<dryajov>
lgierth: it seems like multiaddrs have to parse into tuples, so an address like this would not be valid - /p2p-circuit/p2p/QmTwo
<dryajov>
I’m looking at the multiaddr implementation and thats what it tries to do every time
<dryajov>
has to be an even number of parts
<dryajov>
whyrusleeping: ^^^
<whyrusleeping>
hrm...
<whyrusleeping>
dryajov: are you using go or js ?
<dryajov>
js
<whyrusleeping>
theres your problem
<dryajov>
lol
<whyrusleeping>
the go code doesnt require tuples
<dryajov>
well… thats a problem
<whyrusleeping>
/ip4/1.2.3.4/udp/5001/utp/ws would work
<dryajov>
yeah...
<whyrusleeping>
Yeah, dignifiedquire daviddias[m]
<daviddias>
yeaah
<daviddias>
what?
<whyrusleeping>
your multiaddrs is broken
<daviddias>
dryajov: not true
<dryajov>
:P
<daviddias>
libp2p-webrtc-star is odd
<daviddias>
same for websockets
<daviddias>
it parses in tupples, but some addrs have 0 length fields
<daviddias>
making it a half tupple
<dryajov>
lol, ok… let me take a look at those
<dryajov>
half tupple… like that
<dryajov>
;)
zabirauf has quit [Remote host closed the connection]
zabirauf has joined #ipfs
zabirauf has quit [Ping timeout: 256 seconds]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
<dryajov>
diasdavid: so, p2p-circuit is a [290, V, 'p2p-circuit’] (note the V), which I’m guessing its a variable size addr… in which case it breaks on odd parts
<dryajov>
if I change it to a 0, it works
<dryajov>
I initially had it as [290, 0, 'p2p-circuit’], but lgierth: made an observations that it should be a V...
<daviddias>
It should be 0
<daviddias>
why V?
<dryajov>
I haven’t dugg deep enough, but wondering why is it a V?
<dryajov>
as per the spec… no idea
<dryajov>
let me dig up the issue… I made the change today…
gmcabrita has quit [Quit: Connection closed for inactivity]
galois_dmz has joined #ipfs
galois_d_ has quit [Ping timeout: 246 seconds]
wallacoloo_____ has joined #ipfs
wallacoloo_____ has quit [Client Quit]
wallacoloo_____ has joined #ipfs
<Akaibu>
He seems really into the idea of ipfs to reduce bandwidth
antiantonym has quit [Ping timeout: 260 seconds]
bwerthmann has joined #ipfs
bwerthmann has quit [Ping timeout: 260 seconds]
structuralist has joined #ipfs
SuprDewd has joined #ipfs
WindPower has quit [Quit: ~]
galois_d_ has joined #ipfs
galois_dmz has quit [Ping timeout: 240 seconds]
Akaibu has quit []
structuralist has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
robattila256 has joined #ipfs
galois_d_ has quit [Ping timeout: 246 seconds]
Aranjedeath has quit [Ping timeout: 246 seconds]
galois_dmz has quit [Ping timeout: 246 seconds]
onabreak has quit [Ping timeout: 260 seconds]
Jesin has joined #ipfs
<whyrusleeping>
Akaibu, that link doesnt work for me
bwerthmann has joined #ipfs
galois_dmz has joined #ipfs
bwerthmann has quit [Ping timeout: 240 seconds]
WindPower has joined #ipfs
infinity0 has joined #ipfs
galois_dmz has quit [Read error: Connection reset by peer]
galois_dmz has joined #ipfs
_whitelogger has joined #ipfs
galois_dmz has quit [Ping timeout: 264 seconds]
gpestana has joined #ipfs
Oatmeal has joined #ipfs
bwerthmann has joined #ipfs
john2 has joined #ipfs
bwerthmann has quit [Ping timeout: 264 seconds]
chris613 has joined #ipfs
tmg has joined #ipfs
mg_ has quit [Ping timeout: 264 seconds]
mg_ has joined #ipfs
stoopkid has quit [Quit: Connection closed for inactivity]
bwerthmann has joined #ipfs
bwerthmann has quit [Ping timeout: 268 seconds]
gpestana has quit [Quit: Page closed]
gpestana has joined #ipfs
Caterpillar has joined #ipfs
rendar has joined #ipfs
koalalorenzo has joined #ipfs
koalalorenzo has joined #ipfs
koalalorenzo has quit [Changing host]
caiogondim has quit [Quit: caiogondim]
bwerthmann has joined #ipfs
mildred has joined #ipfs
mildred4 has quit [Ping timeout: 264 seconds]
bwerthmann has quit [Ping timeout: 240 seconds]
koalalorenzo has quit [Quit: This computer has gone to sleep]
koalalorenzo has joined #ipfs
koalalorenzo has joined #ipfs
koalalorenzo has quit [Changing host]
