Timestamps are in GMT/BST.
[0:05] * jay (jcl@ool-18bf6dac.dyn.optonline.net) Quit ("Client exiting")
[0:06] * sanity (~ian@81-178-76-180.dsl.pipex.com) Quit (Read error: 232 (Connection reset by peer))
[0:07] * sanity (~ian@81-178-76-180.dsl.pipex.com) has joined #freenet
[0:49] * Disconnected.
[15:36] * hapi (~PircBot@newton.ncc.edu) has joined #freenet
[15:36] * Topic is 'Upgrade to 5096 (5095 is mandatory) | Unstable: Upgrade to 60241 (60234 is mandatory) | Channel logs: http://newton.matcmp.ncc.edu/~lockej/freenet/chanlog/ | #freenet-politics and #freenet-chat are available for offtopic discussions'
[15:36] * Set by toad_ on Fri Sep 24 14:57:57 EDT 2004
[15:37] <toad_> greycat: possibly, I have heard stuff but mostly it's possible to write them off as honest mistakes
[15:38] <toad_> bbl
[15:39] <plixed> right: "rules of engagement" http://www.imdb.com/title/tt0160797/
[15:54] * jay (~jcl@198.38.10.200) has joined #freenet
[16:31] * Hory (~Miranda@81.196.25.110) Quit (Connection timed out)
[16:38] <spaetz> toad: you still need the simulation boxes? No hurry, just curious when I can shut it down again...
[16:39] <spaetz> it's sad to see the only reaction from the army to these scandals is to ban mobiles with a camera.
[16:39] <spaetz> Doesn't sound like they are honestly trying to be better next time, they just don't want to be caught again.
[16:52] * spaetz (~spaetz@217-162-197-136.dclient.hispeed.ch) Quit ("bye, see you later")
[17:31] * greycat (~wooledg@192.35.79.70) Quit ("Lick Bush in '04")
[17:35] * guido^pe (~unknown@dsl-213-023-247-096.arcor-ip.net) has joined #freenet
[17:44] * FallingBuzzard (~srademach@207.152.112.129) has left #freenet
[17:47] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) Quit (Client Quit)
[18:14] * hobx (~hobx@h196n1fls21o1077.bredband.comhem.se) has joined #freenet
[18:28] * Digger3 (~digger3@kf-dn-tg01-174.dial.kabelfoon.nl) Quit ("Lost terminal")
[18:35] * d-ArkAngel (~Rob@cpc2-midd2-5-0-cust16.midd.cable.ntl.com) has joined #freenet
[18:35] <d-ArkAngel> Evening all
[18:35] <cbreak> hi
[18:40] <d-ArkAngel> I just had an interesting thought... could we, rather than using an arbitrary HTL work out a way to auto tune HTL based on network conditions?
[18:41] <d-ArkAngel> just thinking that when we get slashdotted and there are hundereds of stupid nodes that know nothing about the network then the number of HTL before we get a right answer will be a lot higher than when the network is nicely trained...
[18:42] <cbreak> also, the load due to retries because of failed nodes is higher, which would be further increased by higher htl.
[18:42] <d-ArkAngel> and if each node has a max HTL based on it's own expectations of routing in the current network, and if it recieves a request with a huge HTL (i.e. from a node that has just joind and has no hope of routing quickly) then it could reduce the HTL to it's own MAX HTL
[18:43] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[18:44] <d-ArkAngel> but network load is going to increase in situtaions where huge numbers of nodes join, but converstly hustge amounts of new resources become avaliable too (since there are all these new nodes arround)
[18:44] <cbreak> not at the same time
[18:45] <cbreak> until a node is of service for the network, he has to learn routing and get content in its DS.
[18:45] <d-ArkAngel> no, but the "lag" factor is based on how fast the nodes can learn
[18:45] * lolo-laptop (~lostlogic@68.251.84.226) has joined #freenet
[18:46] <d-ArkAngel> lol, maybe what we need to do is limit the bandwidth on the FTP that freenet.jar is on to be the size of the download devided by the length of time it takes to train a node :-)
[18:47] <cbreak> not good idea.
