notduncansmith has quit [Read error: Connection reset by peer]
pfraze has joined #ipfs
<simoon>
I just had to start the daemon GEEZ
<simoon>
geeeeeez guys
<simoon>
thanks ;)
<simoon>
peace
simoon has left #ipfs [#ipfs]
pfraze has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
sharky has quit [Ping timeout: 260 seconds]
sharky has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
inconshreveable has quit [Remote host closed the connection]
iampims has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<iampims>
Hi there. I've got a question for which I couldn't find any documentation. I’d like to build go apps on top of ipfs and was wondering what the simplest way to connect to the local daemon was. Right now I am using something like `addCmd := exec.Command("ipfs", "add", fname)` but I’d much rather use a client to send a command. Can anyone point me in the right direction. I've tried to emulate what the cli does, but without much
fleeky has quit [Remote host closed the connection]
joshbuddy has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
iampims has quit [Quit: Leaving...]
<whyrusleeping>
davidar: wassup!!
<whyrusleeping>
its friday night, so im not technically here, but whatever
joshbuddy has quit [Ping timeout: 244 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
joshbuddy has joined #ipfs
inconshreveable has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Leer10 has quit [Ping timeout: 246 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ygrek_ has joined #ipfs
joshbuddy has quit [Quit: joshbuddy]
fps has joined #ipfs
<fps>
hi, just heard of ipfs from a visitor in #morphis
<spikebike>
greetings
<spikebike>
have you used #morphis?
f[x] has joined #ipfs
Encrypt has joined #ipfs
ygrek_ has quit [Ping timeout: 250 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<fps>
spikebike: yes
devbug_ has joined #ipfs
<fps>
soo, the implementation of ipfs to check out it the go one?
<spikebike>
yeah, both are being developed, but I think the go one is more functional
Encrypt has quit [Client Quit]
<fps>
i skipped through the ipfs paper once
<spikebike>
fair bit there isn' working yet
<fps>
but admit i only really skipped through it. i guess i should return to it before asking silly questions :)
captain_morgan has quit [Ping timeout: 246 seconds]
<spikebike>
DHT works, finding peers works, static files, video playback of streams, etc.
<fps>
ah cool, somewhat similar to morphis then
<spikebike>
encryption doesn't
<fps>
what are anonymity and privacy concerns
<fps>
oh
<spikebike>
but is fairly high on the priority lsit
rendar has joined #ipfs
<fps>
cool
<spikebike>
the bitcoin type stuff doesn't work
<fps>
i wonder if morphis and ipfs differ too much for the authors to collaborate :)
<spikebike>
yeah so no privacy/anonymity, encrypt before you put stuff in ipfs for now
devbug_ has quit [Ping timeout: 265 seconds]
<fps>
i mean when you have the key for a value you get it back in clear at your endpoint anyways, right
ygrek_ has joined #ipfs
<fps>
so you're just talking about interpeer traffic that's not encrypted yet?
<spikebike>
I'm not one of the devels, just tinkered a bit
<fps>
ah, same here with regard to morphis :)
<spikebike>
might be transport encrypted, but nothing to prevent anyone from slurping your files if they know the hash
<fps>
yeah, that's natural though
<fps>
i'd say
<fps>
if you want confidentiality of your file contents, encrypt them yourself :)
f[x] has quit [Ping timeout: 264 seconds]
<fps>
interesting that ipfs aims at providing that in the system itself
<fps>
if i understood you right
gaboose_ has joined #ipfs
<spikebike>
ipfs is pretty open, there's discussions of using cjdns or whatever that encrypted networking layer is called
<spikebike>
but yeah ideally you could share/publish/sync files with people of yoru choosing
<spikebike>
caching works pretty well
<spikebike>
and the performance has impressed me
<fps>
btw. if you just want to play with retrieving stuff from morphis
<fps>
i have setup a rate limited proxied node on fps.io:555
<fps>
morphis home page:
<fps>
fps.io:555/sp1nara
<fps>
my home page:
<fps>
fps.io:555/fps7j77/index.html
<fps>
rate limited to 1k/s per connection though ;)
<davidar>
whyrusleeping: hey, just wrestling with pinbot :)
<fps>
spikebike: oh, never heard of cjdns
<fps>
hmm
<fps>
in that node installation i also reduced the required number of bits of an id to 1 to do a prefix search..
borgtu has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<fps>
i wonder if one could do a user space filesystem on top of either ipfs or morphis
<fps>
hmm
<fps>
does ipfs have versions of files under the same key?
