Cryto! 8 June 2013

00:00:30 <pzuraq> but unless I get a decent job, or donations start rolling in, I'm not going to be able to host it forever on my dime
00:00:53 <joepie91> you can run this off a $5 VPS for the foreseeable future.
00:01:38 <joepie91> does your job pay so bad that you can't pay a $5 or maybe $10 VPS...?
00:01:47 <joepie91> in which case you should probably not bother wasting your time with said job :P
00:02:40 <pzuraq> my job pays terribly, I'm a student!
00:02:46 <pzuraq> going into debt D:
00:02:51 <pzuraq> also, it's not just a listing site
00:02:54 <joepie91> so terrible that you don't have a disposable $5-$10?
00:02:59 <joepie91> well, then what is it? :P
00:03:09 <pzuraq> we are going to be doing some pretty intensive analysis to find what we call Cycles
00:03:23 <pzuraq> which are when we have a multi-person trade event
00:03:44 <joepie91> okay?
00:05:01 <pzuraq> hold on a sec
00:05:04 <pzuraq> need to find a graphic
00:05:32 <pzuraq> ah
00:05:33 <pzuraq> http://betternature.files.wordpress.com/2012/05/f10-2_barter3.png?w=500
00:05:37 <pzuraq> essentially, this scenario
00:05:50 <pzuraq> our current algorithm grows exponentially
00:05:57 <pzuraq> I'm pretty sure it's NP-complete
00:06:54 <pzuraq> er
00:06:59 <pzuraq> sorry, meant the opposite
00:07:03 <pzuraq> it's non-polynomial
00:07:11 <pzuraq> terms get me confused..
00:08:34 <pzuraq> basically, as our user base grows so will our need for better processing equipment. How much better? Not sure yet.
00:08:48 <anonO_o> wow, redeclipse makes my heard palpitate
00:08:51 <anonO_o> heart
00:10:53 <AnonForecast> lol
00:13:57 <joepie91> anonO_o: in a good or a bad way?
00:14:22 <joepie91> pzuraq:
00:14:23 <joepie91> > DIck
00:14:25 <joepie91> Dick *
00:14:48 <joepie91> anyway
00:14:53 <joepie91> pzuraq
00:14:54 <anonO_o> makes it race and hard to breathe.  I was playing locally to test
00:15:03 <anonO_o> if that's AI, it was totally kicking my ass.
00:15:03 <joepie91> processing equipment isn't actually expensive
00:15:15 <joepie91> anonO_o: did it have a 'bot' sign next to it?
00:15:24 <joepie91> the word 'bot' in small font, surrounded by stripes, a sun-like pattern
00:15:28 <anonO_o> didn't look, too busy dying
00:15:31 <joepie91> ah
00:15:35 <joepie91> if it does, it's a bot
00:15:38 <joepie91> if not, it's not
00:15:40 <anonO_o> it was local, right?
00:15:42 <joepie91> oh
00:15:45 <joepie91> right, then it's a bot
00:15:50 <joepie91> AI is pretty meh in Red Eclipse though
00:15:57 <anonO_o> I was gangbanged by bots
00:15:59 <joepie91> I really don't like the bots - they have an unrealistic playstyle
00:16:21 <joepie91> they are incredibly stupid about some things, and in other cases they spot you in a corner when no human would have seen you
00:16:24 <joepie91> very frustrating :/
00:16:43 <pzuraq> yes joepie91?
00:16:49 <joepie91> <joepie91>processing equipment isn't actually expensive
00:16:53 <pzuraq> ah
00:17:00 <joepie91> I'm sure that if you hang around a few hosting IRC channels
00:17:05 <joepie91> you can pick up a few Xeon boards for cheap
00:17:14 <joepie91> colocate them in the cheapest datacenter possible
00:17:21 <joepie91> because hey, it's only off-site analysis anyway
00:17:27 <pzuraq> well, like I said I only want to cover costs and such. I'm trying to keep VC money out of it as much as possible myself, my partners are less keen on that idea.
00:17:28 <joepie91> so who cares about uptime / speed / etc
00:18:11 <pzuraq> I figured you would like the idea of bypassing the monetary system for trade tho ;P
00:18:34 <joepie91> pzuraq: the thing is that you haven't thought about these things
00:18:46 <joepie91> you assume that need processing power == need a bunch of monies to pay for processing power
00:18:59 <joepie91> and meh, trade system really isn't that much better than monetary system
00:20:11 anonO_o has quit (Input/output error)
00:24:26 <pzuraq> you are such an idealist joepie91
00:24:35 <pzuraq> it's why I stick around, tho ;P
00:26:33 <joepie91> actually, this has very little to do with being an idealist
00:26:46 <joepie91> and a lot to do with being realistic, and _thinking_ about things rather than dismissing ideas beforehand
00:27:07 <joepie91> the "ah, that's not going to work anyway, not even going to think about that" thing that a lot of people do, and that seems to apply here as well
00:27:23 <joepie91> completely separate from what I believe is good/necessary/whatever, you just didn't think this through :P
00:27:29 <joepie91> not enough at least
00:28:28 <pzuraq> I wouldn't say that. It's just that you look at things in the future and see the trends, and you know what it's going to become and you're absolutely right.