[18:47] <d-ArkAngel> I know, that was mostly a joke :-)
[18:48] <d-ArkAngel> I see what you mean with the HTL thing causing load. but to some extent we have to expect peaks of load when people join the network.
[18:49] <d-ArkAngel> it's either with people retying requests, or with requests with large HTL's
[18:50] <d-ArkAngel> with large HTL requests, at least there's an increased chance that you'll get a data source for your request, and therfor have someone to add to you RT as a known good location to find a area of the keyspace...
[18:51] <d-ArkAngel> or would retyed requests be better for learning?
[18:51] <cbreak> If I'd know, Toad wouldn't be simulating all the time :)
[18:52] <d-ArkAngel> anyway, it was just something that popped into my head while working on the threaded sim code.
[18:53] <d-ArkAngel> which I shall now get back to :-)
[18:54] <d-ArkAngel> one thing that this might help with is increasing the number of people in your RT (improve the speed a node can get new peers when it joins the network)
[18:54] <cbreak> what's the max number of connections by default?
[18:54] <cbreak> For Mac OS X it's 128.
[18:54] <d-ArkAngel> 200
[18:55] <d-ArkAngel> for linux
[18:55] <d-ArkAngel> (mines set to 200, and I've not specifically got a max connections line in my .conf file
[18:55] <cbreak> and how many are used?
[18:55] <cbreak> My node only uses about 64 Connections.
[18:56] <d-ArkAngel> I'm at 41 outs at the moment
[18:56] <d-ArkAngel> but my node isn't very old
[18:57] <d-ArkAngel> Connections open (Inbound/Outbound/Limit) 74 (33/41/200)
[18:57] <cbreak> Here, the total number of connections more or less stays the same.
[18:57] <cbreak> But during the day, outbound decreases, and Inbound increases... strange...
[18:58] <d-ArkAngel> well they're all bidi anyway aren't they?
[18:58] <cbreak> yes.
[18:59] <d-ArkAngel> so it's down to needing to establish connections. I suppose if you use your node a lot you'll have more outbounds, because you node feels the need to connect to nodes. but if it's just passing them though it makes no new interconnects right?
[18:59] * jay (~jcl@198.38.10.200) Quit (".")
[18:59] <d-ArkAngel> but it will still receive interconnects from outside, hence inbound outnumbers outbound during times you're not using the node
[19:00] <cbreak> hmm... sounds reasonable...
[19:00] <d-ArkAngel> but when you use the node you may connect to the DS of nodes that give you answers
[19:00] * Zxcvb (~somebody@client-216-176-88-97.consolidated.net) has joined #freenet
[19:00] <d-ArkAngel> all pure guesswork of course :-)
[19:00] <d-ArkAngel> but that's the kinda guy I am ;-)
[19:01] <Zxcvb> not sure if this is relevant, but has anyone contacted indymedia about using freenet for their site?
[19:01] <d-ArkAngel> indymedia?
[19:01] <cbreak> afaik kind of an independand news agency
[19:02] <Zxcvb> an independant media site where anyone can become a journalist, from what I understand
[19:02] <Zxcvb> it was shut down recently, they are running on backups but it is likely the backup servers will be siezed as well
[19:02] <Zxcvb> http://nyc.indymedia.org/ has info on the shutdown
[19:03] <d-ArkAngel> well let them know about freenet then :-)
[19:03] <cbreak> If they need reliability and a huge audience (which they probably will), I doubt freenet is the right platform for them...
[19:05] <d-ArkAngel> no, but if they get all their hosting nicked by the FBI then it's better than the site being totally offline, and it might get us some nice new node ;-)
[19:06] * d-ArkAngel worries about getting slashdotted tho
[19:06] <d-ArkAngel> how "big" is indymedia?
