<lgierth>
d10r: yes that's correct. also files are chunked to 256KB blocks to improve deduplication
<d10r>
ok, thanks. And is there currently a way to check how many replicas the network has of a file? At least approx.
<Mateon1>
d10r: `ipfs dht findprovs` returns a list of peers that own a certain hash. You might want to check all hashes under a file or directory, though, as some nodes might only have parts of the file.
bastianilso___ has quit [Quit: bastianilso___]
<d10r>
perfect, thx!
<d10r>
one more question: is there an easy way to compute the ipfs hash of a file in the local file system? The only way I currently know is to add it.
wallacoloo_ has quit [Remote host closed the connection]
d10r has quit [Ping timeout: 256 seconds]
ygrek has quit [Ping timeout: 260 seconds]
Kane` has quit [Remote host closed the connection]
hoboprimate has quit [Quit: hoboprimate]
haileris has joined #ipfs
rodarmor has joined #ipfs
<rodarmor>
I'm evaluating IPFS, and I'm trying to figure out how it compares to bittorrent.
<rodarmor>
Is there information anywhere on how IPFS peers exchange data? Does it use a tit-for-tat incentivizaiton strategy like bittorrent?
<rodarmor>
And is the P2P wire protocol documented anywhere?
<rodarmor>
Ahhh, I just found the section in the whitepaper on bitswap, looks like that wasn't there the last time I read it
f4 has joined #ipfs
shizy has quit [Quit: WeeChat 1.6]
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
<rodarmor>
Is it possible to crawl/search IPFS for a particular piece of data?
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
<haileris>
rodarmor: I don't believe that IPFS incentivizes yet. But I'm no expert.
<rodarmor>
haileris: How does my IPFS node retrieve files then? Doesn't there need to be some kind of incentivization for other nodes to send me blocks?
<rodarmor>
Or do they just altruistically send each other blocks, with incentives planned to come later?
sdiver has quit [Quit: WeeChat 1.6]
f4 has quit [Read error: Connection reset by peer]
<rodarmor>
Another random question: Can I configure IPFS to only talk to peers on a whitelist?
f4 has joined #ipfs
sdiver has joined #ipfs
geemili has quit [Ping timeout: 244 seconds]
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
chris613 has quit [Quit: Leaving.]
<haileris>
rodarmor: Your IPFS node just looks, and gets from (I think) what seems like the best places. You can see who has a hash with "ipfs dht findprovs HASH".
galois_dmz has joined #ipfs
<rodarmor>
haileris: Hmm, okay, interesting. So the bitswap stuff from the paper isn't implemented yet?
f4 has quit [Read error: Connection reset by peer]
<haileris>
rodarmor: Bitswap happens, but I don't know details. There's lots of bitswap. Maybe a couple GB per day in idle nodes.
<rodarmor>
haileris: Okay, cool, I'll keep digging. Thanks for your help ^__^
f4 has joined #ipfs
<haileris>
rodarmor: About whitelisting. I'm sure it's possible. You can "ipfs swarm connect "ADDRESS". So then just disable everything else, seednodes etc.
<rodarmor>
haileris: But perhaps the nodes that you connect to will gossip about other nodes, which you'll then connect to?
<haileris>
rodarmor: I haven't looked into that. One moment.
<rodarmor>
haileris: There's the swarm.AddrFilters config object, but I haven't looked into the details of that to see if it allows whitelisting multiple addresses
<haileris>
rodarmor: Yes, in "ipfs swarm filters" you can specify wildcards.
<haileris>
... and multiple filters.
<rodarmor>
haileris: Could you point me to somewhere (code is fine) I can read the format of those filters/what kinds of wildcards are allowed?
kulelu88 has quit [Quit: Leaving]
<haileris>
rodarmor: Not a coder. Commands are "ipfs swarm filters add <address>" and "ipfs swarm filters rm <address>.".
<rodarmor>
haileris: Ah, okay cool, then presumably I can do something like `filters add my-buddy` and `filters rm *` to get whitelist behavior
galois_dmz has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
<whyrusleeping>
rodarmor: they are just normal cidr's but put into a multiaddr format
<haileris>
rodarmor: Default config has ""AddrFilters": null" so you _add_ filters to block.
<rodarmor>
whyrusleeping: haileris: In that case it seems like it won't allow whitelisting, since if my buddies are all on different networks there is no filter which will block everyone else.
