<ipfsbot>
[go-ipfs] whyrusleeping created feat/reqlog-time (+1 new commit): https://git.io/vgQVj
<ipfsbot>
go-ipfs/feat/reqlog-time f55803c Jeromy: add command to change keep time for reqlog objects...
<lgierth>
whyrusleeping: is there a quick way of pinning the dependencies?
<whyrusleeping>
dependencies for a gx package?
<lgierth>
yeah so i can gx install offline
<whyrusleeping>
'gx deps bundle'
<whyrusleeping>
gives you a hash containing all the deps
<lgierth>
oh yeah that works
<lgierth>
nice
<whyrusleeping>
i'm writing a testing thing called earl
<ipfsbot>
[go-ipfs] whyrusleeping opened pull request #2344: add command to change keep time for reqlog objects (master...feat/reqlog-time) https://git.io/vgQrJ
Encrypt has quit [Quit: Quitte]
<lgierth>
whyrusleeping: question about contexts -- do the context timed out errors still happen on the gateway? i know they did, but i can't reproduce them right now
<whyrusleeping>
i don't know that i see them
<whyrusleeping>
they still should be happening... i think
<lgierth>
mh ok -- because they end up as 404 responses iirc, so i'd need to look at the response body in multireq to see if it's a legitimate 404
<M-mubot>
mh ok has -1 points, -1 of which is for they end up as 404 responses iirc, so i'd need to look at the response body in multireq to see if it's a legitimate 404.
<jbenet>
whyrusleeping: awesome stuff with gx :)
<lgierth>
or just patch the goddamn gateway :)
<lgierth>
M-mubot: shut up
<jbenet>
whyrusleeping: it really works so well :) <3
<whyrusleeping>
jbenet: :)
<whyrusleeping>
i've put a lot of effort into it while nobody was looking
<lgierth>
whyrusleeping: you might need to have multiple Checks() no?
<whyrusleeping>
lgierth: the idea is that it just runs once at the end, and it can perform as many different things in the function as it needs
<whyrusleeping>
whats your thought there?
<ianopolous>
nice work, whyrusleeping!
<whyrusleeping>
(this is very open to input, i'm spitballing ideas)
<whyrusleeping>
ianopolous: :) thanks!
<lgierth>
but eventually you wanna see where it starts doing funny stuff
<whyrusleeping>
lgierth: true...
<lgierth>
i mean where Act() start being off
<whyrusleeping>
Act() can return an error at any time
<whyrusleeping>
which halts the entire thing
<lgierth>
so you're Check'ing in Act? :)
<whyrusleeping>
i mean, i guess you could...
<whyrusleeping>
hrm
<whyrusleeping>
maybe `Check` isnt needed then?
<whyrusleeping>
and everything for that just runs in `Act`
<lgierth>
i get why the usual unit test like procedure doesn't really fit
<lgierth>
trying to tell a story, not specify an api
Matoro has quit [Ping timeout: 248 seconds]
<lgierth>
you could have a high level funtion what tells the story, and move the details into $steps
<lgierth>
i have my love-hate relationship with cucumber
grandmaster has joined #ipfs
<lgierth>
(where the given/then/when steps make up the high level function, so to speak)
<whyrusleeping>
yeah...
<whyrusleeping>
i'm aiming for more of a 'what will real life usage look like' type of tests
<lgierth>
not saying cucumber is a good idea to introduce haha
<lgierth>
mh yep
<lgierth>
aha!
<lgierth>
} else if err == context.DeadlineExceeded {
<lgierth>
so basically anything <500 and !=408 is legitimate
<whyrusleeping>
ooooh, interesting
<lgierth>
yeah need to fix that -- just wanna get rid of the 503s first before i go full rabbit hole :)
<lgierth>
multireq right now thinks that <400 is legitimate, anything else counts as a failure
<lgierth>
if both targets fail, you get the 403
<lgierth>
*503
ppham_ has joined #ipfs
compleatang has quit [Quit: Leaving.]
<lgierth>
on the other hand, that line ^ implies that the context timeouts are caused by requests in fact timing out. then i should never see that e.g. in the browser though
<lgierth>
GET for example by definition does not time out
<lgierth>
unless your request headers are so big that they don't fit in a typical IP packets
ppham has quit [Ping timeout: 260 seconds]
compleatang has joined #ipfs
Matoro has joined #ipfs
<whyrusleeping>
hrm... they definitely shouldnt be
<whyrusleeping>
the faliures occur in clusters too
<lgierth>
and it's likely with the same host?
<lgierth>
i mean each cluster with the same host
<whyrusleeping>
likely, *very* hard to tell
<whyrusleeping>
i'd have to manually resolve the dns myself before making the http request to get that info
ppham_ has quit [Remote host closed the connection]
jedahan has joined #ipfs
Ronsor has quit [Ping timeout: 240 seconds]
Ronsor has joined #ipfs
Ronsor is now known as Guest93192
mkelly has joined #ipfs
mkelly has quit [Client Quit]
ppham has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping deleted fix/t0065-timing at 11f9e7f: https://git.io/vg7ea
M-eternaleye is now known as eternaleye
pepesza has quit [Ping timeout: 256 seconds]
<ipfsbot>
[js-ipfs-api] dignifiedquire created greenkeeper-stream-http-2.1.1 (+1 new commit): https://git.io/vg7vu
<ipfsbot>
js-ipfs-api/greenkeeper-stream-http-2.1.1 eacbe08 greenkeeperio-bot: chore(package): update stream-http to version 2.1.1...