00:28:49 <joepie91> ?
00:28:57 * joepie91 is unsure how to interpret that sentence
00:30:03 <pzuraq> but, for the moment at least, we are in a transitionary period where the combination of social/technological/political forces in the world aren't there yet. Not that they couldn't be, at least technologically speaking.
00:30:36 <joepie91> eh.
00:30:40 <pzuraq> what I mean is there are going to be some number of steps between the world of today and the world you envision. Maybe it's one, gigantic revolutionary leap. Maybe it's many small steps.
00:30:43 <joepie91> you can use a $5 VPS for the frontend
00:30:54 <joepie91> you can pick up cheap Xeon boards from hosts for processing
00:31:02 <joepie91> I am unsure what else you think needs to be accomplished before the above is a possibility?
00:31:08 <joepie91> unless the topic changed and I didn't notice?
00:31:08 <pzuraq> and, what if the site grows to several million users?
00:31:16 <joepie91> when do you expect that to happen?
00:31:20 <joepie91> overnight?
00:31:32 <pzuraq> I don't really, I'm pretty sure it's going to flop
00:31:47 <joepie91> let's assume that it does not flop
00:31:48 <pzuraq> but it's always a good idea to have a plan, just in case
00:32:02 <pzuraq> maybe 2-3 years, maybe longer.
00:32:03 <joepie91> it's always a good idea to have a plan for realistic scenarios, and just a bit beyond *
00:32:15 <joepie91> this way, you may as well start planning for all of the universe including aliens to use your site
00:32:17 <joepie91> but that wouldn't make sense either
00:32:29 <joepie91> the chance of that actually happening fast enough to take you by surprise, is so small that it's negligible
00:32:52 <pzuraq> that I can agree with
00:32:52 <joepie91> in other words: when you start getting users flooding in, figure out a way to do things
00:33:04 <joepie91> graph out your signup rates
00:33:13 <joepie91> if you see a continued sharp rise, you'll have to start thinking about more hardware
00:33:18 <joepie91> or more hosting, rather
00:33:24 <joepie91> and really, the only relevant thing is the frontend
00:33:27 <pzuraq> indeed
00:33:31 <joepie91> the analysis is a nice feature, but it's not urgent
00:33:40 <joepie91> it doesn't matter whether you do it right now, or whether you have a backlog of a month
00:33:44 <joepie91> that's something that can be solved later
00:34:03 <joepie91> in other words, you grab a few cheapass OVH boxes as backend servers for the user-facing part
00:34:15 <joepie91> load-balance with haproxy on some cheapo VPS with a lot of bandwidth on a reliable host
00:34:35 <joepie91> and you have a duct taped solution that will hold up against your rise in userbase long enough to think of a better plan, for under 50 euro a month
00:35:47 <joepie91> to put things into perspective
00:35:57 <joepie91> anonnews handled 300k unique visitors in one day on a 512MB VPS
00:36:02 <joepie91> and back then it didn't even use memcache
00:36:18 <joepie91> I think you're going to be fine with a $5-10 VPS for the foreseeable future
00:36:51 <pzuraq> sure, but does it have to account for users actually generating content? Working with the database?
00:36:55 <pzuraq> Also, it's PHP based no?
00:37:10 <joepie91> define 'working with the database'
00:37:17 <joepie91> what does your site have to do beyond inserting and retrieving records?
00:37:21 <joepie91> and yes, it is PHP based
00:37:49 <pzuraq> on every insert it has to run that algo to find cycles
00:37:53 <pzuraq> so there's that
00:38:04 <joepie91> also, anonnews currently has a forum, I have frequent spikes of 200-300k uniques a day with heavy forum usage including a lot of spambots (not counted in those uniques)
00:38:07 <joepie91> using memcache now
00:38:13 <joepie91> resource usage is hardly noticeable
00:38:17 <pzuraq> cool
00:38:20 <joepie91> hardly a blip on a graph
00:38:27 <joepie91> no, it doesn't have to run that algo on every insert
00:38:34 <joepie91> if it does, then your system isn't engineered well enough
00:38:46 <joepie91> for starters, why aren't you caching stuff?
00:39:04 <joepie91> and beyond that, why are you even updating things live? from what I understand so far, I see no need for realtime feedback from said algo
00:39:04 <pzuraq> every insert could potentially have created a new cycle, so we need to check for new cycles
00:39:06 <joepie91> so why not queue it?
00:39:16 <joepie91> potentially, yes
00:39:17 <joepie91> is it relevant?
00:39:19 <joepie91> at that point?
00:39:29 <joepie91> at that immediate point after insertion?