<spikebike>
ipfs has a fuse module/mount
<fps>
coool
<spikebike>
or just use ipfs t interact
<spikebike>
no key=sha256(file) basically
<spikebike>
but there is IPNS for name -> file mappings that aren't content based
<fps>
oh interesting
<fps>
yeah, the key = hash(file) is basically the same in morphis
<spikebike>
and the merkledag is kidna git like for handling changes/diffs, etc.
<fps>
though there's a notion of updateable keys which i haven't fully grokked yet
<fps>
but which i used for fixing stuff in my homepage :)
<fps>
internally it uses some versioning
<fps>
hmm
Leer10 has joined #ipfs
akhavr1 has joined #ipfs
akhavr has quit [Read error: Connection reset by peer]
akhavr1 is now known as akhavr
kerozene has quit [Ping timeout: 244 seconds]
bedeho has quit [Ping timeout: 240 seconds]
nomoremoney has joined #ipfs
pfraze has joined #ipfs
VegBerg has joined #ipfs
<nomoremoney>
Hey guys, stupid question here that I tried to figure out myself but I haven't been able to so far. I'm playing around with IPFS and thought it would be fun to build a twitter clone on top of IPFS. My idea is to have some sort of shared datastore that each node that runs the server/client gets. But I'm unsure how to deal with changes and how to notify all the clients of the change. Anyone have any idea or pointers for me what I can
<nomoremoney>
look into to get started?
kerozene has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
pfraze has quit [Ping timeout: 244 seconds]
giodamelio has joined #ipfs
<gaboose_>
ive heard a pub/sub type of communication between nodes are in their plans
<gaboose_>
Not sure what the progress of it is, but that could be used to notify changes
devbug_ has joined #ipfs
<nomoremoney>
@gaboose_ that sounds like something that can help me. Is there any information out there about that?
<davidar>
fps, not sure, have to wait until it's worked out in more detail I guess
<nomoremoney>
@davidar, awesomeness! Seems like re-captcha doesn't like the domain name though
<nomoremoney>
fyi
ygrek_ has quit [Ping timeout: 252 seconds]
<fps>
btw, does ipfs use kademlia's property of keys and node ids sharing the same address space?
<davidar>
nomoremoney: oh, what error are you getting?
<fps>
and its way of looking up nodes using the shortest paths with respect to the XOR metric?
<nomoremoney>
davidar, "ERROR: Invalid domain for site key" in the re-captcha widget
<davidar>
nomoremoney: during signup?
<nomoremoney>
davidar, when confirming the signup, after clicking the verify link in the email
<davidar>
nomoremoney: which domain does the verify link take you to? ipfs.io?
<davidar>
cc Matthew
<fps>
if so, i might just have thought about a way to do away with polling and just using kademlia update propagation to do pub/sub :)
<fps>
*thought of
<davidar>
fps: not sure, check the ipfs whitepaper, but yet, ipfs is heavily based on kademlia
<nomoremoney>
@davidar, yup ipfs.io
<davidar>
fps: cool, I'm not sure if there's an issue for submitting ideas about pubsub, but I can create one
<davidar>
for you if you like?
<spikebike>
ipfs is planning on pub/sub, but the approach hasn't been pinned down afaik
<davidar>
nomoremoney: thanks for the bug report, I'll try to get it fixed
<nomoremoney>
davidar, on second look, the link in the email points to vector.im but redirects me to ipfs.io
<fps>
davidar: i have a github account myself..