<ipfsbot>
[js-ipfs-api] dignifiedquire deleted greenkeeper-stream-http-2.1.1 at eacbe08: https://git.io/vg7vM
tmg has quit [Quit: leaving]
tso has joined #ipfs
<tso>
sup
<tso>
whoops
<tso>
wrong channel my bad but as it reminds me
<tso>
how the **** is **** IPFS doing!? are we the distributed *permanent* web yet!?
<tso>
don't forget that you people have done an amazing job so far and to not lose your way :D
edubai__ has quit [Quit: Connection closed for inactivity]
sivachandran has quit [Remote host closed the connection]
sivachandran has joined #ipfs
sivachandran has quit [Client Quit]
jaboja has joined #ipfs
elima_ has joined #ipfs
reit has quit [Ping timeout: 260 seconds]
VegemiteToast has quit [Quit: Leaving]
Shane has quit [Ping timeout: 272 seconds]
rafajafar has joined #ipfs
jedahan has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
blanu has joined #ipfs
Tv` has quit [Quit: Connection closed for inactivity]
voxelot has quit [Ping timeout: 252 seconds]
M-davidar has quit [Quit: node-irc says goodbye]
ulrichard has joined #ipfs
m0ns00n2 has quit [Quit: undefined]
srenatus has joined #ipfs
voxelot has joined #ipfs
<rafajafar>
anyone up? Kinda curious about permanence. Can't IPFS files drop out of existence if not enough people have full copies?
<tso>
interesting you're concerned about permanance not being permanent
<tso>
I'm concerned abotu the inverse
<Stskeeps>
rafajafar: permanence is a bit of a social problem, you could always pay others to host you or get incentivised to host other people's files
<Stskeeps>
(IMHO)
Not_ has quit [Ping timeout: 240 seconds]
<rafajafar>
tso: I made a ContentID system. The only ones interested in systems like ContentID are content owners. Sites like YouTube find it frustrating.
<rafajafar>
With IPFS, I can have people register and protect their content and use my system as a single point of entry for getting and distributing their videos without needing YouTube and their massive server farm
ppham has quit [Remote host closed the connection]
<tso>
so you are in favor of copyright
<rafajafar>
however, tso, if the videos can disappear, then I need to figure out how I'd account for that
<rafajafar>
tso, most artists are, so I am
<tso>
OK
<rafajafar>
I'm in favor of accredation
<rafajafar>
and uniqueness
<rafajafar>
I'm not in favor of takedowns, by the way... I'm also not in favor of automated systems taking full control
<tso>
<rafajafar>
I'd rather just flag things as potentially infringing content and let the host decide, and if they host is using my directory service improperly, I block their ability to discover content
<rafajafar>
on top of my contentID system, there's a jury-of-peers approach to 2nd round matching.1st, thorough algorithm. If contested... 2nd, jury of peers. If contested, pay fee, we'll get a human look at it. If you lose, though, we keep the fee. If you win, we flag the other guy as a potential false flagger.
<rafajafar>
and we give the fee back
<rafajafar>
IPFS sounds ideal honestly... i woulda done this three years ago if IPFS was around
<rafajafar>
again, thoug, I need to figure out the permanence. I might charge permanance insurance, and if a file is losing seeds, I can throw up our seeder at a metered cost to the customer and send them an email alert.
micxjo` has quit [Read error: Connection reset by peer]
_fil_ has joined #ipfs
r04r is now known as zz_r04r
micxjo` has joined #ipfs
zz_r04r is now known as r04r
micxjo` has quit [Remote host closed the connection]
M-rschulman has quit [Quit: node-irc says goodbye]
<dignifiedquire>
M-davidar: works for me
<M-davidar>
it's not just me, there's a few people telling me they're having trouble too :/
ugjka has quit [Quit: Leaving]
niekie has quit [Ping timeout: 276 seconds]
chris6132 has joined #ipfs
niekie has joined #ipfs
chris6131 has quit [Ping timeout: 264 seconds]
Encrypt has joined #ipfs
<haad>
I'm also seeing a blank page
<Kubuxu>
Probably depends which gateway you get.
<Kubuxu>
dig ipfs.io
ugjka has joined #ipfs
<M-davidar>
I take it the v0.4.0 transition isn't going entirely smoothly?
<M-davidar>
I can't even access the site over ipfs :(
tmg has joined #ipfs
ugjka has quit [Quit: Leaving]
ipfsrocks has joined #ipfs
Guest14828 has joined #ipfs
ipfsrocks has quit [Ping timeout: 250 seconds]
zorglub27 has joined #ipfs
<lgierth>
it's back
<lgierth>
sorry for the inconvenience
<lgierth>
i need to set u palerts
G-Ray has joined #ipfs
<M-davidar>
:)
<M-davidar>
If the make install hangs while fetching dependencies and if you are running a daemon locally, ensure that it is on at least version 0.4.0-dev. If you don't have a build that is new enough, run the make install with your daemon turned off and the deps will be fetched through the ipfs.io gateways.