[19:07] * d-ArkAngel stops pretending that he's some kinda desssion maker, and goes back to being a code grunt ;-)
[19:07] <Zxcvb> it is not that big, and each story is a seperate page
[19:08] <d-ArkAngel> I meant big as in popularity
[19:08] <Zxcvb> not sure, but isn't freenet designed to handle floods by mirroring?
[19:08] <d-ArkAngel> floods of requests isn't too much of a problem
[19:08] <d-ArkAngel> but floods of new nodes makes the network unhappy
[19:08] <cbreak> freenet mirrors everything that passes through a node with a given probabilty.
[19:10] <d-ArkAngel> since it takes time for the new nodes to properly integrate with the network, and during that time they have poor routing ability, and if there is a reasonable % of the network that is made up of "stupid" (i.e. new) nodes then routing sucks and performance drops
[19:10] <Zxcvb> sounds like it might be better for people to continually create offsite mirrors, and publish on freenet whatever gets taken down
[19:11] <d-ArkAngel> well there's no reason the site couldn't exist in freenet
[19:11] <cbreak> also, freenet has only static content. no updating.
[19:11] <Zxcvb> that could be a problem, but I guess no comments is better than no content
[19:11] <d-ArkAngel> the downside is that inserting is slow, and people couldn't just switch over to using freenet overnight
[19:12] <cbreak> it makes things a bit harder for a news agency to stay update.
[19:12] <cbreak> one could wget the page and insert it in freenet on a reqular base (DBR, with a 12 or 24 Hour cycle)
[19:12] <d-ArkAngel> cbreak: how does the message boards work in freenet?
[19:12] <Zxcvb> cbreak: would the idea of publishing on freenet if it is taken down, but not running the whole thing on freenet?
[19:12] <Zxcvb> cbreak: your idea would work as well
[19:13] <d-ArkAngel> could the code to get the latest post be coded in javascript?
[19:13] <cbreak> d-ArkAngel: they post ksks with a deterministic name
[19:13] <Zxcvb> cbreak: to keep up to date, all you would do is post a new page for the new story, and replace the main page that links to the stories
[19:13] <d-ArkAngel> so you can calculate what the ksk of the n'th post is going to be?
[19:14] <cbreak> like KSK@frost/boardname/post-2004.10.06-1
[19:14] <Zxcvb> cbreak: a postfreenetd could probably handle that
[19:14] <cbreak> freenet does not allow for script code in freepages.
[19:14] <d-ArkAngel> dang
[19:14] <d-ArkAngel> css?
[19:14] <cbreak> (because it's a huge, and I mean HUGE security hole)
[19:15] <cbreak> css works, I think.
[19:15] <cbreak> at least on my page :)
[19:15] <Zxcvb> what is the key that lets you update/replace information?
[19:15] <Zxcvb> hopefully with a passphrase
[19:16] <cbreak> DBRs redirect based on the clients time.
[19:16] <cbreak> thats the most dynamic key type.
[19:16] <Zxcvb> no, the one that lets you run a normal static website, and update/replace the content
[19:16] <cbreak> combined with an SSK, it allows for updateable keys, which are read only for the masses.
[19:16] <Zxcvb> SSK, I think
[19:16] * toad_ (toad@toad-with-underline.active.supporter.pdpc) Quit (Remote closed the connection)
[19:16] <cbreak> no.
[19:17] <d-ArkAngel> well a site could have a CSS sheet that contains all the stylation information, so a news article could be posted using the same look feel... and then rather than having "editions" you could have "storys"
[19:17] <cbreak> SSK is only a Public/Private key, where only one can post.
[19:17] <Zxcvb> yes, that is the one
[19:17] <cbreak> but you can't update SSKs.