<fps>
but let me just spin it to you
<fps>
i'm coming from a morphis background though, so details might differ.. and it might not be applicable to ipfs
<spikebike>
fps: if you write it up anywhere post the link
<davidar>
fps: sure, I meant just creating an umbrella issue that you could attach a comment to :) but you're welcome to do it yourself ofc
<fps>
davidar: go ahead :)
<fps>
in a nutshell: in morphis documents can be updated while retaining their key
<davidar>
fps: you're welcome to discuss it here first if you like, but I don't think any of the core team members are around atm :)
<fps>
so if you just setup a node with a key close to the document you're interested in being notified of changes of
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<fps>
then kademlia update propagation will request your node to store the new file under the key
<fps>
so a subscription would just be setting up a node. ipfs uses S/Kademlia so you have to do a POW to generate a key that's close to the key you're interested in getting update notifications of
<davidar>
ah ok, was just about to ask how that would work
<nomoremoney>
fps, I keep reading about morphis but can't find anything about it with google. Could you maybe provide a link for curious people?
<davidar>
is that stable?
<fps>
yes
<fps>
i setup a proxied node under fps.io:555
<fps>
rate limited to 1k/s
<davidar>
I know that the ipfs network likes to rebucket stuff depending on participating nodes
<fps>
dmail and uploads are broken through that proxy
<fps>
but you can retrieve keys like fps.io:555/sp1nara
<davidar>
no, I mean can you always assume that your generated key will be "near" the document key, if the network changes
<nomoremoney>
fps, is a project of yours? Looks interesting for sure
<fps>
nomoremoney: nah, just an active user of it :)
<fps>
nomoremoney: thufir in #morphis is the main dev
<fps>
davidar: a general property of kademlia routing is that nodes and document keys share the same adress space
<M-matthew>
davidar: ah, hm, yes
<M-matthew>
captcha will be broken if the domain doesn't match
<fps>
davidar: for static files the key is just the hash of the document like in ipfs
<fps>
morphis has a notion of updateable keys though.. [which if less than 32k just gets stored under that key - if more, there's a merkle tree yadayada]
<M-matthew>
not much we can do about that off the top of my head, short of a different strategy to captcha during signup
<M-matthew>
will have a think
<davidar>
Matthew: apparently the confirmation email links to vector.im, but then redirects to the originating domain?
<fps>
so, in kademlia documents are stored on nodes who's ids are close to the document key with respect to the xor metric
<fps>
so, just create a node with an id close to the document you're interested in and you'll get store requests for it
<davidar>
fps: yeah, but that depends on where the bucket boundaries are, no?
<davidar>
fps: like I said, I'm not an ipfs dev, so not sure
<fps>
davidar: just brainstorming myself :)
<fps>
davidar: not an expert either
<davidar>
fps: but yeah, sounds like a reasonable idea, I'll make an issue to coordinate ideas about pubsub in a sec
nomoremoney has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<fps>
oh, i think in morphis the node id is again a hash of a key
<fps>
let's see
<fps>
so you'd have to do some work to mine for a key close to the document you're interested in, too
<fps>
i suppose that might also be the downfall of the scheme. with an appreciable network size the work required to mine for such a key would be prohibitive
<iHedgehog>
Trying to run ipfs on server in local network and connect from browser. I have changed corresponding ips in config. The I am tried to connect to webui from another machine in the network. Webui interface is loaded but no any info and I got 403 errors in browser console.
<RX14>
the webui doesn't work over the network, it is a static page which uses javascript to connect to the ipfs api on localhost
<RX14>
therefore it only works on the machine running ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<iHedgehog>
It would be cool if it worked.
<iHedgehog>
And I have error in terminal when I run ipfs on any machine
<iHedgehog>
ERRO[13:20:06:000] Path Resolve error: no link named "fontawesome-webfont.woff2" under QmWjympWW8hpP5Jgfu66ZrqTMykUBjGbFQF2XgfChau5NZ module=core/server
<RX14>
iHedgehog, that's an expected error
<RX14>
the woff2 file doesn't exist but it falls back to the woff file
<RX14>
and works correctly
<iHedgehog>
Ok, I just noticed.
<iHedgehog>
Is there some interesting places in network already? Maybe some list of sites/apps?
notduncansmith has quit [Read error: Connection reset by peer]
Quiark has joined #ipfs
<RX14>
Blame, would you randomise the gateways list for every request to spread the load... if gateway x is always at the top of the list it will be used the most.
<blame>
I've considered a random k of n approach. Right now it polls All of then
<RX14>
well, in terms of bandwidth mine would be at the top...
<RX14>
maybe have where they are based?