00:40:13 <pzuraq> then or shortly after, yes. It's first come first serve, and any items which are part of a cycle need to be removed from the pool so that another cycle can't interfere with the original one
00:40:23 <pzuraq> an item cannot belong to two cycles
00:40:37 <joepie91> you'll probably have to elaborate on the concept of said cycles more
00:40:44 <joepie91> because at this point I have no idea what you mean
00:40:58 <joepie91> try describing it from a user point of view
00:41:41 <pzuraq> so, users can make offers on items
00:42:01 <pzuraq> which is better described as the item "wanting" another items, in terms of our db
00:43:08 <pzuraq> if item A wants item B, item B wants item C, and item C wants item A then we can create a cycle which allows us to trade the items without an intermediary (cash)
00:43:44 <pzuraq> user A gets item B, user B gets item C, and user C gets item A, and they all give there respective items to who they are supposed to
00:43:58 <pzuraq> users don't actually have to organize these cycles, they just make the offers
00:44:29 <pzuraq> a typical trade between two people of two items would be a two item cycle, a special case
00:46:36 <joepie91> right
00:46:42 <joepie91> I still don't see how this warrants real-time processing
00:47:07 <joepie91> instead of retrying every single cycle on an insert, create a list of 'wanted' items to complete a cycle every time you re-analyze
00:47:19 <joepie91> and queue a re-analyze job after an insertion to happen _eventually_
00:47:36 <joepie91> that should already cut processing power requirements and thus job execution time
00:47:43 <joepie91> second point, seeing as it's first come first serve anyway
00:47:49 <joepie91> does it really matter whether a trade is executed now or in 2 days
00:47:54 <joepie91> as long as they are executed in order?
00:48:17 <joepie91> also, re: that wanted items list
00:48:23 <joepie91> every time a new item is inserted, it's checked against the list
00:48:28 <joepie91> no calculations, just a simple match
00:48:42 <joepie91> if it's in there, queue a process job with high priority
00:48:58 <joepie91> the chance of a needs-two-items trade being completed faster than a needs-one-item trade is miniscule
00:49:28 <joepie91> thus it's safe to assume that a needs-one-item trade was placed earlier than any needs-two-items trade offer that might exist during a later-executed queued job
00:49:34 <joepie91> when executing the needs-one-item trade immediately
00:49:51 <joepie91> there you go, need for realtime heavy processing obliterated
00:51:40 * joepie91 pokes pzuraq in case he didn't see response yet
00:53:19 <pzuraq> that... doesn't really make sense to me
00:53:51 <pzuraq> It does matter if a trade is executed now or 2 days from now, we need to inform the users that they have a trade immediately so they can actually send the items
00:54:10 <joepie91> no, you don't 'need'
00:54:12 <joepie91> you 'want
00:54:13 <joepie91> 'want' *
00:54:15 <pzuraq> it's not about when an *item* is inserted
00:54:23 <pzuraq> no, I'm going to go with need
00:54:27 <joepie91> then you're wrong
00:54:33 <pzuraq> under the basis that I would not use the site if it took that long
00:54:35 <joepie91> there's a very big difference between "this has to absolutely be done to ensure the site works"
00:54:41 <joepie91> and "this has to be done to make users happy"
00:54:57 <pzuraq> making users happy is part of my definition of it working
00:55:14 <joepie91> if you consider those the same thing, then yes, you're going to have problems with this
00:55:28 <joepie91> 'working' is purely a technical matter here
00:55:37 <joepie91> is the site reachable, do buttons work, etc
00:56:05 <joepie91> the whole scenario that I just sketched with a trade being executed _eventually_ is something that will only ever happen when your infrastructure is overloaded
00:56:21 <joepie91> with the setup you have in mind right now, the moment your site is overloaded, everything crashes and burns
00:56:26 <joepie91> and the entire site stops working
00:56:27 <joepie91> literally
00:56:42 <joepie91> and it's going to cost copious amounts of money to prevent that from happening
00:56:58 <joepie91> with the solution(s) I just gave above, when your stuff gets overloaded, it will as-graciously-as-possible degrade service
00:57:07 <joepie91> avoiding impact on critical parts of the infrastructure as best as it can
00:57:37 <joepie91> the site keeps working, trades keep getting executed, no discrepancies in the internal bookkeeping, no inexplicably disappearing trades or delays
00:57:59 <joepie91> when things get delayed because it's overloaded, that is measurable, and the only impact it has, is that the trade execution speed lowers
00:58:03 <joepie91> that's it./
00:58:06 <pzuraq> ok, fair enough. It's not that I'm not planning to have it degrade gracefully, and yes if it gets overloaded it will have to do that.
00:58:50 <joepie91> and honestly, "sorry guys our systems are overloaded due to unexpected interest, we'll have it back up and running in 3 days" is a whole lot better than "whoops, website vanished from the net" or "whoops, site is gone because owners are bankrupt because of hosting costs"
00:59:27 <pzuraq> Fair enough, but I wouldn't want to have to say 3 days. Very few communities survive that kind of down time.
00:59:42 <joepie91> and at the point where you get hit with a traffic spike that you don't expect to last, you can always spin up a few temporary VPSes at some pay-per-hour hosting thing
00:59:45 <joepie91> to offload computing power
01:00:01 <pzuraq> Yeah, I kind of want to use Heroku if possible.
