rest-- they're usually need-based only. (it may even be worth removing the periodic slots for them)
<dignifiedquire>
jbenet: can you ping me tomorrow when you have a moment, for organizational stuff around ipfs youtube channel/hangouts on air things?
<jbenet>
dignifiedquire: thanks for setting up the recording stuff
<jbenet>
dignifiedquire: yep, can do
<ion>
NeoTeo: Yes, switching the default chunker is a major thing, not to be done without a good reason. I would say switching from fixed chunking to a rolling hash is worth it.
<dignifiedquire>
thanks
<ion>
Especially when IPFS is still in alpha.
<NeoTeo>
ion indeed as soon as possible so we don't get too much of the "old" chunker. But the very fact that you can choose one seems odd.
<dignifiedquire>
jbenet: are there any limitations to adhere to on ipfs/distributions other than, it has to run in a browser?
<ion>
dignifiedquire: It should require 1024×768.
<jbenet>
dignifiedquire: yeah-- (a) it should work on mobile too (at least to see it and copy links, etc), (b) it should be very basic-- with everything in one page. (sectioned off) (c) the composition isn't good right now, but the required sections are all there: title, description, clear platform links, gif of the dist,
<jbenet>
dignifiedquire: this is based on the https://iojs.org and nodejs pages, which are very good at being to the point (iojs one specially)
<jbenet>
ideally it would just be a bit of css fixing the positioning to make what's there look ok
<jbenet>
(right now it feels cluttered, etc)
grahamperrin has joined #ipfs
<dignifiedquire>
jbenet: okay thanks will keep that in mind
<dignifiedquire>
jbenet: so just the latest versions on the main page with older/nightlies on a linked page?
<jbenet>
dignifiedquire: yeah, those links can go to the ipfs dir listings
<whyrusleeping>
i need to learn more about javascript first
rjeli has quit [Ping timeout: 245 seconds]
bren2010 has quit [Ping timeout: 245 seconds]
devbug has quit [Read error: Connection reset by peer]
cemerick has quit [Ping timeout: 245 seconds]
rjeli has joined #ipfs
bren2010 has joined #ipfs
hjoest has quit [Ping timeout: 246 seconds]
ygrek has quit [Ping timeout: 260 seconds]
devbug has joined #ipfs
hjoest has joined #ipfs
<jbenet>
hey everyone-- can the leaders of their hangout trim the unnecessary parts of their hangout on youtube? it wont scale for one person to do it. it will also get you to rewatch yourself as feedback :)
nham has joined #ipfs
elima has quit [Ping timeout: 260 seconds]
grahamperrin has joined #ipfs
<whyrusleeping>
jbenet: how do?
elima has joined #ipfs
grahamperrin has left #ipfs [#ipfs]
CarlWeathers has joined #ipfs
devbug has quit [Read error: Connection reset by peer]
devbug has joined #ipfs
voxelot has joined #ipfs
Not_ has joined #ipfs
true_droid has joined #ipfs
Not_ has quit [Client Quit]
wiedi has joined #ipfs
domanic has quit [Ping timeout: 245 seconds]
domanic has joined #ipfs
harlan_ has joined #ipfs
gozala_____ is now known as gozala
gozala has quit []
gozala has joined #ipfs
harlan__ has joined #ipfs
domanic has quit [Ping timeout: 250 seconds]
<harlan_>
jbenet are you planning to talk POST at 2:30? hangout link will be here if so?
<jbenet>
harlan_ wasn't planning to. Not much progress since last hangout
<NeoTeo>
whyrusleeping super cool music player :) Let the feature request begin!
<whyrusleeping>
NeoTeo: :D i'm going to try and make a friend of mine work on it for me, he seemed really interested
<dignifiedquire>
jbenet: I can do the trimming tomorrow morning
<dignifiedquire>
jbenet: (until I have given access and thaught everyone how to do so)
anticore has quit [Read error: Connection reset by peer]
<NeoTeo>
whyrusleeping Only trouble is that to listen to your own music you have to put it out there which is legally a bit of a gray-zone I assume.