<blame>
Actually, mine is
<RX14>
so you can see which are in europe and which are in the US?
<RX14>
mine is gigabit in Paris
<blame>
Make a issue! Heck I'd love some help.
<atgnag>
Is there way to find a list of files I'm hosting?
<atgnag>
(that aren't pinned)
<blame>
We could make a map and a log and stuff
chriscool has joined #ipfs
<blame>
It really is meant to be a mvp to see if people want something better
<RX14>
hmmn, i don't do python
<RX14>
i can make issues though :P
<blame>
I'm about to be grocery shopping and not in a programming headspace. I'm trying to design a sukkah in my head for a friend
<blame>
I designed pyhp as a hack so I could avoid learning/dealing with php. It would be fairly easy to learn I expect
dignifiedquire has joined #ipfs
<voxelot>
mine is an OVH in france lol
<voxelot>
looks like you nginx is down too atm
<voxelot>
been awhile since i've touched py, let me see if i can randomize our gateways
<gaboose_>
has anyone done an app on ipfs that does 'aggregation'?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
dignifiedquire has quit [Quit: dignifiedquire]
<blame>
voxelot: name has changed to cachewarmer.blamestross.com
<atgnag>
How can I see what I've added?
<blame>
There is a list of cached files in the webui
<gr8>
what is the difference between IPFS and Gnunet? Both seem to be decentralized file-serving networks using content based adressing
<gr8>
for example, why don't you simply implement that IPFS thing using Gnunet infrastructure?
<gr8>
Gnunet aims to be extensible, afaik
chriscool has joined #ipfs
<lgierth>
gr8: we'd be happy for someone knowledgable to write down the differences and similarities between ipfs and gnunet: https://github.com/ipfs/faq/issues/2
<lgierth>
that'd be a first step for integration
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<substack>
I'm writing up a spec for doing content-addressed versioning inside of html <meta> and <link> tags. Does anyone know of any prior art? All I've found so far is http://tools.ietf.org/html/rfc5829
<substack>
this should work super well with ipfs and service workers :D
<blame>
How do folks feels about a federated/decentralized ipfs search engine rather than a distributed one?
_Vi has quit [Ping timeout: 268 seconds]
<spikebike>
gr8 are you familiar with gnunet?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<gr8>
spikebike: no
rschulman_ has joined #ipfs
<spikebike>
ah, I have some past experience, there are pretty large differences.
<spikebike>
gnunet wouldn't be a particularly good thing for IPFS to build on
<voxelot>
Blame: been thinking about a search engine as well, not even sure how to go about creating a decentralized one atm
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
<voxelot>
whyrusleeping's idea was to send out requests to the dht and hold files to satisify the bitswap as a way to do searching
atomotic has joined #ipfs
<spikebike>
voxelot: step one would be centralized search
<ipfsbot>
[go-ipfs] chriscool created generalize-peerid-check (+6 new commits): http://git.io/vZry8
<ipfsbot>
go-ipfs/generalize-peerid-check 5c118a2 Christian Couder: sharness: generalize test_check_peerid()...
<ipfsbot>
go-ipfs/generalize-peerid-check 0aa5936 Christian Couder: t0060: simplify peerid check...
<ipfsbot>
go-ipfs/generalize-peerid-check 9563cb2 Christian Couder: t0110: add peerid check...
<pjz>
. o O ( make it work, then make it fast )
<voxelot>
right, maybe all of us wiht servers can work together to reduce load atm
<pjz>
oh, hm, what if by convention there's the reverse of a robots.txt: a list of files we're hosting
<blame>
voxelot: id do it in a federated way, multiple servers act as "centralized" then let them exchange and merge databases via ipfs
<voxelot>
thought about that too, but could be a privacy issue
<spikebike>
pjz: heh, basically an index.html?
<voxelot>
yeah, let's figure some federated thing out Blame
<pjz>
spikebike: sure
<blame>
It could be an index.json
<spikebike>
voxelot: er, no privacy issue, just like HTML, if you don't want it published it's not in html
<blame>
But yeah
<voxelot>
yeah
<spikebike>
well you could take it to the next level
<pjz>
spikebike: but it should be different from the base-url because that's likely to be more fo an intro to your site kind of thing
simonv3 has quit [Quit: Connection closed for inactivity]
<chriscool>
morning whyrusleeping and everyone!