* M-davidar
turns off daemon, gx still tries to fetch through local daemon, fails miserably
<achin>
bielewelt: for what it's worth, adding large directories seems to work pretty well. i've not added something quite as large as 26GB, but i suspect that "ipfs add" is, at the moment, a bit more stable than "ipfs tar add"
<achin>
so if you can afford the disk space, unpacking the tar first might be another way forward
<daviddias>
hey richardlitt, shall we go back to the original schedule of sprints (libp2p -> js-ipfs -> go-ipfs)?
<daviddias>
bielewelt: the Datastore is only for DHT records and other meta stuff, files get stored in the blockstore
<bielewelt>
I see.
<daviddias>
there is no limit of file size, everything gets chunked
<daviddias>
however, it is good to know that adding with the tar command will untar the file and add each of its files, one by one
<bielewelt>
I thought the tar sub-command would improved de-duplication.
<lgierth>
it's just kind of a shortcut so that you don't have to unpack the tar before adding
<lgierth>
deduplication happens one way or the other
reit has quit [Ping timeout: 272 seconds]
leer10 has quit [Ping timeout: 260 seconds]
disgusting_wall has joined #ipfs
ipfsrocks has joined #ipfs
ppham has joined #ipfs
ashark has joined #ipfs
<richardlitt>
daviddias: yes!
ppham has quit [Ping timeout: 252 seconds]
M-matthew has joined #ipfs
M-david has joined #ipfs
M-nated has joined #ipfs
M-whyrusleeping has joined #ipfs
M-osener has joined #ipfs
M-jbenet has joined #ipfs
M-amstocker has joined #ipfs
M-eternaleye has joined #ipfs
M-trashrabbit has joined #ipfs
M-ikreymer has joined #ipfs
M-mistake has joined #ipfs
M-oddvar has joined #ipfs
HostFat_ is now known as HostFat
M-kode54 has joined #ipfs
M-hash has joined #ipfs
M-cryptix has joined #ipfs
M-victorm has joined #ipfs
M-edrex has joined #ipfs
M-erikj has joined #ipfs
M-jgrowl has joined #ipfs
M-roblabla has joined #ipfs
M-davidar has joined #ipfs
M-kegan has joined #ipfs
M-mubot has joined #ipfs
M-davidar-test has joined #ipfs
M-jfred has joined #ipfs
M-rschulman1 has joined #ipfs
ovoviviparousnes has left #ipfs [#ipfs]
M-giodamelio has joined #ipfs
M-alien has joined #ipfs
M-cocci has joined #ipfs
M-zick has joined #ipfs
M-staplemac has joined #ipfs
ovoviviparousnes has joined #ipfs
M-Ingo has joined #ipfs
M-GereG has joined #ipfs
M-fil has joined #ipfs
M-ryanwilliams has joined #ipfs
M-Peer2Peer has joined #ipfs
M-harlan has joined #ipfs
M-rschulman has joined #ipfs
M-matthew has quit [Quit: node-irc says goodbye]
<daviddias>
cool, thank you. Could you update the sprint issue then, for everyone following
<lgierth>
richardlitt: posted my update -- i'll have to head out in a bit and will be back at 7pm utc for infrastructure hangout and the other hangouts
bielewelt has quit [Quit: Leaving.]
<lgierth>
ok /me out, see you later
<richardlitt>
lgierth: sounds good
<richardlitt>
sorry, was testing a bear.
<richardlitt>
thanks Kubuxu, too
<richardlitt>
30 minutes until sprint.
cemerick has joined #ipfs
<noffle>
o/
<voxelot>
\o
<richardlitt>
noffle: o/
<haad>
\o
<richardlitt>
~~~~o/
ygrek has joined #ipfs
<alu>
\o/
<richardlitt>
./o\.
<richardlitt>
>o< (It's a tie fighter)
<voxelot>
<(^.^<)<(^.^)>(>^.^)>
<daviddias>
|-o-|
tawar has joined #ipfs
<dignifiedquire>
whaaats happening
<alu>
memes
<daviddias>
zummm zummm
<daviddias>
pew pew
<richardlitt>
pew pew pew
* noffle
is installing nixos on his new laptop, to evaluate as a dev environment
<tawar>
naive question: In order to enable ipfs ... would a disconnected private deployment be possible. I.e., private network -- no access to internet. Ever. Or is that... non-trivial?
M-jon has quit [Remote host closed the connection]
Matoro has quit [Ping timeout: 248 seconds]
<achin>
once data has been downloaded to an ipfs node, it can then be disconnected from the internet and still happily serve its downloaded content to local clients
<noffle>
daviddias: a semi-recent 12" thinkpad -- not going to fall into the trap of non-upgradable macbooks again
<dignifiedquire>
richardlitt: oh boy, there was not much to study in January 2013, was there :D
devbug has joined #ipfs
<richardlitt>
Actually, I failed my exams that semester
<richardlitt>
So.... awesome.
<dignifiedquire>
:D
<dignifiedquire>
I wonder why
<daviddias>
an x250?