[19:18] <Zxcvb> indymedia could just replace their SSK every 24 hours or whenever they are shut down
[19:18] <Zxcvb> or is it CSK that lets you do that
[19:18] <cbreak> If they use DBR (Thats not a key type), they HAVE to update it each 24 hours
[19:18] <cbreak> or it will become inaccessible.
[19:19] <Zxcvb> so it is not possible to change/remove a SSK?
[19:19] <cbreak> If they used edition based pages, they can publish a new one at will.
[19:19] <cbreak> no.
[19:19] <cbreak> SSKs are like KSKs, only that they are read only without the private key.
[19:19] <Zxcvb> what is a KSK then?
[19:19] <cbreak> (Afaik :)
[19:20] <cbreak> A Key type with a clear text name.
[19:20] <d-ArkAngel> so the frontpage could be a DBR, and then the "storys" could be SSK
[19:20] <cbreak> as in KSK@gpl.txt
[19:20] <cbreak> d-ArkAngel: DBR Pages are most likely SSKs.
[19:20] <cbreak> I doubt anyone interested in maintaining a freepage would chose CHK or KSK in favor of SSK.
[19:21] <d-ArkAngel> right, but you see what I mean...
[19:21] <cbreak> A DBR main page, that links to edition stories. yes :)
[19:22] <d-ArkAngel> and you could store all the "static" elements like the banner, css file, etc in a single page that you update less often to improve performance by allowing it to cache more "deeply" into the network
[19:23] <Zxcvb> of course, if DBRs can no longer be published, there is a problem
[19:23] <cbreak> unless you use a container page, all files are independent keys.
[19:23] <Zxcvb> unless you can detect which SSK the DBR is currently redirecting to
[19:23] <cbreak> you can. they are deterministic :)
[19:24] <Zxcvb> ah, so if there are no more updates, just publish the SSK instead
[19:24] <cbreak> ?
[19:24] <d-ArkAngel> it works like this...
[19:25] <Zxcvb> suppose indymedia publishes their site with a DBR
[19:25] <d-ArkAngel> a DBR is just an SSK page that has some code built in so that at a certain date time that page automaticaly redirects you to a new version of the page.
[19:25] <Zxcvb> the fbi/nsa/cia/whatever shut down indymedia completely, no more updates are possible
[19:25] <d-ArkAngel> and the new version of the page has a caclulatable key (think of it for example having the date on the end)
[19:26] <cbreak> example of a DBR Page: SSK@sPolFAwS69lG2Le%7eTTRzNEVarcEPAgM/DFI//
[19:26] <cbreak> it isn't maintained.
[19:26] <d-ArkAngel> cbreak how quick are DBR's?
[19:26] <cbreak> The last inserted DBR Page is at SSK@sPolFAwS69lG2Le~TTRzNEVarcEPAgM/40b28c80-DFI//
[19:26] <Zxcvb> now nobody can access it because there are no more updates
[19:26] <cbreak> Quick? The update interval is up to the publisher.
[19:26] <d-ArkAngel> could it be fast enough to run a webcam on? :-)
[19:27] <cbreak> Someone (Fish?) had a webcam with a 10 Min DBR :)
[19:27] <cbreak> he even had a streaming radio over freenet.
[19:27] <Zxcvb> cbreak: the update interval would be 60 times a second for a webcam
[19:27] <cbreak> but that's a long time ago.
[19:27] <cbreak> Zxcvb: I doubt you could insert that ammount of data in freenet.
[19:28] <Zxcvb> cbreak: one update per frame, if you are doing realtime webcam video
[19:28] <d-ArkAngel> maybe if you insert the data ahead of time? like running the cam on a long delay....
[19:28] <cbreak> Humans only see 25 FPS :)
[19:28] <Zxcvb> d-ArkAngel: that would work
[19:28] <d-ArkAngel> that way you could theoreticaly do inserts in parallel.