<whyrusleeping>
chriscool: heyo! good late afternoon (right?)
<chriscool>
whyrusleeping: yeah it's 9pm here
captainbland has joined #ipfs
rendar has quit [Ping timeout: 268 seconds]
voxelot has quit [Ping timeout: 256 seconds]
voxelot has joined #ipfs
<atgnag>
Why is the webui telling me I'm connected to 0 peers?
rendar has joined #ipfs
<atgnag>
Is the webui iffy for anyone else?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
joshbuddy has joined #ipfs
chriscool has quit [Ping timeout: 250 seconds]
joshbuddy has quit [Client Quit]
voxelot has quit [Ping timeout: 244 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
chriscool has joined #ipfs
<spikebike>
atgnag: what was ipfs swarm peers say?
<atgnag>
spikebike: Nevermind, I figured out you simply have to wait.
<spikebike>
heh
<atgnag>
It can only be as responsive as the network I guess.
mildred has joined #ipfs
<atgnag>
I kinda wish it would run run swarm peers periodically and cache it for the webui though.
ion has joined #ipfs
<ion>
Hi. IPFS looks awesome. Have you thought about specifying a standardized method of chunking files according to a rolling hash? That simple algorithm would provide rsync/zsync-like file updates.
<atgnag>
Hmm… dad doesn't seem to work.
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<richardlitt>
ion: If no one has, open an issue about it! Sounds cool.
<ion>
Thanks, I’ll look.
<richardlitt>
ion: np. :)
<atgnag>
Dag, not dad.
<richardlitt>
:D
<richardlitt>
If I make a PR in a random repository, who do I give it to for code review? Who is in charge of PRs for each repository? Is this list maintained anywhere?
<ion>
The owner(s) of the repository should receive an email about the pull request.
<richardlitt>
ion: Yeah, that works for personal ones - I'm specifically talking about who is in charge of ipfs/website, ipfs/community, ipfs/faq... and so on.
notduncansmith has quit [Read error: Connection reset by peer]
uhhyeahbret has quit [Ping timeout: 256 seconds]
<Igel>
afternoon.
rschulman_ has joined #ipfs
rschulman_ has quit [Ping timeout: 244 seconds]
<ipfsbot>
[node-ipfs] RichardLitt created feature/improve-readme (+1 new commit): http://git.io/vZoKd
<ipfsbot>
node-ipfs/feature/improve-readme 1fc1771 Richard Littauer: Added a Contribute section, with small rewrites...
<ipfsbot>
[node-ipfs] RichardLitt opened pull request #16: Added a Contribute section, with small rewrites (master...feature/improve-readme) http://git.io/vZoKN
domanic has joined #ipfs
mildred has quit [Ping timeout: 240 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ygrek_ has joined #ipfs
rschulman_ has joined #ipfs
mildred has joined #ipfs
<mafintosh>
jbenet, daviddias can i help implementing bitswap in js? i wanna use it for other things as well. is there any work done on this atm?
<ianopolous>
hi guys, quick question: with the http api, is the add command meant to also pin (that's what I'm finding)?
<richardlitt>
mafintosh: Might be good to open an issue here https://github.com/ipfs/node-ipfs and ping daviddias in case anyone else wants to help, too.
<richardlitt>
mafintosh: or I could open it now and just ping both of you?
notduncansmith has quit [Read error: Connection reset by peer]
<richardlitt>
who was it that needed help with setting DNS records on Digital Ocean programmatically? Was mentioned in one of the calls in the sprint on Monday, the infrastructure one I think.
mildred has joined #ipfs
mildred has quit [Ping timeout: 260 seconds]
VegBerg has quit [Quit: Leaving]
rschulman_ has quit [Quit: rschulman_]
reit has quit [Ping timeout: 264 seconds]
rschulman_ has joined #ipfs
reit has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Encrypt has quit [Quit: Sleeping time!]
amstocker has quit [Ping timeout: 244 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
dignifiedquire_ has joined #ipfs
higuita has quit [Ping timeout: 264 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
M-trashrabbit has joined #ipfs
voxelot has joined #ipfs
<lgierth>
richardlitt: i and daviddias
<richardlitt>
lgierth: what repo? A new one?