<achin>
tawar: you have to interact with the larger peer swarm to get the content initially
<tawar>
achin, Lemme be explicit: If I had a zillion node private network. Global. But no data can exist outside of the internal network. (a private CDN) can that be implemented with IPFS as it exists today?
cemerick has quit [Ping timeout: 240 seconds]
<achin>
but after that, you don't need to. if you request QmHashFoo from your local ipfs node, and your node already has that content locally, it can serve it directly and immeditally, without contacting the swarm
<Kubuxu>
tawar: nope
<tawar>
Kubuxu, Thanks.
<tawar>
Kubuxu, RFE! ;)
<noffle>
daviddias: x140e. $200 on ebay new and slightly better spec'd than my 2012 macbook air
<Kubuxu>
There is plan to allow connections with pre-shared key.
<achin>
a github issue might already exist for this scenario
* tawar
nods
<noffle>
daviddias: the height is a bit weird to get used to vs the flat air
<noffle>
some minor wrist pain
<daviddias>
noffle: wow, what a deal
<daviddias>
haven't played with any yet, but it looks good on the pictures
<noffle>
daviddias: yeah! I can probably make $300 profit selling the mba
micxjo has joined #ipfs
<richardlitt>
Alright!
<noffle>
daviddias: what I'm starting to realize is that macro work is great (installing packages, rolling back, reproducible systems), but micro can be challenging (symlink soup, can't do ad-hoc system tweaks without system rebuilds nuking your changes)
<richardlitt>
We all ready for this sprint?
<richardlitt>
I'll take that resounding answer as a ... yes?
<noffle>
daviddias: but that may be acceptable. I expect to manage my homedir ad hoc, but nixos for system-level config)
<noffle>
richardlitt: +1
<dignifiedquire>
ready
drnet has joined #ipfs
<richardlitt>
Alright. First up we have lars, who... is not here
<daviddias>
noffle: true, the lower the level, it gets harder to find all of the conviniences that we have on the applications level for fast iteration and testing cycles
<richardlitt>
So, I'll just C&P his stuff, I guess?
<whyrusleeping>
Bus is taking longer than expected. Eta ten min
<dignifiedquire>
whyrusleeping: tell the driver to hurry ;)
<noffle>
daviddias: well said :)
<whyrusleeping>
I guess things are slower in construction season
<richardlitt>
whyrusleeping: want us to wait?
<richardlitt>
Or can you follow along via chat?
Matoro has joined #ipfs
<noffle>
daviddias: playing with nix will also help me work on getting them ipfs support
<daviddias>
nice!
<daviddias>
it will also help you get it working faster :D
<richardlitt>
Lars incoming ====
<noffle>
exactly
<richardlitt>
### @lgierth sprint update
<richardlitt>
This was a rather slow sprint for me -- plenty of good gateway debugging with @noffle and @whyrusleeping though.
<richardlitt>
- 0.4.0
<richardlitt>
- [x] blog post about v04x/v03x, tweeted ipfs/blog#27
<richardlitt>
- [x] fixed blog post ipfs/blog#28
<richardlitt>
- [~] update pinbot
<richardlitt>
- [x] refine multireq's status code precedence whyrusleeping/multireq@152e37c
<richardlitt>
- [x] CR
<richardlitt>
- docker
<richardlitt>
- [ ] ~~daily cronjob for jbenet/go-ipfs~~ handed to @jbenet: ipfs/ops-requests#20
<richardlitt>
- [~] blog post
<richardlitt>
- [ ] ~~profile polishing~~
<richardlitt>
- [x] branches and tags
<richardlitt>
- infrastructure
<richardlitt>
- [x] debug gateway v03x hangs
<richardlitt>
- [x] debug git + ipfs.io hangs
<richardlitt>
- [ ] captain's log
<richardlitt>
- [ ] tests
<richardlitt>
- [ ] simpler website deployments
<richardlitt>
- others
<richardlitt>
- [x] feedback on gx + go get whyrusleeping/gx-go#2
<daviddias>
- [~] `js-ipfs files add` and `add -r` https://github.com/ipfs/js-ipfs/pull/67 (it works, but I want to merge it when I get the nested dirs issue figured out)
<daviddias>
- I listed this before as 'extra', but it really is part of libp2p, namely, the whole endeavour of packet switched networks over RTCDataChannels:
<daviddias>
- [~] https://github.com/diasdavid/webrtc-explorer/pull/10 - Almost there! Really really happy with the reboot, the amount of coverage, quality of the API and structure of the code, makes webrtc-explorer so much more useful
<daviddias>
- [x] use and work with Jeromy at debugging problems in IPFS that were causing registry-mirror to crash. The last one was a oomd, but there is a patch on go-ipfs landing ['soon'](http://static.fjcdn.com/pictures/Soon_15db89_2831712.jpg)
<richardlitt>
After dignifiedquire's question is answered. :D
<daviddias>
dignifiedquire have several groups of tests and then in those groups, bring browsers to join the network progressively and not all at the same time
<dignifiedquire>
daviddias: okay, ta
<richardlitt>
Sweet.
<dignifiedquire>
richardlitt: go go go
<richardlitt>
Beginning:
calavera has joined #ipfs
<richardlitt>
This week, I finished PRing the HTTP API, basically. Now it just needs a lot of CR from people.
