ploop has quit [Remote host closed the connection]
gmoro has quit [Ping timeout: 268 seconds]
ploop has joined #ipfs
asyncsec has joined #ipfs
gmoro has joined #ipfs
null_radix has joined #ipfs
stevenaleach has joined #ipfs
<whyrusleeping>
sure thing
<kevina>
thanks
caiogondim has quit [Quit: caiogondim]
Akaibu has quit []
gmoro has quit [Ping timeout: 260 seconds]
gmoro has joined #ipfs
MDude has joined #ipfs
ckwaldon has quit [Remote host closed the connection]
ckwaldon has joined #ipfs
galois_dmz has quit [Remote host closed the connection]
spilotro has quit [Ping timeout: 260 seconds]
dimitarvp has quit [Read error: Connection reset by peer]
metasyn[m] has joined #ipfs
spilotro has joined #ipfs
wmoh has joined #ipfs
m_anish has quit [Ping timeout: 255 seconds]
wmoh has left #ipfs [#ipfs]
btmsn has joined #ipfs
mildred1 has quit [Ping timeout: 260 seconds]
mildred1 has joined #ipfs
_whitelogger has joined #ipfs
leeola has quit [Quit: Connection closed for inactivity]
m_anish has joined #ipfs
_whitelogger has joined #ipfs
ofdm has quit [Ping timeout: 260 seconds]
wkennington has joined #ipfs
ofdm has joined #ipfs
Aranjedeath has quit [Quit: Three sheets to the wind]
palkeo has joined #ipfs
palkeo has quit [Quit: Konversation terminated!]
akkad_ has joined #ipfs
Oatmeal has joined #ipfs
ilackarms has quit [Ping timeout: 260 seconds]
asyncsec has quit [Quit: asyncsec]
Oatmeal has quit [Ping timeout: 240 seconds]
<vivus>
Hello all. Will it be okay to ask blockchain questions here?
wkennington has quit [Read error: Connection reset by peer]
screensaver has joined #ipfs
sirdancealot has joined #ipfs
rendar has joined #ipfs
Fab-Mat-rix is now known as fabrixxm
mildred2 has joined #ipfs
Caterpillar has joined #ipfs
fnlintred887 has quit [Ping timeout: 258 seconds]
tom3 has joined #ipfs
s_kunk has joined #ipfs
s_kunk has joined #ipfs
s_kunk has quit [Changing host]
espadrine has joined #ipfs
cxl000 has joined #ipfs
maxlath has joined #ipfs
aeternity has joined #ipfs
Guest218473[m] has joined #ipfs
rcat has joined #ipfs
dgrisham has quit [Ping timeout: 255 seconds]
dgrisham has joined #ipfs
robattila256 has quit [Ping timeout: 260 seconds]
nunofmn has joined #ipfs
fnlintred887 has joined #ipfs
dimitarvp has joined #ipfs
jkilpatr has quit [Ping timeout: 260 seconds]
maxlath has quit [Ping timeout: 268 seconds]
gmcabrita has joined #ipfs
Guest45287 has quit [Ping timeout: 255 seconds]
aeternity has quit [Ping timeout: 240 seconds]
jkilpatr has joined #ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Boomerang has joined #ipfs
maxlath has joined #ipfs
nannal_ has quit [Ping timeout: 252 seconds]
nannal has joined #ipfs
Intensity has quit [Ping timeout: 258 seconds]
OstlerDev has quit [Ping timeout: 260 seconds]
<timthelion[m]>
Is there any reliable way of sending someone a link to an ipfs resource which would then be loaded by their ipfs daemon? I'm thinking of something like http://localhost:8080/ipfs/kjlkjklj but not depended on which port the other person uses. The reason I want this, is that I want to use client side AES decryption, without actually having to trust a public gateway.
<timthelion[m]>
This should be possible, because the local ipfs daemon would verify that the hash was correct. So the client side decryption code would not be modified.
<timthelion[m]>
However, the localhost link doesn't work if the user selects a different port than 8080.
<timthelion[m]>
And that seems highly likely to hapen, given how popular 8080 is.