<richardlitt>
DO has an API with Oauth that you can use to set DNS records dynamically.
<lgierth>
richardlitt: yeah somethign along those lines would be great
<lgierth>
richardlitt: i added you to ipfs/infrastructure
<lgierth>
just make a new directory there or so. otherwise i can also make you a repo if that's more convenient
<richardlitt>
lgierth: I can make a repo, or just work on it in my own account. Will see what I can do.
<richardlitt>
Thanks.
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<richardlitt>
lgierth: Cool. So, there's a few good wrappers for the DO api. I'm not exactly sure where this code fits into ipfs/infrastructure, and I couldn't find a relevant open issue about it. I can make an issue there just stating this, if I'm not missing something?
amstocker has joined #ipfs
ReactorScram has quit [Ping timeout: 255 seconds]
ReactorScram has joined #ipfs
<rabbitface1>
n00b question: ipns requires the original node to be online in order to work, right? is there any plan to work around this eventually?
<lgierth>
richardlitt: oh hey i think it can be just a little standalone tool in whatever language you see fit. you give it a domain and a /ipfs/<hash> or /ipns/<hash> multiaddr, and it sets up the A and TXT record, or updates the TXT if it was already set up
<lgierth>
rabbitface1: yes totally! it's in the works
<lgierth>
/ipns/* will be first-class objects that can be pinned by other nodes, so that they keep them up-to-date automatically
<lgierth>
and keep them available while the originating node is unavailable
<lgierth>
richardlitt: i set up a new site just yesterday, let me see, i had a screenshot of the dns set up
<wasabiiii>
So, I really like this idea of using IPFS to deliver content... but have you guys thought much about how to deliver server interactivity? Like, say I want to post something to some database, and get a response back?
<rabbitface1>
lgierth: awesome. I was worried about ipns names as single points of failure
<lgierth>
the A record should ideally be 8 records, for the hosts in infrastructure.git/solarnet/hosts
<lgierth>
wasabiiii: not sure, generally you'd build your apps differently with ipfs. it's not really geared towards an application model where you have a central backend database
<lgierth>
you'll be able to post/put through the http gateway soon, though, if that's what you're asking
<richardlitt>
lgierth: So, this is for domains which you have credentials to manage on DO? Hmm. I may not understand DNS enough. Looking through DO dev docs now
atgnag has quit [Ping timeout: 264 seconds]
<richardlitt>
Ok. I can make this, I'm just not really sure why or how this is useful.
<richardlitt>
lgierth: still confused as to whose credentials I should use though. (Apologies; newbie as far as interacting with IPFS, although I've known about it for a while)
atgnag has joined #ipfs
<lgierth>
we're hosting the ipfs.io website through this (github/ipfs/website), but since ipns isn't complete yet, we need to point the TXT record to /ipfs/<hash> instead of /ipns/<hash>
<lgierth>
so we need to edit the TXT record everytime we make the smallest change to the website
tsenior has joined #ipfs
<lgierth>
it's ipfs add -r build/ipfs.io, then log into digitalocean, navigate to the ipfs.io domain, edit the record
<richardlitt>
lgierth: Got it. I'll make a dummy account on do to test this, makes sense. Thanks!
dignifiedquire_ has quit [Quit: dignifiedquire_]
<lgierth>
ideally there'd be a single make task in website.git, which builds the site, adds it, and calls your tool
<lgierth>
the next website we wanna host like this is dist.ipfs.io (the downloads page)
<lgierth>
and ipn.io could use a facelift and new domain :)
<lgierth>
and we want everybody to be able to easily host websites like this, be it with their gateway or ours
<richardlitt>
Is there a reason we don't use GitHub pages?
groxx has joined #ipfs
<richardlitt>
haha one thing at a time, I'm a volunteer still :)
<substack>
is it common to use an ipfs: protocol in URLs?
domanic has quit [Ping timeout: 244 seconds]
rendar has quit []
amstocker has quit [Ping timeout: 246 seconds]
<lgierth>
richardlitt: because we can! :)
<lgierth>
these websites are hosted right out of ipfs