<whyrusleeping>
Yeah, there isnt whitelisting yet
<whyrusleeping>
we're working on having 'private networks'
<whyrusleeping>
which allows you to select who you connect to
<haileris>
whyrusleeping: Interested in an OnionCat update?
f4 has quit [Read error: Connection reset by peer]
galois_dmz has quit [Ping timeout: 260 seconds]
f4 has joined #ipfs
lkcl has quit [Ping timeout: 260 seconds]
lkcl has joined #ipfs
<rodarmor>
whyrusleeping: Is the bitswap credit stuff implemented yet?
<whyrusleeping>
haileris: sure, hows that all goin?
<whyrusleeping>
rodarmor: not really, we track it, but we don't limit any peers by it
<haileris>
@whyrusleeping: If I want hybrid clearnet/OnionCat node to bridge OnionCat-only nodes to clearnet, I can't run with "--routing=dhtclient".
<whyrusleeping>
huh
<haileris>
So then I'm interested in ways to manage throughput.
<whyrusleeping>
yeah... we havent gotten to implementing bandwidth limiters yet
<whyrusleeping>
its difficult to do without choking the system
<rodarmor>
whyrusleeping: Okay, interesting. I would really like to use IPFS for my application, but that's the one thing that makes me kind of nervous. BitTorrent has a bunch of flaws and ugly bits, but bittorrent swarms are unquestionably robust, even in the face of malicious participants, and IPFS might not be as robust yet, since the incentivization structure hasn't been worked out/tested
f4 has quit [Read error: Connection reset by peer]
<haileris>
@whyrusleeping: If that's a surprise for you, I can test more to verify that it wasn't a fluke.
f4 has joined #ipfs
<whyrusleeping>
haileris: do you mind documenting some of this in an issue somewhere? my brains not firing on all cylinders, spent all day debugging a networking issue with no progress.
<rodarmor>
whyrusleeping: Although, I that the mechanisms that make bittorrent work are well understood, and if the IPFS network attracted enough usage to make malicious peers a problem, the IPFS community would probably prioritize working out those issues to keep the network healthy.
<whyrusleeping>
rodarmor: correct
<whyrusleeping>
we havent put much effort into that side of things because "it just works" right now
<haileris>
@whyrusleeping: Yes, I'll test more, and then look for an appropriate issue to port to.
<haileris>
... port to.
<haileris>
...post to. Damn!
<rodarmor>
whyrusleeping: I think that's totally reasonable.
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
lkcl has quit [Ping timeout: 260 seconds]
Sebastian_CB has joined #ipfs
Sebastian_CB has quit [Ping timeout: 260 seconds]
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
Aranjedeath has quit [Quit: Three sheets to the wind]
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
gigq has quit [Ping timeout: 260 seconds]
gigq has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
haileris has quit [Quit: WeeChat 1.0.1]
haileris has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
haileris has quit [Client Quit]
haileris has joined #ipfs
mib_f6526ca has quit [Ping timeout: 260 seconds]
f4 has joined #ipfs
haileris has quit [Client Quit]
haileris has joined #ipfs
haileris has quit [Client Quit]
haileris has joined #ipfs
haileris has quit [Client Quit]
lkcl has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
wallacoloo_ has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
lkcl has quit [Ping timeout: 265 seconds]
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
lacour has quit [Quit: Leaving]
<keks>
whyrusleeping: i just looked at flushCopy again. I think this will always break iff the request has not bee read completely and the response is larger than 4096 bytes, i.e. one full buffer read
<whyrusleeping>
hrm...
<whyrusleeping>
thats unfortunate
<whyrusleeping>
is the problem that we're abusing something?
<whyrusleeping>
or is it a 'bug' in go's http code?
<keks>
that would mean we need to complettely buffer the request before responding, but apparently that's how http works in go
<keks>
i think they reuse the request body buffer for the the responsewriter
<keks>
so it's 0alloc by default
<keks>
oh it's midnight at your place. not gonne bother you :)
<keks>
typing in a crowded bus is tricky
f4 has quit [Read error: Connection reset by peer]
<whyrusleeping>
keks: hrm...