<richardlitt>
- [x] Sprint Management
<richardlitt>
- [x] Write the weekly roundup
<richardlitt>
- [ ] Send it out
<richardlitt>
- [x] IPFS API :star: :star: :star: - I finished up the last of the PRs.
<richardlitt>
Please check out these pulls this week, other people: github.com/ipfs/http-api-spec/pulls. Need some overview before merging.
<richardlitt>
Also, the weekly is in a bit of dire straights. Probably going to make this week a double. I don't know, juan always wants more for it, but it's hard for me to source stuff. If you all have anything cool, please add it to the weekly
<noffle>
richardlitt: can you link folks here to where they'd submit that?
<richardlitt>
yeah. It's... really annoying right now, because there's a weekly for the week before last, there should be one for last week, and now I need to make one for this coming week, but the one from three weeks ago is still open
<richardlitt>
that makes like three different places people could submit stuff
<dignifiedquire>
this seems to be hanging there some time now
<richardlitt>
no, the most recent version of the go-ipfs repo
<richardlitt>
because I don't want to do doc fixes for docs that are already changed
<richardlitt>
ok
Codebird has joined #ipfs
<richardlitt>
thanks all for helping out
<richardlitt>
anything else?
<dignifiedquire>
richardlitt: let's plan for a day of content checking together for those blog posts and the install docs later this week
<whyrusleeping>
<dignifiedquire>
so we can get this out of the door
<richardlitt>
dignifiedquire: sounds good
Tv` has joined #ipfs
Protocol_under has joined #ipfs
<richardlitt>
dignifiedquire: PR the distro, and I'll go forth with the other four areas
<richardlitt>
or three, whatever they are
<dignifiedquire>
richardlitt: okay
<richardlitt>
noffle: go ahead, I think it's your turn
<Codebird>
I seem to have found a key which, if you attempt to pin it, crashes the daemon. Strange. Wonder what's doing on there.
* noffle
---
<noffle>
Summary: some go-ipfs + js-ipfs development, and ipget planning. ipfs-hyperlog gives us access to the [hyper* ecosystem](https://www.npmjs.com/search?q=hyperlog), which has modules to let ipfs app developers build things like kv-stores and event logs that work in node+browser. This could help in quickly getting pubsub off the ground, too.
<daviddias>
Codebird: which key?
<whyrusleeping>
Codebird: file an issue with the key please, thats really interesting
<daviddias>
noffle: It would be really great if we can bring ipfs-dag interface enhancements to ipfs-merkle-dag
<Codebird>
Could someone test that and see if it's something about that key, or a problem with my instance alone?
<noffle>
daviddias: yeah I'd really like that. and basically make ipfs-merkle-dag into ipfs-dag-service
<whyrusleeping>
noffle: looks pretty good, i had some feedback on some of your PRs
<daviddias>
we can even break ipfs-merkle-dag into ipfs-merkle-dag-service and ipfs-merkle-dag-node
<whyrusleeping>
not sure if you saw that yet
<noffle>
whyrusleeping: I saw! getting to them soon :) thanks
<noffle>
daviddias: yep. hopefully the author is amenable
<daviddias>
noffle: ah, for sure, vijayee is great, I've full push perm and code is MIT licensed
<daviddias>
I already had factored out block and block service in the past
<noffle>
daviddias: great; he sounded agreeable
<Codebird>
It crashes in under a minute for me - not just the error I pasted, but the daemon terminates too.
<whyrusleeping>
Codebird: thats no bueno. I'll take a look
<Codebird>
Does it crash you as well?
<whyrusleeping>
noffle: anything blocking you?
gordonb has joined #ipfs
<whyrusleeping>
Codebird: havent checked, focusing on sprint stuff ATM
<whyrusleeping>
Codebird: what version of ipfs are you running?
<noffle>
whyrusleeping: no blockers
<Codebird>
0.4.0-dev
<whyrusleeping>
noffle: alright, cool.
<whyrusleeping>
anyone else have anything for noffle?
<Codebird>
Upgraded from 0.3.something, I had to run the migration tool.
<daviddias>
re: ipfs-dag name; I've mixed feelings calling it ipfs-dag as it might confused people, DAG is very generic name which would work for an IPLD graph or a MerkleDAG graph.
<daviddias>
We have been calling the MerkleDAG our protobuf implementation and IPLD our JSON+CBOR
<daviddias>
Just one more thing
yellowsir has quit [Quit: Leaving.]
<daviddias>
let's write the core API docs and start with the merkle-dag-service and merkle-dag-node
<daviddias>
richardlitt, wanna help on that end too?
<richardlitt>
sgtm
<richardlitt>
sure
<daviddias>
awesome :D
<whyrusleeping>
dignifiedquire: wanna go now?
<daviddias>
ok, whyrusleeping that is all I had
<noffle>
daviddias: yeah I'm happy to visit new naming ideas
<dignifiedquire>
sure
<richardlitt>
dignifiedquire: go ahead
<richardlitt>
thanks whyrusleeping
<daviddias>
thank you, noffle :)
* dignifiedquire
incoming
<richardlitt>
dias: make a planning issue for those docs?