<whyrusleeping>
NeoTeo: yeah... i'm not really sure how that works
<NeoTeo>
personal DRM. Yuck!
<whyrusleeping>
i'm getting really upset at iptables right now
<whyrusleeping>
packets are being forwarded nicely, but i cant make ssh connections to the box itself from outside of its lan...
<whyrusleeping>
freaking half an hour of fighting with this
<ion>
whyrusleeping: Well, IPFS is going to support encrypted, private sharing, right? The player should work fine with that, shouldn't it?
<whyrusleeping>
ion: yeap
<harlan__>
jbenet: sure makes sense
nekomune has joined #ipfs
ygrek has joined #ipfs
gamemanj has quit [Ping timeout: 260 seconds]
<achin>
whyrusleeping: are the incoming packets being fowarded, too? maybe there is a DROP rule somewhere in the FORWARD table?
<whyrusleeping>
achin: hrm... i'm noticing that the Send-Q has a bunch of packets in it
<whyrusleeping>
not sure why
<achin>
that doesn't sound normal
<whyrusleeping>
yeah... it really doesnt
<whyrusleeping>
i don't understand any of this
<whyrusleeping>
one of my forwarding commands is working fine
HoboPrimate has joined #ipfs
<achin>
is the iptables/forwarding machine a NATing router?
<whyrusleeping>
yeah
<whyrusleeping>
and the nat works, my internal machines can make normal internet connections
<dignifiedquire>
it kinda sucks not being able to read the history in between
<whyrusleeping>
dignifiedquire: not sure, i invited the bot back in, but i think its actually offline
<dignifiedquire>
hmm
<dignifiedquire>
oh right I just checked, it
<dignifiedquire>
’s dead for other channels as well :(
<achin>
whyrusleeping: tcpdump is a really useful tool here. for example, does an incoming packet hit the WAN iface? does it hit the LAN iface? does it hit the interface of the internal machine (the one behind the NAT)
<whyrusleeping>
good point! one sec...
<dignifiedquire>
whyrusleeping: do you know if we are running this bot ourselves, or if they are running it for us?
<whyrusleeping>
they are running it
<achin>
whyrusleeping: (from a 3rd party external machine, i'll use netcat to send packets on a unique port, so i can filter just on that port numberr and not get drowned in a bunch of not-related packets)
<dignifiedquire>
okay, I sent them a tweet, let’s see if we get sth back
harlan_ has quit [Ping timeout: 252 seconds]
harlan__ is now known as harlan_
dignifiedquire has quit [Quit: dignifiedquire]
<harlan_>
maybe we should run our own. it's open source...
amade has quit [Quit: leaving]
<ipfsbot>
[go-ipfs] whyrusleeping created Dignifiedquire-fix/access-controll-headers (+3 new commits): http://git.io/vBddI
<ipfsbot>
go-ipfs/Dignifiedquire-fix/access-controll-headers eedd5d5 dignifiedquire: Add correct access control headers to the default api config...
<ipfsbot>
go-ipfs/Dignifiedquire-fix/access-controll-headers e09c397 dignifiedquire: Try to fix tests...
<ipfsbot>
go-ipfs/Dignifiedquire-fix/access-controll-headers b036b23 dignifiedquire: Add access-control-headers to gateway config...
domanic has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to Dignifiedquire-fix/access-controll-headers: http://git.io/vBdds
<ipfsbot>
go-ipfs/Dignifiedquire-fix/access-controll-headers bdc1b27 Jeromy: hard code things...
null_radix has quit [Remote host closed the connection]
elima has quit [Ping timeout: 250 seconds]
rombou has joined #ipfs
devbug has quit [Ping timeout: 260 seconds]
patcon has quit [Ping timeout: 245 seconds]
patcon has joined #ipfs
go1111111 has quit [Ping timeout: 260 seconds]
pinbot has joined #ipfs
<lgierth>
cryptix: pinbot back :/
<lgierth>
sorry for missing sync and hangouts folks :S bad scheduling and org on my part...
computerfreak has quit [Quit: Leaving.]
<whyrusleeping>
if i ever meet Rusty Russells (author of iptables) i might not be able to prevent myself from smacking him
<rendar>
when ipfs divides files into chunks, the size of these chunks *must* be fixed, or can vary?