espadrine has joined #ipfs
maxlath has joined #ipfs
bwerthmann has joined #ipfs
monkhood_ has joined #ipfs
monkhood has quit [Ping timeout: 240 seconds]
bwerthmann has quit [Ping timeout: 268 seconds]
koalalorenzo has quit [Quit: This computer has gone to sleep]
<whyrusleeping>
PSA: DHT traffic might go up a little for 0.4.7 (It went down in 0.4.5), We fixed an issue that was causing dht queries to end early, resulting in values not being put out to enough peers
<whyrusleeping>
The upside of this is that ipns should be more reliable
ianopolous has joined #ipfs
kobajagi has joined #ipfs
kobajagi has quit [Remote host closed the connection]
koalalorenzo has joined #ipfs
koalalorenzo has quit [Client Quit]
eater has quit [Ping timeout: 260 seconds]
koalalorenzo has joined #ipfs
ShalokShalom has quit [Remote host closed the connection]
eater has joined #ipfs
Mizzu has joined #ipfs
ianopolous has quit [Remote host closed the connection]
mildred has quit [Ping timeout: 240 seconds]
koalalorenzo has quit [Quit: Sto andando via]
<reit>
playing with 0.4.7 rc1, currently i am getting different hashes for `ipfs add [FILE]` and `ipfs add --nocopy [FILE]` even though it's the same file
galois_dmz has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
galois_dmz has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
galois_dmz has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
galois_dmz has quit [Remote host closed the connection]
palkeo has joined #ipfs
palkeo has joined #ipfs
palkeo has quit [Changing host]
_whitelogger has joined #ipfs
ianopolous has quit [Read error: Connection reset by peer]
ianopolous has joined #ipfs
palkeo has quit [Ping timeout: 258 seconds]
bwerthmann has joined #ipfs
bwerthmann has quit [Ping timeout: 260 seconds]
jkilpatr has joined #ipfs
<cblgh>
cc victorbjelkholm ^
<cblgh>
i like the screenshots in the readme, way too many repos just don't have any
ianopolous has quit [Ping timeout: 246 seconds]
<cblgh>
nice, simple api too
Encrypt has joined #ipfs
atrapado_ has joined #ipfs
bwerthmann has joined #ipfs
galois_dmz has joined #ipfs
tmg has quit [Ping timeout: 260 seconds]
bwerthmann has quit [Ping timeout: 268 seconds]
galois_dmz has quit [Remote host closed the connection]
robattila256 has quit [Quit: WeeChat 1.7]
skeuomorf has joined #ipfs
robattila256 has joined #ipfs
espadrine has quit [Ping timeout: 240 seconds]
onabreak has joined #ipfs
bwerthmann has joined #ipfs
bwerthmann has quit [Ping timeout: 240 seconds]
<ShalokShalom>
how can WebAssembly help IPFS?
<Mateon1>
ShalokShalom: It could provide a smaller (and faster) "binary" for the browser version, currently it's ~2MB
<ShalokShalom>
kk
<ShalokShalom>
i mean, can it replace the JS rewrite?
<ShalokShalom>
and run the go version directly?
<Mateon1>
Not Go, as Go doesn't compile to LLVM
<apiarian>
pinkieval: that looks similar to my pinbase project (planning to have multiple pinbases mirroring part or all of each-others pins, based on user requirements)
<apiarian>
yup, looks pretty similar, though my focus is more on helping get my non-ipfs-node-running friends and family to use IPFS. a step toward the distributed end through personal-trust-based decentralization
<apiarian>
(also supporting a possible future search system based on the aliases and other metadata that individual parties could link to their pins)
bwerthmann has joined #ipfs
<apiarian>
(mostly solving that main problem of IPFS for me: i don't particularly want to maintain a note on my desktop with hash-to-filename links)
<Ronsor>
that forms an irc network anyone can link to
<Ronsor>
the big part is
<Ronsor>
linking uses IPFS pubsub
bwerthmann has joined #ipfs
caiogondim has joined #ipfs
mg_ has quit [Ping timeout: 260 seconds]
robattila256 has joined #ipfs
<Ronsor>
fun fact: I coded half of the IRCd's ipfs-backend on a phone; that was painful
Drew[m] has joined #ipfs
bwerthmann has quit [Ping timeout: 260 seconds]
<apiarian>
Ronsor: you do do that? ssh into a machine, local editor?