<dignifiedquire>
## @dignifiedquire sprint update
<dignifiedquire>
### Webui
<dignifiedquire>
- Redux PR
<dignifiedquire>
- [x] Streaming logs
<dignifiedquire>
- [~] Peers page
<dignifiedquire>
### Distributions
<dignifiedquire>
- Code Signing
<dignifiedquire>
- [x] Added document about the signing process for review
<daviddias>
"- [x] increase dht K-Value to 20 (from 10)" decrease then, right? Or is it now 20?
<strings>
hello, how do I get these gx import paths to work with go get?
<richardlitt>
That is a lot of merges.
<whyrusleeping>
daviddias: ;to 20'
<dignifiedquire>
whyrusleeping: ipns a little faster, how much are we talking?
<daviddias>
whyrusleeping: oh, wasn't it 20 before?
<voxelot>
2x the peers in bucket = 2x the speed? :)
<daviddias>
was that done in order to increase IPNS?
<Kubuxu>
dignifiedquire: from 10s each resolution to ~2.5
<whyrusleeping>
dignifiedquire: not sure percentage wise, but i'm seeing two second resolves
<noffle>
whyrusleeping: richardlitt: do we have a doc on this now (see strings msg above)?
<whyrusleeping>
strings: uhm, what do you mean? you want to be able to fetch them with go get?
<daviddias>
voxelot: the double of file descriptors
<Kubuxu>
daviddias: we require 16 confirmations for record from DHT, we were publishing new record only to 10 nodes
<richardlitt>
noffle: not sure what you're talking about
<dignifiedquire>
whyrusleeping: it's something :)
<whyrusleeping>
daviddias: it was 10 before in the kbuckets
<strings>
import "gx/ipfs/QmQopLATEYMNg7dVqZRNDfeE2S1yKy8zrRh5xnYiuqeZBn/goprocess": import path does not begin with hostname
<daviddias>
Kubuxu: and that required increasing the kbucket?
<noffle>
strings: you'll need to do a git clone of the go-ipfs repo into its proper location, then run make -- we're working on this
<Kubuxu>
daviddias: that is question to whyrusleeping
<whyrusleeping>
strings: we don't support using go get anymore, it makes it really hard to vendor your code properly
<strings>
I can build locally
<daviddias>
whyrusleeping: I remember having a conversation in the Summer about reducing the k-bucket size from 20 because it were opening (or trying to) too many connections
<strings>
assumeing gx is a bootstrapping issue?
<richardlitt>
ahh
<richardlitt>
afaik we don't have a doc for this
<noffle>
strings: yes it's not fully integrated into 'go get' yet :)
<daviddias>
chriscool: I've a summary there why it was created (https://github.com/diasdavid/piri-piri#why-does-it-exist) but the gist is that in order to test WebRTC enabled apps, specially if they involve 2 or more browsers at the same time, conventional browser testing frameworks don't work, because they run tests on the browser individually and there is no way to
<daviddias>
orchestrate several browsers at the same time
<noffle>
chriscool: nice
<daviddias>
chriscool: nice, it is super cool how IPFS always keeps Windows on the loop
<richardlitt>
chriscool: awesome work
Matoro has joined #ipfs
<chriscool>
thanks all!
piririgua has joined #ipfs
<dignifiedquire>
chriscool:
<richardlitt>
nginnever: you here?
<voxelot>
yup! real quick list
<voxelot>
js-peer-id
<voxelot>
[x] run karma + webpack tests
<voxelot>
[x] get PR merged (ty daviddias)
<voxelot>
js-ipfs
<voxelot>
[x] node-forge with webpack (ty dignifiedquire)
<voxelot>
Webui
<voxelot>
[x] found some things to work on here
<voxelot>
EOF
<dignifiedquire>
voxelot: join us for the apps hangout, will be talking about the webui :)
<richardlitt>
voxelot: :D
<voxelot>
will do
<noffle>
voxelot: what were the things you found to work on re webui?
<daviddias>
voxelot: awesomely done with forge and getting all of that working
<Kubuxu>
I have one issue with Windows if someone want to check it out: start daemon, Ctrl-C it, start it again, it will complain that other daemon is running (lock file is in .ipfs), start it again, it starts normally
<voxelot>
Webui
<voxelot>
[ ] begin redux saga (issue #227)
<voxelot>
[ ] research enzyme tests on components
<voxelot>
[ ] document in contibuting.md
<noffle>
ah ha :)
<voxelot>
just getting started
<daviddias>
voxelot: I remember you mentioning that you had a PR for js-ipfs with the new js-peer-id, can you PR it for CR and Merge?
<noffle>
didn't realize you culled all of the non-complete items from the issue
<richardlitt>
Is that everyone?
<voxelot>
daviddias: yeah thats on my TODO list, wasnt sure i f i should put TODO's or just what is done for sprint
ppham has joined #ipfs
<daviddias>
here is just what is done
<voxelot>
gotcha
<richardlitt>
OK
<richardlitt>
I think tha'ts it
<richardlitt>
Thanks all!
<daviddias>
thank you for herding :D
<noffle>
richardlitt: thanks for making it happen :)
<richardlitt>
np. :)
<dignifiedquire>
thanks everyone for making awesome things!
<richardlitt>
OK!