01:00:02 <joepie91> if you expect it to be a continued increase in users, buy an old server and colo it somewhere cheap
01:00:12 <joepie91> heroku is expensive and not terribly efficient from what I've read
01:00:22 <pzuraq> scales nicely though
01:00:24 <joepie91> afaik Joyent would be cheaper, for example
01:00:25 <pzuraq> super easy to setup too :)
01:00:33 <joepie91> also, 3 days is nothing
01:00:38 <joepie91> because yes
01:00:43 <pzuraq> depends on the community
01:00:45 <joepie91> very few communities survive that kind of downtime
01:00:50 <joepie91> the thing is
01:00:53 <joepie91> you don't _have_ downtime
01:01:02 <joepie91> that's why I sketched that whole setup above
01:01:14 <joepie91> the only thing you have is 3 days of backlog and delayed trade executions
01:01:17 <joepie91> most of which won't even be noticeable
01:01:32 <joepie91> because people generally don't see that "hey, there's a possible trade, but the system hasn't picked it up yet"
01:01:42 <joepie91> as far as they are concerned, the final item for the three-way-trade was just added 2 days later
01:01:45 <pzuraq> I would notice... because I clicked "trade" and I still haven't gotten my item...
01:01:45 <joepie91> than when it really was
01:01:52 <joepie91> you'll have queued trades anyway
01:02:12 <joepie91> queued-because-of-trading-delay and queued-because-of-missing-item are indistinguishable
01:02:15 <pzuraq> are, forgot to mention, we have a view called "on the table" where users can essentially "buy it now"
01:02:17 <joepie91> to an end user
01:02:23 <pzuraq> you click here, you get that item now
01:02:27 <joepie91> sure, that's fine
01:02:49 <joepie91> that doesn't even have to interact with the backend db
01:02:50 <joepie91> er
01:02:53 <joepie91> with the backend processing system *
01:03:03 <joepie91> trade now is purely a database operation
01:04:16 <joepie91> look, things _are_ going to break at some point
01:04:19 <pzuraq> hmm, good point, we could definitely cut down there
01:04:20 <pzuraq> yes
01:04:21 <joepie91> that's unavoidable
01:04:25 <joepie91> so, design for it
01:04:42 <joepie91> design your things to have minimal (visible) impact to the end user
01:04:44 <joepie91> when something breaks
01:05:27 <pzuraq> we plan to
01:05:57 *** Xeross is now known as Xeross|AFK
01:08:27 <joepie91> also
01:08:29 <joepie91> I am going to brb
01:08:30 <joepie91> potentially sleep
01:08:31 <joepie91> :P
01:08:53 <pzuraq> gnight, sleep well
01:14:42 speakeasy (speakeasy@cryto-D25FDC50.range86-152.btcentralplus.com) has joined #crytocc
01:15:30 <twitchyliquid64> joepie91: OMGGGG
01:15:41 <twitchyliquid64> WHY DID NOT ONE TELL ME ABOUT TWITTER BOOTSTRAP
01:15:58 <twitchyliquid64> everyone knows I HATE website styling and am terrible at it
01:16:46 <twitchyliquid64> now I don't have to! horay for the Apache license!
01:16:57 <twitchyliquid64> <3
01:16:58 <twitchyliquid64> :D
01:17:00 <twitchyliquid64> :DDDD
01:17:25 <pzuraq> also check out Foundation by Zurb
01:17:55 speakeasy has quit (Ping timeout)
01:27:43 <waflel> hi
01:28:05 kvasir (kvasir@cryto-9B6B837F.hsd1.ca.comcast.net) has joined #crytocc
01:43:07 <iceTwy> Django Unchained was an entertaining film to watch
01:43:15 <iceTwy> so Tarantino-esque
01:43:42 <kvasir> very very classic tarantino :D
01:44:10 <iceTwy> yup, loved it. I also liked his appearence at the end of the film
01:44:20 <iceTwy> He decided to have himself blown up lol
01:45:46 <iceTwy> Going to sleep now
01:45:48 <iceTwy> Night!
01:46:01 iceTwy has quit (User quit:  Quit)
01:54:36 zest has quit (User quit:  hf)
02:11:38 HiveResearch has quit (Ping timeout)
02:57:04 LastOneStanding has quit (User quit:  <censored> you guys, I'm going home.)
03:33:23 pzuraq has quit (Input/output error)
03:37:22 kvasir has quit (User quit:  <censored> <censored> my computer turned itself off again)
04:13:31 pzuraq (pzuraq@cryto-3B273C29.ucsc.edu) has joined #crytocc
04:14:02 pzuraq has quit (Input/output error)
05:05:04 ttmbRAT (ttmbRAT@cryto-54383BCE.us-west-1.compute.amazonaws.com) has joined #crytocc
05:06:27 ttmbRAT has quit (User quit:  Connection closed)
05:17:49 HiveResearch (HiveResear@developers.developers.developers) has joined #crytocc
05:21:38 Angelina (chatzilla@Angelina.users.cryto) has joined #crytocc
05:30:02 kvasir (kvasir@cryto-9B6B837F.hsd1.ca.comcast.net) has joined #crytocc
05:43:28 HiveResearch has quit (Ping timeout)
06:11:39 HiveResearch (HiveResear@developers.developers.developers) has joined #crytocc
06:41:29 kvasir has quit (User quit:  <censored> <censored> my computer turned itself off again)
07:19:07 pzuraq (pzuraq@cryto-D09C67DC.resnet.ucsc.edu) has joined #crytocc
08:02:18 <ShadowDemon> I just had the hottest pepper I've had in quite some time
08:02:29 <ShadowDemon> Some dried asian pepper
08:02:42 <AnonForecast> orly?