OstlerDev has joined #ipfs
OstlerDev has joined #ipfs
OstlerDev has quit [Changing host]
Intensity has joined #ipfs
nunofmn has joined #ipfs
Boomerang has quit [Remote host closed the connection]
citizenErased has joined #ipfs
<achin>
how about just giving them an https://ipfs.io/ipfs/QmHash url, and letting the person manually load that hash in their own gateway if they have one?
Foxcool has joined #ipfs
tom3 has quit [Quit: WeeChat 1.7.1]
<vivus>
Hello all. Will it be okay to ask blockchain questions here?
khoarau[m]1 has joined #ipfs
<achin>
go ahead and ask. if it's not appropriate, someone will tell you :)
leeola has joined #ipfs
<timthelion[m]>
achin: Well, that means that the person has to trust the gateway. I guess I could lead to a landing page that says: "Do not enter your password here unless you're sharing cat photos. For real security use your own gateway." And Then a little link to localhost and the hash for more savy users.
N80[m] has quit [Ping timeout: 245 seconds]
gmanrwp[m] has quit [Ping timeout: 245 seconds]
Foxcool has quit [Ping timeout: 252 seconds]
<achin>
you'd have to trust the gateway (and the application) in both cases
<timthelion[m]>
well the hash is trusted... Perhaps sent via gpg encrypted mail
aeternity has joined #ipfs
<achin>
you could just send the hash and let the user dedide for themselves how best to view it
<timthelion[m]>
but you own local gateway should be trusted to verify that the hash is correct.
<achin>
i suppose if someone is willing to trust the ipfs code base, they would also be willing to trust https://ipfs.io/, but maybe that's not true
<timthelion[m]>
not at all true. I've read some of the code. But I have no idea what is on the server...
<timthelion[m]>
while I cannot read all of the code, if a lot of people are reading it and trying to understand it, it only takes one of us to find a flaw/backdoor.
<achin>
is the concern that ipfs.io might be logging your access? sending you malicious data? something else? all of the above?
<timthelion[m]>
There are 121 contributors to the repo, which is a lot of parts of the code that have been understood by a lot of people
aeternity has quit [Ping timeout: 258 seconds]
<timthelion[m]>
achin, that it could send you a website that was different than the code that was supposed to bellong to the hash.
<timthelion[m]>
That the hash of the contents could be different
<achin>
ok
<lidel>
timthelion[m], if you have mentioned browser add-on, it will detect public urls that start with https://ipfs.ip/ipfs/ and replace request with one to localhost. it hever hits ipfs.io, but is much easier to share
<timthelion[m]>
Because the trusted code would prompt for the aes password and decrypt it, but a comprimised version of the code could send the password to the server or something.
<lidel>
*never
<achin>
indeed that would be possible and hard to detect in a browser
<timthelion[m]>
I want the UX to be as close to "hit this link to the private story in your email client" as possible.
kvda has joined #ipfs
<achin>
but your specifically targetting users who are running their own IPFS daemon?
<timthelion[m]>
well I'd prefer not to, but I've yet to figure out any way to get around that requirement.
<timthelion[m]>
I thought of making a bookmarklet like link, but that doesn't seem to work due to anti-xxs
<lidel>
browser add-on will eventually ship with js-ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<achin>
could you run your own public gateway, and send URLs that use your own gateway?
Foxcool has joined #ipfs
<timthelion[m]>
I could. But I'm probably even less trustworthy to third parties than ipfs.io
N80[m] has joined #ipfs
<achin>
i thought you were the one sending out the ipfs hashes?
gmanrwp[m] has joined #ipfs
<timthelion[m]>
well, I want it to be possible for anyone to do it ;)
<timthelion[m]>
does the firefox plugin re-write only ipfs.io links or also, for example, ipfs.hobbs.cz links?
aeternity has joined #ipfs
jrwr has quit [Changing host]
jrwr has joined #ipfs
emunand[m] has joined #ipfs
<Atrus[m]>
It'll do ipfs.io and /ipfs/hash links. You can list the ipfs.hobbs.cz as a gateway, but that's on a per plug in basis
<timthelion[m]>
aha
tom1 has joined #ipfs
<timthelion[m]>
They think that matching */ipfs/<valid-hash> is too broad?