<whyrusleeping>
so adding that extra flush in 'fixes' it?
f4 has joined #ipfs
xelra has quit [Ping timeout: 250 seconds]
<keks>
what fixes it is that i removed it from the top
f4 has quit [Read error: Connection reset by peer]
<keks>
but this still isn't the right way to do it
<keks>
also i don't have an intuition about how the cmds.res behaves
xelra has joined #ipfs
<keks>
my idea was to move the flush...but at that point it should have already flushed...
f4 has joined #ipfs
<keks>
i guess we need to differentiate between input-heavy cmds (like add) and output-heavy cmds (like get or how it's called)
<keks>
so when we receive an add we perform the entire task before responding and flushing
<keks>
hmm no we should just process the whole request up front either way
<keks>
or do we have requests that have both a lot input and a lot of output?
<keks>
sorry for thinking out loud here
f4 has quit [Read error: Connection reset by peer]
f4 has joined #ipfs
<keks>
my idea was adding a large buffer either to buffer the request body or the response body, but we would need to choose which one we use based on the cmd at hand, eg buffer input for get and buffer output for add. but i'm not sure this is the best way to do it
<whyrusleeping>
hrm...
<whyrusleeping>
thats difficult too
<whyrusleeping>
do different http libs work differently?
<keks>
idk, but the only thing others can do is add buffers, and we can do that as well. just need to figure out the best way to do that
<keks>
maybe just make the ipfs daemon consume the whole request before starting to write
<keks>
that would be the best way, but i'll have to see how we can make that happen
ulrichard has joined #ipfs
f4 has quit [Read error: Connection reset by peer]
lkcl has joined #ipfs
cemerick has joined #ipfs
raitucarp has quit [Quit: Page closed]
bertschneider has joined #ipfs
<M-hash>
i'm only skim-reading these last few comments, but iirc there was a indeed bug in the go stdlib http bits about flushery that got fixed at some point in the last few point releases. you may want to make sure you're comparing go versions as you poke whatever-it-is.
robattila256 has quit [Ping timeout: 252 seconds]
wallacoloo_ has quit [Quit: wallacoloo_]
ivo_ has quit [Ping timeout: 260 seconds]
tmro has quit [Ping timeout: 260 seconds]
Qwertie has left #ipfs ["Leaving"]
tmro has joined #ipfs
ivo_ has joined #ipfs
ulrichard_ has joined #ipfs
ulrichard_ has quit [Read error: Connection reset by peer]
dignifiedquire has quit [Quit: Connection closed for inactivity]
<geoah>
(yes yes I can get it from ipfs but I'm using govendor instead of gx and breaks my ci :D)
Encrypt has quit [Quit: Quit]
cemerick has quit [Ping timeout: 265 seconds]
neuthral has quit [Quit: leaving]
herzmeister has joined #ipfs
arpu has quit [Ping timeout: 265 seconds]
Encrypt has joined #ipfs
arpu has joined #ipfs
Encrypt has quit [Quit: Quit]
Mizzu has joined #ipfs
soloojos has quit [Ping timeout: 245 seconds]
Sebastian_CB has quit [Quit: Leaving]
soloojos has joined #ipfs
ulrichard_ has joined #ipfs
hoboprimate has joined #ipfs
<A124>
mib_27dh283 What do you mean by 'assemble'?
<A124>
And what do you refer as not workable?
ulrichard_ has quit [Quit: Ex-Chat]
<mib_27dh283>
A124: I am experimenting with writing merkledag/unixfs on my own (outside of ipfs-go) and am experimenting with my own chunking alogorithms
<mib_27dh283>
I cut a file (in fact the latest tar of ipfs-go) into newline-delimited chunks to see how it will behave: it wasn't... great :)
<A124>
mib_27dh283 I'm interested in the chunking, can you point me at any repo you have?
<A124>
Well cutting anything on newline boundaries is not great. Also you might want to try master with the experimental raw leaves.
<A124>
*anything binary.
rgrinberg has joined #ipfs
<mib_27dh283>
the raw leaves are for ipld alone no?
<mib_27dh283>
A124: it's a bunch of bash and ruby, nothing runnable
<A124>
Not sure what you mean ipld alone. The new system to store files will have raw leaves for files also. (unless I have blackout)
<mib_27dh283>
at this point I am simply exploring what the system can do - finding its limits will get me a better idea what I am looking for as an optimum
<A124>
You might want to try crystal-lang. Ruby, while insanely expressive and joy to work with has performance problems.
<mib_27dh283>
A124: my understanding was that unixfs (the thing currently accessible via ipfs.io/ipfs/ is essentially frozen with no new dev )
<mib_27dh283>
am I mistaken?
<A124>
mib_27dh283: why(r)usleeping: just use master and pass the --raw-leaves flag to add
<A124>
That is, on master, on on release version.