<Ronsor>
ssh
<Ronsor>
into my home server
<apiarian>
emacs?
<Ronsor>
where i do my dev work
* apiarian
has tried using vim through ssh, but found the mode switches super awkward on a screen keyboard
dhorn[m] has joined #ipfs
dhorn[m] has left #ipfs [#ipfs]
<Ronsor>
i use nano
<Ronsor>
only need one command: ^X-Y (save and exit)
<apiarian>
aha
<apiarian>
well that's pretty impressive that you're able to work like that. trying to work on a small screen probably requires you to keep a lot of the code in your head. and the docs and references.
<Ronsor>
i keep lots of code in my head
<Ronsor>
and docs
<Ronsor>
and stuff
<Ronsor>
i never stop thinking about my code
<Ronsor>
except sleep
<apiarian>
does nano have any support for things like code completion or jump to def/docs ?
<Ronsor>
nope
chriscool has joined #ipfs
<Ronsor>
it just edits text
<Ronsor>
it does have syntax highlighting though
<apiarian>
nice. i've stopped using syntax highlighting for my golang work, but still find it helpful for nutty things like perl (my main language at work)
<apiarian>
so, this is an irc server which can talk to other irc servers to set up a decentralized network which coordinates over ipfs?
<Ronsor>
yeah
<apiarian>
how do irc servers normally work?
<apiarian>
manual configs?
<Ronsor>
normally you have to configure your server to link with another server
cemerick has joined #ipfs
<Ronsor>
now on startup they just automagically join the network of servers via ipfs
<apiarian>
do you need to do anything to keep bad actors out?
arkimedes has joined #ipfs
<Ronsor>
yes
* apiarian
has only a vague understanding of irc servers, having messed with some bits of client-side irc in the past
<Ronsor>
i'm hoping i just ignore them
<Ronsor>
but you do understand the issue
<Ronsor>
though ngircd is by default configured to disallow remote die/squit/connect/etc
<Ronsor>
so that helps prevent lots of abuse
<apiarian>
what's to stop me from claiming that my haxor-irc-server.net is part of your network
<Ronsor>
you're supposed to
<Ronsor>
anyone can join
<Ronsor>
that is the point
<apiarian>
so... what's to stop my rogue server from intercepting and altering all of the messages from the unsuspecting users connecting to it?
palkeo has quit [Ping timeout: 260 seconds]
<Ronsor>
well
<Ronsor>
you could do that
<Ronsor>
we won't stop you
<Ronsor>
we only care if you affect users on other servers
<Ronsor>
users foolishly choose to connect to your server; that's their fault
<Ronsor>
your server won't get automatically added a round-robin or something
espadrine has quit [Ping timeout: 240 seconds]
<apiarian>
aha, so this is a personal trust sort of model?
<apiarian>
like, my users trust me and explicitly connect to my server in order to communicate with the rest of the network?