08:03:13 <ShadowDemon> My mouth tasted metallic and my stomach is burning from the inside right now
08:03:38 <ShadowDemon> Serves me right for eating asian food in a dark room
08:05:00 speakeasy (speakeasy@speakeasy.users.cryto) has joined #crytocc
08:06:25 <ShadowDemon> Yay
08:06:30 <ShadowDemon> Ice cream killed it
08:07:49 pzuraq has quit (Input/output error)
08:08:08 <ShadowDemon> joepie91: btw, i do agree with the octal php issue
08:08:41 <ShadowDemon> it's quite stupid
08:23:13 pzuraq (pzuraq@cryto-3B273C29.ucsc.edu) has joined #crytocc
08:44:30 HiveResearch has quit (Ping timeout)
08:53:22 monod (~pmpf@monod.users.cryto) has joined #crytocc
08:53:46 <monod> good evening/morning
08:53:53 <Angelina> GReets
08:56:55 <monod> ooooh
08:56:59 <monod> Angelina
08:57:02 <monod> welcome back
08:57:21 <monod> please, remind me what you were into? cryptography?
08:59:08 <Angelina> [off[ Somewhat
08:59:35 <Angelina> Hahah
09:02:03 HiveResearch (HiveResear@developers.developers.developers) has joined #crytocc
09:03:14 speakeasy has quit (User quit:  Nettalk6 - www.ntalk.de)
09:06:16 <MK_FG> It is in murrica!
09:08:29 <ShadowDemon> Meow
09:11:26 mama has quit (Ping timeout)
09:11:43 <ShadowDemon> Much more tasty as well
09:17:23 <monod> guys
09:17:28 <monod> how could it be true?
09:17:59 <monod> pi written to the (200*10^6)-th digit
09:18:05 <monod> 200m digits
09:18:14 <monod> is ~190MB
09:18:25 <monod> if you compress it with gzip
09:18:32 <monod> IT GETS TO 90MB o_O
09:19:04 <monod> this is absolutely strange, so before thinking there's low entropy in the first 200m digits of pi, I have to do some more researches
09:19:19 <monod> but that's what I immediately thought
09:19:28 <monod> but how can pi have low entropy??
09:20:01 <monod> if it would be true, then we could predict further digits
09:20:06 <monod> afk a minute
09:20:08 <monod> or two
09:20:11 <monod> or three/four
09:23:05 <MK_FG> Heh
09:23:21 <MK_FG> Trick is that you don't write that 190M as binary, I guess
09:23:32 <MK_FG> You write 190M of ASCII digit-characters
09:24:03 <MK_FG> Which vary only one-two bits in each consecutive byte, I think
09:24:21 <MK_FG> So I'm actually surprised that compression there is not 10x
09:26:14 <monod> but it's 5x
09:26:23 <monod> anyway, hmm I think it may right
09:26:36 <monod> how can I convert such a file to a binary file?
09:30:12 <MK_FG> I don't know how any huge-mantissa formats past just double-precision float go, but given that you probably don't need much interoperability there anyway...
09:31:10 <MK_FG> ...I'd just accumulate digits till they overflow 256, then write that byte out big-endian style and do acc=acc%256, rinse, repeat
09:35:11 <monod> I didn't get it: read the file, digit by digit, until the 256th digit?
09:42:53 <MK_FG> No, but I think it's a bit trickier than I thought above anyway
09:43:07 <monod> hmm okay
09:43:41 <MK_FG> Have counter, read each digit, multiply by 10*(counter++)...
09:44:12 <MK_FG> ...add that to accumulator, eventually writing-out the part that gets left behind
09:44:41 <MK_FG> (as new numbers only add up to one end of that potentially-infinite bitstream)
09:45:28 <MK_FG> There's gotta be some Knuth on how to do that
09:48:23 <MK_FG> You want the Knuth? YOU CAN'T HANDLE THE KNUTH!
09:48:37 <MK_FG> (c) someone on twitter
09:48:45 <monod> so, multiplying by 10.. what is it supposed to do? Oh lawl
09:48:57 <monod> lol for the joke
09:48:59 <monod> anyway
09:49:00 <MK_FG> Integers you read are base-10
09:49:04 <monod> ype
09:49:07 <monod> yep*
09:49:46 <MK_FG> So you multiply by 10 for every next one to get it's value in the number
09:49:50 pzuraq has quit (Input/output error)
09:50:05 <monod> so it's 10's powers then?
09:50:13 <monod> instead of multiplying
09:50:15 <MK_FG> Though you should mul on 10^(counter++), not 10*...
09:50:20 <monod> ok
09:50:27 <monod> and also
09:51:05 <monod> this way you store them in reversed order, is this what you meant with big-endian? where most significant digits are at the right end of the numer?
09:51:07 <monod> number*
09:51:37 <monod> (dunno if that't the definition)
09:51:42 <ShadowDemon> lol
09:51:52 <MK_FG> Yeah, so you can shift left side to disk eventually
09:52:14 <monod> "shift left side to disk"?