<mib_27dh283>
A124: and will ipfs.io/ipfs understand that, or not at present?
<mib_27dh283>
( and if no - how long roughly before it does? )
rgrinberg has quit [Ping timeout: 250 seconds]
<A124>
Also as for chunking, you should try content dependent chunker if you want universal solution. As for ideal the problematic thing is standardization across community. The fixed block size is apparent. But if you have uncompressed or multi member tar files, you can slice them: header, file, padding, repeat.
<mib_27dh283>
A124: nod, I am just not that far along yet, that's step #2 :)
<mib_si2js31>
is there some example of how one would reference a git sha1 directly under this model?
<mib_si2js31>
( it doesn't have to be functional - pseudo-structure would do, I just can't figure out what the doc is trying to say )
Mizzu has joined #ipfs
hoboprimate_ has joined #ipfs
kulelu88 has joined #ipfs
galois_dmz has joined #ipfs
<dignifiedquire>
whyrusleeping: npm-on-ipfs now writes with 32 processes into the same ipfs daemon, this should give it some more stress :)
maxlath has joined #ipfs
chris613 has joined #ipfs
galois_dmz has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
<mib_si2js31>
I have to disconnect, but will read the log in the evening if someone drops a link about git-on-ipld
mib_si2js31 has quit [Quit: Page closed]
galois_dmz has quit [Ping timeout: 260 seconds]
maxlath has quit [Remote host closed the connection]
maxlath has joined #ipfs
geemili has joined #ipfs
maxlath has quit [Remote host closed the connection]
maxlath has joined #ipfs
hoboprimate_ is now known as hoboprimate
maxlath has quit [Remote host closed the connection]
maxlath has joined #ipfs
cemerick has joined #ipfs
lacour has joined #ipfs
BostonEnginerd has quit [Quit: WeeChat 1.5]
robattila256 has joined #ipfs
M-moyamo has joined #ipfs
M-moyamo has left #ipfs [#ipfs]
maxlath has quit [Remote host closed the connection]
maxlath has joined #ipfs
ianopolous_ has joined #ipfs
maxlath has quit [Ping timeout: 250 seconds]
ianopolous has quit [Ping timeout: 244 seconds]
taaem has quit [Ping timeout: 260 seconds]
Encrypt has joined #ipfs
herzmeister has quit [Quit: Leaving]
<A124>
What is zumwalts handle on irc?
<A124>
dignifiedquire ^
<dignifiedquire>
I think flyingzumwalt
<dignifiedquire>
but he doesn't seem to be online
<A124>
Yeah, wondered about that. Well, definitely there should be category "Help"
<A124>
Which would be a place to ask, so the other categories are not polluted with questions
ianopolous has joined #ipfs
<dignifiedquire>
yeah, I think he is in the middle of creating some topics if you refersh :D
<dignifiedquire>
nevermind that was from the system
ianopolous_ has quit [Ping timeout: 244 seconds]
shizy has joined #ipfs
bastianilso___ has quit [Quit: bastianilso___]
cemerick has quit [Ping timeout: 252 seconds]
kulelu88 has quit [Ping timeout: 252 seconds]
bertschneider has joined #ipfs
s[m] has joined #ipfs
shizy has quit [Ping timeout: 260 seconds]
Encrypt has quit [Quit: Quit]
DiCE1904 has quit [Ping timeout: 260 seconds]
<A124>
dignifiedquire Coding category is for fellow coders to help each other implement third party stuff.
<dignifiedquire>
A124: I see
<dignifiedquire>
A124: done
chriscool has quit [Ping timeout: 265 seconds]
geemili_ has joined #ipfs
geemili has quit [Ping timeout: 260 seconds]
<A124>
Great. Just you should likely move those two messages to Announcements
yoosi has quit [Remote host closed the connection]
yoosi has joined #ipfs
travis_ has joined #ipfs
otaci has joined #ipfs
<travis_>
Hello everyone, I have a total noob question that I expect one of you could answer pretty easily :)
<travis_>
I am struggling to understand the difference between js-ipfs and js-ipfs-api. I understand that js-ipfs-api is only a client, and requires connecting to a node
shakib[m] has joined #ipfs
<travis_>
But it seems like js-ipfs can be used standalone in the browser, without needing an additional running client
<shakib[m]>
how secure is ipfs?
<travis_>
When should I use js-ipfs-api vs just using js-ipfs alone? Any help would be appreciated
<shakib[m]>
ipfs is the best?