<rendar>
i mean, vary in the same file
go1111111 has joined #ipfs
ashark has quit [Ping timeout: 245 seconds]
<lgierth>
rendar: it can vary -- with rabin fingerprinting (ipfs add -s rabin), they actually vary most of the time
<rendar>
lgierth: i see
<lgierth>
the default chunksize of the default chunker is 256k
<rendar>
lgierth, but when it varies, how you can read from offset X if you don't know in which chunk X starts? you have to read all chunks sizes until X, to get which chunk where X starts
Encrypt has quit [Quit: Quitte]
<ion>
rendar: By a rolling hash. You only need to look at a window to determine split points.
<rendar>
ion: so its a O(1) operation?
<lgierth>
ion: i think they mean readnig
<lgierth>
*reading
<lgierth>
rendar: the root object has links to all the chunks
<rendar>
lgierth: i know that, but if you have to read all links, it takes O(n)
<lgierth>
yep
<lgierth>
gives better deduplication though
<ipfsbot>
[go-ipfs] whyrusleeping created feat/actually-ndjson (+1 new commit): http://git.io/vBFfA
<ipfsbot>
go-ipfs/feat/actually-ndjson 891a8d7 Jeromy: no newlines in json objects...
<rendar>
lgierth: so it really takes O(n) with rolling hashes?
<lgierth>
that i don't know
<ion>
The root object gives the offsets, no?
<rendar>
ion: yep
<lgierth>
do the links contain the size of the child?
<ion>
Yeah
<rendar>
ion: i mean, you have to read *all* the root object, *all* of its objects
<lgierth>
if so then you can pick exactly the link that you're looking for
<tperson>
whyrusleeping: I thought you said you loved iptables lol
<rendar>
ion: what about the root object is really big?
<ion>
rendar: The unixfs file objects are arranged to a tree. The depth is only 1 for files with few chunks.
<rendar>
ion: oh i see, so for big files they became a tree
<lgierth>
sub[-sub-sub]-chunks
<lgierth>
so to say
<jbenet>
daviddias: very nice ipscend blog post!
NightRa has quit [Quit: Connection closed for inactivity]
<ion>
I think “chunk” only refers to the leaf nodes, the branch nodes are something else. I may have this wrong though.
<whyrusleeping>
tperson: no, iptables is the devil
<ipfsbot>
[go-ipfs] whyrusleeping opened pull request #2020: no newlines in json objects (master...feat/actually-ndjson) http://git.io/vBFUj
<whyrusleeping>
daviddias: did you guys *just* want the json output to be properly newline delimited?
<whyrusleeping>
or was there something else?
<daviddias>
ndjson would fix the issue
domanic has quit [Read error: Connection reset by peer]
<daviddias>
However, there are two x headers
<daviddias>
x-stream and x-chunked
<ion>
daviddias: I'd suggest creating the DOM element (hidden) and preloading the image for the “next” version in the history view in case the user selects “next”.
<ansuz>
I just grepped my irc logs for this channel
<ansuz>
grep "the devil" \#ipfs.*.log
<daviddias>
Which I'm still trying to figure out why we need really chunked for something that is a stream
<kyledrake>
jbenet do you have the vector graphic for the libp2p logo box with boxes in it in your slide deck?
<daviddias>
Or to say it correctly,"re'figuring" out
<kyledrake>
Gimp won't cleanly take the black goo out
<kyledrake>
Wait, I might have just got it
devbug has joined #ipfs
voxelot has quit [Ping timeout: 260 seconds]
<achin>
whyrusleeping: if you want another pair on eyes on your tables, let me kow
<ion>
little bobby tables
<jbenet>
https://github.com/ipfs/pm/issues/63 <--- discussion on whether to record the hangouts or not henceforth. (i think it's a nice idea to have these, but im not convinced yet that the benefits outweigh the cons, or that we'll use it the most effective way)
<whyrusleeping>
achin: i gave up, just using ssh forwarding from an internal machine now
<achin>
:D
<achin>
something everyone's done at least 3 or 4 dozen times in the past