<Ronsor>
yeah
<apiarian>
nice
mg_ has joined #ipfs
jkilpatr has quit [Ping timeout: 246 seconds]
bwerthmann has joined #ipfs
mildred has quit [Ping timeout: 246 seconds]
mg_ has quit [Ping timeout: 240 seconds]
Encrypt has quit [Quit: Quit]
mg_ has joined #ipfs
zabirauf has quit [Remote host closed the connection]
maxlath has quit [Ping timeout: 260 seconds]
zabirauf has joined #ipfs
zabirauf has quit [Remote host closed the connection]
zabirauf has joined #ipfs
mildred has joined #ipfs
Akaibu has joined #ipfs
chungy has quit [Ping timeout: 256 seconds]
arkimedes has quit [Ping timeout: 260 seconds]
structuralist has joined #ipfs
<Akaibu>
whyrusleeping: Akaibu, that link doesnt work for me - sorry yea I deleted it after let me reupload it
<Akaibu>
This might be a good reason to implement DHT since this would be a major test of ipfs's ability to work on a large scale
tilgovi has joined #ipfs
maxlath has joined #ipfs
bwn has quit [Ping timeout: 246 seconds]
<Akaibu>
whyrusleeping: one of the guys I told this to that is aware of ipfs said this " Tell them that we need a way to whitelist or at least blacklist hashes from our gateways, and that dht is the killer feature of ipfs.js "
Aranjedeath has joined #ipfs
_desu_ has joined #ipfs
<antonizoon>
So basically. We are currently using nginx directives to blacklist hashes
<antonizoon>
But this is obviously imperfect and does not scale easily
cemerick has quit [Ping timeout: 246 seconds]
<antonizoon>
It would be better to have a whitelist where we direct our servers own ipfs gateway to serve only files that we have ourselves
<antonizoon>
And this whitelist should of course be optionally enabled, not default behavior
<lgierth>
antonizoon: yes i'd love to have a mode of the gateway where it doesn't go and fetch stuff for you, but only uses what it already has locally
<lgierth>
if you wanna take a stab at it i'll help out with review and advice
<antonizoon>
Yeah. Though in our case it could also be a whitelist where you allow other file hashes that you don't have pinned as well
<antonizoon>
E.g. They are hosted on my other storage server
<lgierth>
ok yeah i suppose that'd work too
<antonizoon>
Basically I think it would be best to have some sort of hash lookup table
<antonizoon>
That checks at three levels:
<antonizoon>
1. Whether you have the file locally
chungy has joined #ipfs
<antonizoon>
2. Whether it matches hashes from a hash table, which could be periodically centrally synced with other nodes you own
<Akaibu>
A DHT basically
<antonizoon>
3. Whether it is a hash that is explicitly blacklisted
<antonizoon>
Well a separate issue is that ipfs.js would be able to bring the true benefits of ipfs to the browser
<antonizoon>
But it lacks DHT support at the moment
<antonizoon>
A big Achilles heel
structuralist has quit [Remote host closed the connection]
<antonizoon>
otherwise gateways for ipfs has to be hosted to make your own cdn, or the user must download ipfs, which doesn't easily happen until it reaches critical mass
structuralist has joined #ipfs
<antonizoon>
So ipfs.js would open the true capabilities to a larger audience
<antonizoon>
That's my thoughts. What do you think
chungy has quit [Ping timeout: 260 seconds]
<frood>
does ipfs.js have a solution to CORS/same-origin policy in secured pages?
bwn has joined #ipfs
<frood>
or rather, in pages delivered over https?
<Akaibu>
well there are ways to bypass that tbh
<frood>
sure, like using WebSockets or WebRTC. but there are significant costs involved
<frood>
last I checked Go doesn't have a solid WebRTC implementation either. so that's a rough sell for IPFS.
DiCE1904 has quit [Ping timeout: 246 seconds]
chungy has joined #ipfs
espadrine has joined #ipfs
<Akaibu>
Maybe just a chrome extension?
<Akaibu>
They can bypass same origin policy I'm pretty sure
<Akaibu>
You just need allow permission I think
<Akaibu>
Don't know about Firefox
<Akaibu>
Also are we working on a webassembly version of ipfs?
realisation has joined #ipfs
<frood>
extensions can almost do whatever they want. but that's not a great solution.
<horrified>
4chan on ipfs, i bet that's gonna be fun
zabirauf has quit [Remote host closed the connection]
zabirauf has joined #ipfs
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
zabirauf_ has joined #ipfs
zabirauf has quit [Ping timeout: 246 seconds]
tilgovi has quit [Ping timeout: 246 seconds]
<Akaibu>
horrified: quite fun, but we should take it serious, this is the first major stress test of the ability of ipfs to work at a large scale, neocity is like a Emo goth kid's poetry tumblr page compare to 4chan who has up to 600,000 people a day compared to neocity 20000 a month
Encrypt has joined #ipfs
<Akaibu>
I'm guesstimating but on the actual figures but my point still stands
zabirauf_ has quit [Ping timeout: 264 seconds]
<horrified>
4chan just doesn't seem like the type of content worthy of being put on a "permanent web"
<alu>
honestly im split on it
<alu>
i think the internet archive is a better effort tho
<frood>
I wouldn't worry about it. ipfs is permanent in the same way bittorrent is.