Aaron101- is now known as Aaron1011
Aaron1011 has quit [Changing host]
Aaron1011 has joined #ipfs
<A124>
shakib[m] It is in development, also depends what you mean by security.
<A124>
travis_ js-ipfs is js variant of go-ipfs
<shakib[m]>
people say it's not an anonymous network, is it true?
<A124>
Without the http serevr, etc, or differently said, IPFS node. js-ipfs-api is made for communication with go-ipfs node API.
<A124>
shakib[m] yes.
<shakib[m]>
what's the benefit of it?
<A124>
A lot, you should go https://ipfs.io and other talks/materials on youtube/internet. It is made for content distribution.
<A124>
Today if a lot of people want some files, it overloads servers. With ipfs, clients can be servers also.
<A124>
So it makes stuff faster in situations when it would else be really slow.
<A124>
Also ... data integrity.
<travis_>
Thanks A124
<shakib[m]>
it makes decentralized network more secure?
<travis_>
A124: so is it correct to say that I have the option to either a) use js-ipfs-api and expect my users to run go-ipfs locally or b) have them run a full server within the browser using js-ipfs?
<travis_>
But if that were true, then why would I ever use js-ipfs-api?
<A124>
shakib[m] It is decentralized network.
<A124>
And if enough people use it, it automaticaly becomes distributed.
<A124>
travis_ you can use public gateways without need for any javascript.
<A124>
Apart from that you are correct. and js-ipfs-api can be used on public gateways also (not for writing, but for reading)
<shakib[m]>
i meant ipfs is better than before or not?
<shakib[m]>
A124:
<A124>
What is before?
<shakib[m]>
without ipfs
<A124>
Definitely. But still less then halfway done.
DiCE1904 has joined #ipfs
<A124>
Currently works great for files, but future will be better.
Encrypt has joined #ipfs
<otaci>
does anyone know of an active orbit channel that I could take a look at?
<shakib[m]>
decentralize network isn't an anonymous network?
<ion>
Anonymizing networks are a whole different beast.
<shakib[m]>
then why people wants to use decentralized network?
<ion>
To avoid single points of failure
<ion>
Also possibly to avoid giving too much control over the network to a single entity.
<ion>
The first point was technical, the second one political
<ion>
with a lot of overlap
<shakib[m]>
u mean anonymizing is useless into decentralized network?
ryuuxsky has joined #ipfs
<cehteh>
anonymity needs more than just decentralization
<A124>
otaci if you tell me one I will join one to test.
<A124>
But of course there is #ipfs
<A124>
cehteh Welcome here, hehe.
<A124>
shakib[m] The aim is to solve networking problems, usability problems, etc. Anonymity is not usability problem.
<A124>
It is not made to anonymize you, a lot of people have this confusion.
<shakib[m]>
oh, ok
<A124>
And data bandwidth. Imagine loading some pages off your neighbor instead from US ;)
<cehteh>
A124: ;) .. i am lurking here since long time
<A124>
You feesh ;)
travis_ has quit [Ping timeout: 260 seconds]
<otaci>
A124 - i joined #ipfs, dont see anything
<A124>
I'll download and join.
* whyrusleeping
is pretty sure hes already on #ipfs
ryuuxsky has quit [Quit: Page closed]
bertschneider has quit [Quit: Bye bye]
<A124>
npm install with no output.
<shakib[m]>
goodbye friends. i'm leaving now
<A124>
See you around other time.
<A124>
haad npm install does something, then stops and see only blank line, have no idea
<ion>
shakib: A decentralized network should be able to take advantage of the fact that you and I may be in the very same room with one having some data the other wants. An anonymizing network by definition can not reveal that information to us.
<otaci>
well, thanks for trying A124
<A124>
(h)aad it runs just took few minutes with blank output.
cwahlers has quit [Ping timeout: 250 seconds]
cwahlers has joined #ipfs
<A124>
haad found UI glitch, pressing a tab makes enter not work, and unless person knows to frantically perss shift-tab and enter, he has to re-open
<A124>
Suggestion: make TAB submit username also
<A124>
otaci I'm there.
<whyrusleeping>
yay... found another bug in our stream muxer
* whyrusleeping
grumbles about wanting to just write his own damn stream muxer that works like he expects it to
<A124>
haad Another thing... pick random ports for the built in ipfs.