09:52:45 <MK_FG> With LE byte order, you'll have to know the number size in advance, I think, as least significant byte should be written to the last byte on disk
09:53:02 <ShadowDemon> http://www.amazon.com/CODE-ebook/dp/B004OR1XLA/ref=tmm_kin_title_0
09:53:13 <MK_FG> Assuming pi has infinite digits, you can't just accumulate gigs of these to RAM
09:53:30 <MK_FG> And least-significant part stops changing after few 10*x
09:53:38 <ShadowDemon> :)
09:53:41 <MK_FG> So can be safely written-out and pushed out of RAM
09:54:45 <monod> hmm
09:55:09 <monod> I'm still struggling a bit, but I've understood the fact the you may not know how many digits are there
09:55:16 <monod> even though I "only" have 200m
09:56:05 <monod> "left side" was the left side of the number, the most significant digits, or the left side in this new layout, then the least significant digits?
09:56:51 <MK_FG> least-significant
09:57:04 <HiveResearch> digital analysis allows you to request any digit position of pie that you want
09:57:11 <HiveResearch> *pi
09:57:30 <HiveResearch> even if there are an infinity of them
09:58:26 <monod> what do you mean HiveResearch ?
09:58:29 <monod> :O
09:59:11 <HiveResearch> there is a formula that gives you any digit of pi you want
09:59:12 <monod> there exist algorithms that yeild the n-th digit of pi??
09:59:17 <HiveResearch> yes
09:59:26 <HiveResearch> digital analysis
09:59:29 <monod> did you read it on wikipedia?
09:59:44 <HiveResearch> i read it somewhere
10:00:01 <monod> because I found something similar, but it appeared to only be a summation
10:00:12 <monod> okay, I'll search digital analysis
10:00:22 <HiveResearch> i think that's what it's called
10:00:33 <monod> even though.. we were discussing compression, but thanks anyway :D
10:00:50 <HiveResearch> you could use that for compression
10:01:14 <monod> might it be DIGITS ANALYSIS??
10:01:23 <HiveResearch> mebbe
10:01:45 <monod> oh god, I did this sort of number pattern searching back into my 16 *_*
10:02:08 <monod> I didn't know there already was something like that, but obviously I was expecting so
10:04:28 <monod> "Digital Analysis is the analysis of data to detect abnormal occurrences of specific digits, digit combinations, specific numbers, and round numbers." Hive, so it might be called so
10:04:54 <monod> and may not xD
10:04:59 <monod> just, don't know yet.....
10:05:04 <monod> and, nevermind
10:08:45 monod has quit (User quit:  gotta go)
11:05:54 <ShadowDemon> I'll be back fellas.
11:09:05 ShadowDemon has quit (Ping timeout)
11:18:01 AnonO_o (AnonO_o@anonO_o) has joined #crytocc
11:18:08 <AnonO_o> joe, you're a bastard.
11:18:30 <AnonO_o> I should have never clicked on redeclipse
11:18:41 <AnonO_o> addictive little thing.
12:23:57 Angelina has quit (Input/output error)
12:24:24 Angelina (chatzilla@cryto-BFEA5E32.snydernet.net) has joined #crytocc
12:37:08 AnonForecast has quit (Ping timeout)
12:44:18 AnonO_o has quit (Input/output error)
12:46:10 AnonO_o (AnonO_o@anonO_o) has joined #crytocc
13:14:26 HiveResearch has quit (Ping timeout)
13:28:03 HiveResearch (HiveResear@developers.developers.developers) has joined #crytocc
13:33:30 HiveResearch has quit (Ping timeout)
13:33:57 HiveResearch (HiveResear@developers.developers.developers) has joined #crytocc
13:35:16 mama (anon@cryto-67EE065B.dhcp.dbnet.dk) has joined #crytocc
13:41:49 joepie91 has quit (Ping timeout)
13:42:28 joepie91 (joepie91@cryto-3E6002EF.direct-adsl.nl) has joined #crytocc
13:51:18 joepie91 has quit (Ping timeout)
13:51:55 joepie91 (joepie91@cryto-3E6002EF.direct-adsl.nl) has joined #crytocc
13:53:08 AnonO_o has quit (Input/output error)
14:00:59 foolex has quit (Ping timeout)
14:05:03 S1renide (S1renide@cryto-54383BCE.us-west-1.compute.amazonaws.com) has joined #crytocc
14:05:04 foolex (foolex@AD356075.7DC890E0.CEC56216.IP) has joined #crytocc
14:06:24 S1renide has quit (User quit:  Connection closed)
14:22:10 Luxor (Luxor@cryto-24C3E6D6.red.bezeqint.net) has joined #crytocc
14:24:18 *** Xeross|AFK is now known as Xeross
14:56:19 x has quit (Ping timeout)
14:57:02 x (foobar@C35CA8A8.589C91BA.8F6A2B14.IP) has joined #crytocc
15:19:59 <x> mornin o/
15:21:36 <mama> o/
15:21:46 <x> mama <3
15:21:48 <x> hai :3
15:21:54 <mama> x <3
15:22:01 <mama> hello :D
15:22:07 <x> hai :3
15:22:11 <x> ^_^
15:22:18 <mama> haha
15:22:27 <x> just woke my butt up
15:22:40 <mama> x I expected you will join the meeting yesterday
15:22:52 <mama> only your butt?