aeternity has quit [Ping timeout: 252 seconds]
<achin>
maybe it's too expensive
asyncsec has joined #ipfs
onabreak has quit [Ping timeout: 260 seconds]
<Scio[m]>
it also matches domains with TXT records pointing to ipfs
ilackarms_ has joined #ipfs
maxlath1 has joined #ipfs
maxlath has quit [Ping timeout: 260 seconds]
maxlath1 is now known as maxlath
caiogondim has joined #ipfs
sirdancealot has quit [Ping timeout: 240 seconds]
ashark has joined #ipfs
<vivus>
when storing data on IPFS, is this data immutable?
<achin>
yes
<vivus>
so once I store something on IPFS, it is there forever?
<emunand[m]>
if someone else hosts it, yes
<vivus>
in order to 'update' a piece of data, the old data is stored in the old hash and the new data in a new hash?
<achin>
yep
<emunand[m]>
yes, unless you use ipns, then it redirects you to the new hash
<achin>
the contents of
<achin>
the contents of "QmT78zSuBmuS4z925WZfrqQ1qHaJ56DQaTfyMUF7F8ff5o" will *always* be "hello world"
<achin>
but there might not always be someone hosting this hash
<vivus>
is there a difference to the way data is stored? eg. I want to store my public PGP key on IPFS, is it (by default) public-viewable via a query?
<demize>
If you've added it, and you have a node connected to the rest of the IPFS network, then anyone that has the hash can get it, yes.
<vivus>
so how does 1 store private data then? do I need to encrypt the data locally and then store it? or is it a bad idea to store private data
<achin>
you should encrypt it first, yes
<vivus>
so essentially, I would primarily use IPFS for public-data
<achin>
that would be my recommendation, at least for right now
jungly has joined #ipfs
<vivus>
so IPFS is written in GO?
<achin>
the reference implementation is, yes. but there's also an implementation written in javascript
<vivus>
how do you guys handle BFT? are you using the same structure as bitcoin with Proof of Work?
shizy has joined #ipfs
<achin>
BFT?
<vivus>
Byzantine Fault Tolerance
<r0kk3rz>
theres no need for BFT
<vivus>
why not?
<r0kk3rz>
its not a blockchain
<vivus>
so how is data integrity guaranteed?
<r0kk3rz>
its hash based
<emunand[m]>
but ipfs can be based on one
<vivus>
I don't quite understand how it's not a blockchain if it's hash-based. Isn't bitcoin based on SHA256 hashes?
<r0kk3rz>
hashes have many uses, its true
<r0kk3rz>
ipfs is a lot closer to bittorrent than bitcoin
<r0kk3rz>
both use hashes
jleon has joined #ipfs
<vivus>
is there an infographic that shows how it works?
<achin>
git is also a useful comparison
<achin>
(also being merkledag)
<SchrodingersScat>
why can't i hold all these merkles
<achin>
(in fact, distributed git on top of ipfs will be Really Cool)
<Scio[m]>
vivus: Hashes in IPFS are used to address content. If you have a hash of something in IPFS (no matter how big) you'll get that data directly (or split into a merkle-tree if it's too big.) While that content can be a blockchain, IPFS doesn't concern itself with any ordering of sorts and by extension doesn't require any consensus or any of the hard problems associated with it.
Boomerang has joined #ipfs
wmoh has joined #ipfs
<Scio[m]>
vivus: This recent short talk (short by Juan's standards :) covers whys and hows IPFS and some of the associated technologies: https://www.youtube.com/watch?v=Bqs_LzBjQyk
wmoh has left #ipfs [#ipfs]
kvda has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<achin>
the paper is a useful read, too
<vivus>
so IPFS splits my data up into small pieces using hashes (like how bittorrent does) ?
<achin>
yes
<Scio[m]>
(the paper is also surprisingly non-scary!)
<vivus>
okay, let me first watch that link and come back with questions (if any)
tsetair__ has quit [Read error: Connection reset by peer]
<Scio[m]>
(Also, unlike bittorrent, because ipfs is all about the merkle dag, you don't have to throw away an entire swarm because you just changed a small bit of it, large parts of the unchanged bits get reused)