[19:28] <Zxcvb> cbreak: not true, try forcing 25 FPS on a game like doom
[19:29] <d-ArkAngel> no, cbreak is right
[19:29] <Zxcvb> cbreak: or set your monitor's refresh rate to 25
[19:29] <cbreak> I played doom with 25 Frames a few years ago.
[19:29] <d-ArkAngel> the difference is that doom at 25fsp has no "speed blur"
[19:29] <cbreak> My Monitor has a refresh rate of arround 30 frames (or 60?)
[19:29] <d-ArkAngel> i.e the objects can move faster across the screen then the screen is updateing
[19:29] <Zxcvb> not to mention that in many games, 25fps will make the entire game run at less than 50%
[19:29] <Zxcvb> run at less than 50% speed
[19:30] <cbreak> no one would programm a game that stupid.
[19:30] <cbreak> Frames are independant from the game mechanic, I would think.
[19:30] <Zxcvb> for PC games you might be correct
[19:30] <d-ArkAngel> yes and no
[19:31] <Zxcvb> start up epsxe and set the frames per second to 120, the game will run at double speed
[19:31] <d-ArkAngel> in real life if something moves across your field of view, light is reflected to your eyes from every position it passed thought
[19:31] <Zxcvb> I have even had a game crash on me because I was using a 386 which was too fast
[19:32] <cbreak> Zxcvb: Stupid programmers.
[19:32] <d-ArkAngel> yup, FPS is not related to game speed
[19:32] <Zxcvb> cbreak: the same people later used a trick to allow access to 32 bit flat memory while in real mode (no DPMI, no protected mode, etc)
[19:32] <d-ArkAngel> if it were then how would games work multi player when people have different specs of graphics cards?
[19:33] <Zxcvb> d-ArkAngel: so why does changing the FPS in epsxe change the game speed?
[19:33] <d-ArkAngel> because of the way it's changing the FPS by altering the programs perception of "time"
[19:33] <Zxcvb> d-ArkAngel: unless consoles are different, and actually do use FPS as the game clock
[19:33] <Zxcvb> d-ArkAngel: PAL games often run slower than NTSC ones for example, though newer systems have overcome that limit
[19:34] <cbreak> consoles are different because they often sync to the output device.
[19:34] <cbreak> PAL Has 25 FPS, NTSC has 23.97
[19:34] <d-ArkAngel> and they're designed to be a specific spec beforehand
[19:34] <Zxcvb> d-ArkAngel: typically, a PAL game is made to be the same speed by altering the delay loop
[19:34] <d-ArkAngel> for example everyone knows what speed the proccesor is in a PS2, it's always the same
[19:34] <cbreak> (Of course PAL realy has 50 FPS Interleaved, ...)
[19:35] <Zxcvb> d-ArkAngel: take a NES game, and run it on a PAL console and NTSC one
[19:35] <d-ArkAngel> so they don't need to do "real time" codeing of anything. they KNOW what time it is based on how many frames that have drawn
[19:35] <Zxcvb> d-ArkAngel: the PAL game will have slower sound/music, slower gameplay, etc
[19:35] <cbreak> Zxcvb: PAL has more frames per second. It should be faster if you are correct.
[19:35] <Zxcvb> d-ArkAngel: no, pal is 50, ntsc is 60
[19:36] <d-ArkAngel> UT 2004 doesn't change speed one any platform at any number of FPS
[19:36] <Zxcvb> d-ArkAngel: it must compensate somehow
[19:36] <d-ArkAngel> there are TWO destinct things going on
[19:36] <d-ArkAngel> the physics engine
[19:36] <cbreak> All games I have written draw frames if they have time to do so.
[19:36] <d-ArkAngel> which knows where things are, and what is happening
[19:37] <d-ArkAngel> and the graphics engine, which draws a pretty picter of what you can see.
[19:37] <d-ArkAngel> they run independantly on an interrrupt wait cycle
[19:37] <cbreak> and they calculate the new state of the world if they have time to do so. Both are independant tasks.