15:23:24 <x> ohh sorry :( was up most of the night researching
15:23:36 <x> lol yus, my head is still asleep
15:23:45 <mama> hahahah
15:24:03 <x> brb a bit :)
15:24:07 <mama> you can join #ideas any time
15:24:20 <mama> i have to go afk too
15:24:24 <mama> TTYL
15:47:24 HiveResearch has quit (Ping timeout)
15:56:33 Luxor has quit (Ping timeout)
16:04:16 iceTwy (iceTwy@iceTwy.users.cryto) has joined #crytocc
16:04:20 <iceTwy> Hao
16:04:22 <iceTwy> Hai*
16:07:11 x has quit (Input/output error)
16:16:17 Luxor (Luxor@cryto-24C3E6D6.red.bezeqint.net) has joined #crytocc
16:16:47 Luxor has parted #crytocc (Leaving)
16:21:39 kvasir (kvasir@cryto-9B6B837F.hsd1.ca.comcast.net) has joined #crytocc
17:21:02 kvasir has quit (User quit:  three sheets to the wind)
17:24:59 Angelina has quit (Ping timeout)
17:32:09 Angelina (chatzilla@cryto-70E29873.torservers.net) has joined #crytocc
17:50:41 emptyreddata (emptyred@856A53CC.B454AB78.6B89ED8F.IP) has joined #crytocc
17:57:48 AnonO_o (AnonO_o@anonO_o) has joined #crytocc
18:21:44 Angelina has quit (User quit:  ChatZilla 0.9.90 [Firefox 17.0.5/20130401162833])
18:36:20 ShadowDemon (alexgurrol@cryto-507D6D5.hsd1.ca.comcast.net) has joined #crytocc
19:53:06 zest (zest@cryto-BBDC76E6.tor.uwaterloo.ca) has joined #crytocc
19:55:23 BigAnon (BigAnon@E929C8BA.495D65D2.80CBFD1A.IP) has joined #crytocc
20:12:12 x (foobar@cryto-4ABAF172.as5577.net) has joined #crytocc
20:47:30 monod (~pmpf@monod.users.cryto) has joined #crytocc
20:47:45 <monod> hello people
21:11:27 <iceTwy> hi
21:16:14 <iceTwy> herp
21:16:30 <iceTwy> I hate people relying on someone else's reputation to prove a point
21:16:48 <iceTwy> This Arch user's going all like, "yeah, this Arch official dev is totally correct to tell you not to use XX"
21:16:51 <iceTwy> >why?
21:17:07 <iceTwy> >"because he's an official dev!"
21:17:09 <iceTwy> >_<
21:25:54 <MK_FG> It's almost always the case though
21:26:19 <MK_FG> Just as you rely on your school teachers and book authors for even basic math knowledge
21:29:47 <iceTwy> True..
21:29:49 <iceTwy> Well.
21:30:05 <iceTwy> We're mostly demonstrating stuff in maths. So I wouldn't really be okay with this point
21:31:46 <MK_FG> Should work with pretty much every thing you haven't tested or unable to derive yourself, I think
21:32:05 <iceTwy> Derive? Lawl. That only applies to functions though
21:32:14 <iceTwy> We don't /only/ study function-related content.
21:32:29 <MK_FG> Nah, I meant in a more general sense
21:32:46 <MK_FG> Like, "You say earth is round, why?"
21:33:05 <iceTwy> Oh
21:33:11 <MK_FG> And while there's simple "these reputable scientists told me so" reply...
21:33:27 <MK_FG> You can also construct gozillion proofs based on physics yourself ;)
21:33:40 <iceTwy> True! And that's where curiosity kicks in...
21:33:46 <iceTwy> I love being curious. Always been, always will be
21:33:55 <iceTwy> Probably wouldn't be chatting with you if I wasn't curious
21:37:40 <joepie91> <iceTwy>I hate people relying on someone else's reputation to prove a point
21:37:43 <joepie91> appeal to authority? :P
21:38:03 <iceTwy> Not really
21:38:15 <iceTwy> When people are just repeating what they've been told
21:38:16 <joepie91> also, personally, I couldn't give a rats ass how supposedly knowledgable someone is about something
21:38:27 <iceTwy> Because someone of a certain reputation has told them so, I just hate it
21:38:30 <joepie91> if he cannot support his statement with arguments or explanation, I consider it invalid
21:38:33 <iceTwy> That
21:38:56 <joepie91> iceTwy: yes, isn't that pretty much what an appeal to authority is...?
21:39:12 <iceTwy> well...
21:39:15 <iceTwy> what kind of authority
21:40:28 <joepie91> https://en.wikipedia.org/wiki/Appeal_to_authority
21:40:39 <joepie91> The argument from authority can take several forms. As a statistical syllogism, the argument has the following basic structure:[1]
21:40:39 <joepie91> Most of what authority A has to say on subject matter S is correct.
21:40:39 <joepie91> A says P about subject matter S.
21:40:39 <joepie91> Therefore, P is correct.