<A124>
And IPFS ... not realized as I ran orbit... killed the internet again... insert meme: That's great.
gmcquillan__ has joined #ipfs
gmcquillan__ is now known as gmcquillan
<A124>
whyrusleeping Sorry to bother you with this. Is thee any progress on the router killer things. It is very unpleasant if you are not in control of the router.
<whyrusleeping>
thats kinda ish what i'm working on right now
<whyrusleeping>
my current 'project' is connection closing
<whyrusleeping>
so we don't keep hundreds of connections open at a time
<A124>
Oh, right, good. Well as this connection was open when it did 'it', irc stays open at least, but web is dead for me now.
<whyrusleeping>
overflows your routers tables?
<A124>
No idea, probably something like that.
<A124>
The open connections keep working though. On similar situations on other routers it knocks down completely.
<A124>
On some places we run 2000-2005 old routers, metal boxes, overheating, with transformer PSUs. Why? Static memory and electonics that does not die. This is not the case.
<A124>
As the twitter handle says, "Internet of S**t"
herzmeister has joined #ipfs
pyrazole has joined #ipfs
dlight has joined #ipfs
Mizzu has quit [Quit: WeeChat 1.6]
<dignifiedquire>
A124: Internet of S**t is hilarious
mildred has joined #ipfs
<dignifiedquire>
and sad but true
wallacoloo_ has joined #ipfs
<frood>
router failure may be bufferbloat related. common problem with this sort of thing.
Encrypt has quit [Quit: Quit]
b100d has joined #ipfs
A124 has quit [Quit: '']
<whyrusleeping>
no food in the fridge == looks like soylent for dinner again
<achin>
rule 1: always keep bread and cheese on hand
<Akaibu>
whyrusleeping: is soylent even that good tasting?
<whyrusleeping>
achin: i ate all the bread and cheese
<whyrusleeping>
Akaibu: soylents really not that bad
<achin>
amature move my friend
<whyrusleeping>
it tastes like milk that you previously had cereal in
<whyrusleeping>
achin: i don't have a car. and its been rainy so i couldnt make it to the grocery store
<achin>
:(
<whyrusleeping>
Yeah... i might be starting to regret thinking "man, i'll save so much money by riding a motorcycle everywhere"
<Akaibu>
And do you even need to shit when you (drink/eat?) it long term whyrusleeping?
<whyrusleeping>
Akaibu: yeah, everythings pretty normal
<whyrusleeping>
i don't think i've gone more than a day without eating real food though
<whyrusleeping>
i normally break down and order some thai food
<achin>
yum
<whyrusleeping>
yeah...
<Akaibu>
Heh, having a cheat day on your soylent diet
<whyrusleeping>
is it really a diet if its just being lazy?
<Akaibu>
A diet is just what you eat
<Akaibu>
Not actually a plan or anything
<Akaibu>
It's a common misconception
<Akaibu>
Just like how we landed on the moon
<Akaibu>
It never actually happened
<achin>
i still have a few bags of the powered stuff
<achin>
but i rarely eat them
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
gmcquillan has quit [Quit: gmcquillan]
vtomole has joined #ipfs
<whyrusleeping>
yeah, the powdered stuff kinda sucks
bastianilso___ has joined #ipfs
<whyrusleeping>
i recommend the pre-bottled mix stuff
<achin>
what's the shelf life of the pre-mixed stuff?
<achin>
i do kinda dig that the powered stuff will last like a year
<achin>
(in dry form)
<whyrusleeping>
achin: about a year
<whyrusleeping>
the 'best by' date on the bottle i'm drinking is august 11 2017
<achin>
ah, cool
<achin>
i didn't realize it was so long
<whyrusleeping>
Yeah
<whyrusleeping>
we have quite a large supply of it in our pantry
maxlath has joined #ipfs
maxlath has quit [Client Quit]
dlight has quit [Ping timeout: 260 seconds]
<dignifiedquire>
whyrusleeping: can't you get sth delivered?
<whyrusleeping>
i can
<whyrusleeping>
but that costs money, and i spent all my money on graphics cards
<whyrusleeping>
<.<
<whyrusleeping>
dignifiedquire: also, i'm changing the api for ipfs swarm peers
<whyrusleeping>
you should update it in your side too
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
<dignifiedquire>
why you change that?
<dignifiedquire>
is there an issue I can read?
<whyrusleeping>
because the previous format was just a list of strings
<whyrusleeping>
and for latency, it just appended latency as a string to the address