<frood>
availability is a function of popularity.
<alu>
thats true, its also cost effecient
<Akaibu>
horrified: nazis thought that books made by Jewish authors weren't good enough to be kept
<alu>
fam plz, thats not the same lol
<Akaibu>
It kinda is
<Akaibu>
We don't make desisions on what should and should not be included for the future
<antonizoon>
My position is that ipfs is still hosted by servers
<Akaibu>
We should not make desisions*
<antonizoon>
It's not really permanent until someone else decides to download it
<Akaibu>
^
<antonizoon>
This is why we also would need whitelists in the case of serving this sort of site
<antonizoon>
Since we often have to delete and blacklist certain hashes
<antonizoon>
And I would aim for a more decentralized subscription based blacklist rather than a central one
<antonizoon>
Basically provided by a site provider and having legitimacy as part of its normal moderation
<antonizoon>
Just like the internet today, any system where humans exist will have the good bad and ugly
<antonizoon>
But if you aren't seeking out and downloading the bad and the ugly you are basically making a vote against its continued existence. That's how content should be handled.
<frood>
you aren't really making a vote against it, just declining to make a vote for it
<frood>
making a vote against it would be like, declining to respond to FIND VALUE messages
<frood>
or dropping nodes on the provider list from your routing table.
<antonizoon>
I guess that's where a hash blacklist as suggested prior comes in
realisation has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<antonizoon>
One that you subscribe to per website and is maintained by them as part of normal moderation
<frood>
the adblock filterlist model?
<antonizoon>
Sure
<antonizoon>
E.g. one imageboard publishes a blacklist that applies to the site and any hashes they found. Users of the imageboard would naturally use this blacklist. Or if they disagree with the moderation, they can unsubscribe from it (at their own risk)
<antonizoon>
Other imageboards can choose to subscribe to that first imageboard blacklist with addendums from themselves, or maintain a separate one
<antonizoon>
If ipfs.js were enabled the site would maybe notify with "hash blacklist loaded (click here to disable... At your own risk)"
<antonizoon>
And the servers own ipfs gateways should have a corresponding server side blacklist that prevents such content from being served from our end
<antonizoon>
If the user decides to seek it out anyway with their own ipfs node that is their business
<antonizoon>
If the content even exists by then
<Akaibu>
antonizoon: each user could in turn have there own blacklist for say blocking out intrusive memes such as naked banana and spooderman
<Akaibu>
Like Adblock as was mentioned before
<antonizoon>
Sure, set your own blocks
pled has joined #ipfs
<Akaibu>
And host the blacklist on ipfs that way anyone would would want to also block the memes could just sub to that blacklist, works on a consumer level and a provider level hahah
<Akaibu>
(Provider in this case would be the 4chan mod team)
<antonizoon>
Well we must figure out how blacklists and whitelists would be implemented in ipfs in practice
<antonizoon>
It looks like this is the current approach
<antonizoon>
One blacklist is already given by ipfs.io itself
<antonizoon>
How do they implement it
mildred has quit [Ping timeout: 258 seconds]
<antonizoon>
Well it seems to be more of a proof of concept
mildred has joined #ipfs
pled has left #ipfs ["sic - 250 LOC are too much!"]
<antonizoon>
Another thing is that blacklists should have severity levels, which are decided by the list maintainer's mod team. That way users can decide to disable some blacklists and keep others in place.
<antonizoon>
The lowest could be DMCA/copyright, since it might not be illegal worldwide and could work like how Google displays ChillingEffects
realisation has joined #ipfs
<antonizoon>
Next could be NSFW, then NSFL (non-illegal), then illegal
<antonizoon>
So if you were looking for smut you might turn off DMCA and NSFW blacklists
<antonizoon>
But would still never turn off NSFL or illegal. Unless you really wanted to audit the moderation.
<antonizoon>
And yeah another point all blacklists should be a decentralized, normal moderation tool. They can be shared and collated, but generally we want the scope to remain in one website to avoid central authorities deciding what is blocked.
monkhood has joined #ipfs
john2 has quit [Ping timeout: 240 seconds]
monkhood_ has quit [Ping timeout: 240 seconds]
Intensity has joined #ipfs
<ShalokShalom>
Some people here work on login capability for IPFS