21:41:52 <MK_FG> ...wherein joepie91 appeals to wikipedia to argue that it's called "appeal to authority" ;)
21:42:05 <joepie91> no, not really :P
21:42:21 <joepie91> because references
21:43:56 AnonO_o has quit (Input/output error)
21:51:24 AnonForecast (AnonForeca@AnonForecast.users.cryto) has joined #crytocc
21:57:42 <iceTwy> joepie91: I am a reference
21:57:47 <iceTwy> period
21:57:48 <iceTwy> goodbye
21:57:54 <monod> cya
21:57:57 <iceTwy> no but
21:58:00 <iceTwy> i'm not actually going
21:58:01 <joepie91> ttyl monod
21:58:04 <joepie91> oh
21:58:04 <monod> hahaha
21:58:05 <iceTwy> huehue
21:58:07 <joepie91> I thought you were leavin
21:58:09 <joepie91> leaving *
21:58:10 <joepie91> CONFUSE
21:58:12 <monod> yeah
21:58:14 <iceTwy> monod thought I was leaving
21:58:16 <iceTwy> huehuehue
21:58:17 <MK_FG> DECEPTION
21:58:19 <monod> hahaha
21:58:20 <iceTwy> AW YEAH
21:58:30 <monod> conspiration!
21:58:37 <iceTwy> My little finger's also swollen
21:58:38 <iceTwy> ffs.
21:58:49 <iceTwy> note to self; don't take a volley ball on your little finger next time
21:58:56 <MK_FG> Let's just assume they're terrorist
21:59:14 <iceTwy> MK_FG: no, you're a terrorist
21:59:18 <iceTwy> :<
21:59:32 <MK_FG> Indeed I am, indeed I am...
21:59:56 <monod> lawl
21:59:57 <joepie91> btw, hai emptyreddata!
22:00:07 <joepie91> iceTwy: stupid, stupid
22:00:22 <joepie91> also
22:00:23 <iceTwy> joepie91: impossible, how
22:00:27 <joepie91> I assume all of you
22:00:29 <joepie91> have seen the new zpanel vuln?
22:00:33 <iceTwy> no
22:00:34 <iceTwy> link?
22:00:41 <joepie91> oh dear
22:00:44 <joepie91> you're in for a treat then
22:00:44 <joepie91> sec
22:00:56 <joepie91> http://seclists.org/fulldisclosure/2013/Jun/39
22:00:58 <MK_FG> It's not that interesting without teh drama
22:01:21 <joepie91> well
22:01:29 <joepie91> I'd consider "root your host in 5 easy steps!" kind of interesting
22:01:29 <joepie91> :p
22:01:53 <MK_FG> Ahah
22:02:40 <iceTwy> ffs. really, zpanel?
22:02:50 <joepie91> iceTwy: bonus: accessible to any user
22:02:52 <joepie91> not just admin
22:03:13 <iceTwy> meaning I could just go on any server running zpanel and voila
22:04:15 <emptyreddata> Hi there sir Joepie. What's good?
22:04:39 <monod> who uses zpanel? & what is zpanel used for?
22:04:44 <joepie91> iceTwy: if you have an account there, then yes
22:05:10 <joepie91> emptyreddata: not zpanel :P
22:05:20 <emptyreddata> Haha, is that still going horribly wrong?
22:05:22 <joepie91> monod: too many people, and it's used for hosting
22:05:23 <joepie91> well
22:05:27 <joepie91> a new root command exec vuln was found
22:05:33 <joepie91> accessible to every access level user, this time
22:05:46 <joepie91> including instructions on how to disable firewall, change root pass, and SSH in
22:05:51 <joepie91> so yeah
22:05:58 <joepie91> at least this time there was a decent response from the team
22:06:08 <joepie91> so things are slowly getting better
22:06:54 <emptyreddata> It's good they actually took the route of responding instead of name-calling this time around.
22:07:42 <joepie91> indeed
22:11:34 Angelina (Administra@Angelina.users.cryto) has joined #crytocc
22:14:19 <zest> "f*cken little know it all" lol ...
22:17:18 <iceTwy> Heading to sleep
22:17:19 <iceTwy> Night
22:17:33 <joepie91> night :)
22:17:41 iceTwy has quit (User quit:  Quit)
22:19:29 <monod> see you ice!
22:27:37 <monod> gotat go guys
22:27:47 <monod> for those willing to study physics:
22:28:50 <monod> 1 cinematics 2 dynamics 3 relativity (galileian's) 4 rigid bodies 5) thermodynamics
22:29:08 <monod> it's a short list of main topics in a first physics course
22:29:17 <monod> of a certain level..
22:29:19 <monod> goodbye
22:29:20 <monod> golks
22:29:22 <monod> folks*
22:29:26 monod has quit (User quit:  .)
22:32:59 <zest> where electricity ?
23:06:03 Angelina has quit (User quit:  leaving)
23:19:15 AnonO_o (AnonO_o@anonO_o) has joined #crytocc
23:31:39 x has quit (Input/output error)
23:32:27 ShadowDemon has quit (Ping timeout)
23:46:19 x (foobar@cryto-4ABAF172.as5577.net) has joined #crytocc