<richardlitt>
An the first talk is in twenty minutes
<chriscool>
dignifiedquire and daviddias thanks for the explanations and links about karma-peer and piri-piri it's interesting!
<richardlitt>
Everyone who did awesome stuff this week, if you could update the PR with a comment or a commit for the weekly, that would be super awesome. https://github.com/ipfs/weekly/pull/20
<richardlitt>
Otherwise, I'll go through the logs here later today and pull stuff out and then ping people to CR it
<ipfsbot>
[go-ipfs] whyrusleeping closed pull request #2342: Add information about installing gx into readme (master...feat/gx-in-readme) https://git.io/vg9OR
<dignifiedquire>
whyrusleeping: you coming?
<dignifiedquire>
you were so excited
<whyrusleeping>
i am
Matoro has quit [Ping timeout: 264 seconds]
dawuud has joined #ipfs
Pharyngeal has quit [Ping timeout: 272 seconds]
<noffle>
richardlitt: oh that's awesome! really thrilled someone did this :D :D :D
<richardlitt>
yeah
Pharyngeal has joined #ipfs
Matoro has joined #ipfs
<noffle>
richardlitt: I think it's reasonable to link to from the nodeschool.io website, but maybe it doesn't belong under the nodeschool family of workshoppers
voxelot has quit [Ping timeout: 264 seconds]
<noffle>
richardlitt: whyrusleeping: daviddias: I regret us not leaving an ipfs node hidden in our host's house on svalbard
<noffle>
that'd be awesome to see on the webui world map
<richardlitt>
noffle: lol
<richardlitt>
we should contact him
<richardlitt>
or the guy who runs the kebab shop
<whyrusleeping>
noffle: i can ask him on facebook to set one up
<whyrusleeping>
we should ship him a raspberry pi if we want to do that though
<noffle>
whyrusleeping: agreed
<daviddias>
richardlitt: has two swords behind him
strings has quit [Quit: Page closed]
<daviddias>
sweet
zorglub27 has quit [Ping timeout: 272 seconds]
<richardlitt>
heh.
<richardlitt>
Infrastructure will be running a little bit late today.
<richardlitt>
lgierth: you here, btw?
<Kubuxu>
1. you would have to create .dep 2. get it into repos, this is the hard part
<Kubuxu>
yangwao: yes, just some peer told you that there might be IPFS node there, and your node checked, there was not
<yangwao>
remembers me like pouchdb told me, this 404 is totally normal :D
<Kubuxu>
dignifiedquire: did it work?
mkelly has joined #ipfs
<dignifiedquire>
I think this one works glob.sync("+(!(test))/**/*.js")
<dignifiedquire>
Kubuxu: thanks :)
ppham has quit [Remote host closed the connection]
<Kubuxu>
We could have just go get print better error message.
<noffle>
aw, the display may be borked on this thinkpad -- the bios display diagnostic showing a full blue screen causes the display to become corrupted
<whyrusleeping>
dignifiedquire: yeah, that works-ish, but i want to also be able to move issues once i've addressed them, example:
<whyrusleeping>
i code review something, and i need the author to address some feedback
<whyrusleeping>
so the issue isnt 'complete'
<whyrusleeping>
but its not my problem at the moment
<noffle>
dignifiedquire: whyrusleeping: my sprint setup looks like this: http://www.stephenwhitmore.com/tmp/org.png you can build deep trees for hard/multi-layered problems, but also collapse sections to avoid screen clutter
<dignifiedquire>
whyrusleeping: trello?
<whyrusleeping>
dignifiedquire: closer...
<whyrusleeping>
downside is everyone has to use trello
<noffle>
mark status on TODOs (WAITING, DONE, etc)
<dignifiedquire>
noffle: that's neat, going to try this out
<noffle>
and it's all plaintext, so you can write a script to tranform a sprint orgmode file into e.g. a github sprint issue comment
<whyrusleeping>
noffle: yeah, i'll give org mode a try
<noffle>
whyrusleeping: dignifiedquire: it's a deep rabbit hole and requires lots of iterative customization, so prepared to invest time as the trade-off
ygrek_ has quit [Ping timeout: 265 seconds]
<noffle>
but it's lovely when you get a setup you like
<dignifiedquire>
noffle: I'm used to that in emacs, that's why I can never switch anywhere else, so many things are suddenly missing
<noffle>
orgmode is all I use emacs for -- vim for everything else :)
<dignifiedquire>
lol
<dignifiedquire>
I use emacs for everything, except for shell
<dignifiedquire>
:D
infinity0 has quit [Remote host closed the connection]
<dignifiedquire>
only missing photoshop-mode
<noffle>
heh
metaf5 has quit [Quit: WeeChat 1.3]
<whyrusleeping>
yeap, i'm going to miss the 12:30 meeting. i can make it at 1:30
Encrypt has quit [Quit: Quitte]
`Ronsor is now known as superfish
superfish is now known as Ronsor
m0ns00n2 has quit [Quit: undefined]
Ronsor is now known as Guest67988
infinity0 has joined #ipfs
<richardlitt>
whyrusleeping: I use a massive text file.
<Kubuxu>
whyrusleeping: try: taskworrior
metaf5 has joined #ipfs
Guest67988 is now known as Ronsor`
mkelly has joined #ipfs
<dignifiedquire>
whyrusleeping: will you be around for go-ipfs?
<redfish>
i have peers, but lsof tells me that all are outgoing connections, not incoming
<yangwao>
oh
nicolagreco_ has joined #ipfs
<redfish>
my non-NATed node on a VPS does get incoming connections, so I think IP discovery is not actually working
<yangwao>
firewall?
Matoro has joined #ipfs
Matoro has quit [Read error: Connection reset by peer]
<redfish>
nope, because I can telnet directly to 4001 from outside
<redfish>
my hypothesis is that it never actually discovers its external IP
<noffle>
whyrusleeping: I want to get #1 and #2 done first, but maybe after that we could do a chat to discuss what needs to happen (in more detail)?
Matoro has joined #ipfs
mkelly_ has joined #ipfs
<yangwao>
looks me either
<whyrusleeping>
noffle: sure thing
<yangwao>
idk, maybe some issue with mdns?
<whyrusleeping>
doesnt sound like an mdns issue
<whyrusleeping>
thats very weird
<whyrusleeping>
whats your peerID?
rendar has quit [Ping timeout: 252 seconds]
<yangwao>
wow, telnet shoots of ipfs daemon
<yangwao>
thats really weird.
<redfish>
if that helps info: my two boxes (one NATed other not-NATed) happen to be connected via a VPN and the two ipfs nodes do get connected to each other via the tunnel interface (automatically)
<yangwao>
peerID of my ARM board?
<redfish>
whyrusleeping: mine or yangwao peer ID?
<whyrusleeping>
redfish: yours
<redfish>
sorry, I don't have it handy right now, will report back later
<Kubuxu>
It is ok, it was already degraded from ERROR in new libp2p
<whyrusleeping>
yeah, thats normal. It will be downgraded from an error to a warning
xelra has quit [Remote host closed the connection]
<whyrusleeping>
jinx
<yangwao>
mkey, how to help ipfs network, I should expose 4001?
<Kubuxu>
It should be even lower, IMO.
chriscool has quit [Quit: Leaving.]
<Kubuxu>
Because it will show up in normal work.
<yangwao>
(because I make notes and some manual for others:)
chriscool has joined #ipfs
xelra has joined #ipfs
<yangwao>
should I add external ip if I forward ports?
Oatmeal has joined #ipfs
<whyrusleeping>
yangwao: i don't think so? that should be dynamic
* Codebird
grumblegrumble... how can there be another Codebird on github already?
chriscool has quit [Quit: Leaving.]
Matoro has joined #ipfs
tlevine has quit [Ping timeout: 248 seconds]
<Codebird>
Fine. Codeburd it is.
calavera has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
chriscool has joined #ipfs
<yangwao>
whyrusleeping: idk, but how do you connect to me then?
<whyrusleeping>
yangwao: if your router supports upnp or nat-pmp it does the forwarding automatically
<yangwao>
mm mikrotik here..
calavera has joined #ipfs
Matoro has quit [Read error: Connection reset by peer]
Matoro has joined #ipfs
nicolagreco_ has quit [Quit: nicolagreco_]
chriscool has quit [Quit: Leaving.]
<whyrusleeping>
anyone have a corporate subscription to gandi?
chriscool has joined #ipfs
learningCljs has quit [Ping timeout: 264 seconds]
ylp has quit [Ping timeout: 240 seconds]
Matoro has quit [Read error: Connection reset by peer]
Matoro has joined #ipfs
joshbuddy has quit [Quit: joshbuddy]
<ipfsbot>
[js-ipfs-api] dignifiedquire created greenkeeper-concurrently-2.0.0 (+1 new commit): https://git.io/vgFfO
<ipfsbot>
js-ipfs-api/greenkeeper-concurrently-2.0.0 07ed707 greenkeeperio-bot: chore(package): update concurrently to version 2.0.0...
<sobr>
so, to make sure I understand this correctly, with "ipfs name publish <hash>" it associates a hash with your peer id so that resolving your peer id returns whatever hash is currently associated with it?
<whyrusleeping>
sobr: simply put, yes
<dignifiedquire>
whyrusleeping: when you have a moment, could you looka this repo https://github.com/flynn/go-tuf and let me know if this is code you'd be comfortable of integrating into ipfs-update?
<ipfsbot>
[js-ipfs-api] dignifiedquire created greenkeeper-babel-eslint-5.0.0 (+1 new commit): https://git.io/vgFmb
<ipfsbot>
js-ipfs-api/greenkeeper-babel-eslint-5.0.0 54df2f2 greenkeeperio-bot: chore(package): update babel-eslint to version 5.0.0...
palkeo has quit [Ping timeout: 264 seconds]
Matoro has joined #ipfs
ppham has joined #ipfs
<ipfsbot>
[js-ipfs-api] dignifiedquire deleted greenkeeper-babel-eslint-5.0.0 at 54df2f2: https://git.io/vgFYy
palkeo has joined #ipfs
ppham has quit [Remote host closed the connection]
ppham has joined #ipfs
ppham has quit [Remote host closed the connection]
pfraze has joined #ipfs
<montagsoup>
Thank you so much for not putting annoying little animations all over the IPFS main site. I am getting so sick of scrolling down a page and seeing things fade in and move.