[19:37] <d-ArkAngel> if there is any time left over from what the physics engine is doing then it draws more FPS
[19:38] <d-ArkAngel> that's why games have a min spec.
[19:38] <d-ArkAngel> it's the amount of CPU time needed to run the physics engine, and still have enough CPU time left to do the graphics engine at a reasonable quality
[19:38] <cbreak> I think most games take the distance between two samples of physic calculations to fix the game speed.
[19:39] <Zxcvb> cbreak: so on a fast computer, the game would be unplayably fast, right?
[19:39] <cbreak> nol
[19:40] <cbreak> on a fast computer, more precise calculations are made.
[19:40] <cbreak> if the distance between two physic samples are short, more samples are made per second.
[19:40] <Zxcvb> ok, apparently console games are frame based with vsync on, not timer based
[19:40] <d-ArkAngel> well not new consoles
[19:40] <cbreak> Well, maybe the older ones.
[19:41] <d-ArkAngel> Old ones might be
[19:41] <Zxcvb> xbox apparently is, otherwise cxbx wouldn't be able to run turok at 4x native speed
[19:41] <d-ArkAngel> but only because they're lazy and rather than having something that keeps track of time, they just cound the number of vsyncs
[19:41] <cbreak> Today, games have a varying complexity, so you can't asume the drawing or physic calculations have the same complexity.
[19:41] <Zxcvb> or mabye cxbx just patches the game in memory to remove the standard speed limiting code
[19:42] * guido^pe (~unknown@dsl-213-023-247-096.arcor-ip.net) Quit (Remote closed the connection)
[19:42] <d-ArkAngel> or they're just telling the xbox that timeis passing more quickly than it reallt is.
[19:42] <Zxcvb> d-ArkAngel: or because you have to draw the scanlines yourself because there is no framebuffer
[19:42] <d-ArkAngel> there is always a framebuffer these days, you need to for pixle shaders for a start
[19:43] <cbreak> If you use Open GL, there is allways a double buffering.
[19:43] <d-ArkAngel> and for anti-alasing
[19:43] <Zxcvb> d-ArkAngel: yeah, it is also alot easier than flipping 3-4 bits at just the right time to change the information sent to the electron beam
[19:43] <d-ArkAngel> and for the myrid of other modern effects that need them :-)
[19:43] <cbreak> beams?
[19:44] <cbreak> who uses crts this days?
[19:44] <d-ArkAngel> it's not really easier... think about it. the computer is running at what 1.4Ghz, that's far faster than a TV scanlines
[19:44] <d-ArkAngel> a ramdac on a PC is bearly 400Mhz
[19:45] <Zxcvb> d-ArkAngel: no, suppose to control color you only had 3 bits
[19:45] <Zxcvb> d-ArkAngel: the game has to keep track of where the electron beam on the monitor is and flip those bits many times per second to draw a picture
[19:45] <cbreak> no.
[19:46] <cbreak> it paints the picture in a buffer.
[19:46] <d-ArkAngel> so on a console it's probably more like 150Mhz, so you've got to flip the 3 bits every 10 cycles or so.
[19:46] <Zxcvb> d-ArkAngel: you have no hardware to help, merely one bit that turns on or off red output, one bit that turns on or off green, and one that turns off blue
[19:46] <d-ArkAngel> yeah, that's called a ramdac.
[19:46] <Zxcvb> cbreak: that is what I am talking about, you didn't have a buffer
[19:46] <d-ArkAngel> it's not the CPI that does that
[19:46] <d-ArkAngel> CPU even
[19:46] <Zxcvb> d-ArkAngel: no ramdac, the .exe itself has to do it
[19:47] <d-ArkAngel> lol
[19:47] <cbreak> You can't rely that two operations are sequential
[19:47] <d-ArkAngel> you'll be telling me there is no CPU next, it's just the exe
[19:47] <d-ArkAngel> the exe does no such thing
[19:47] <cbreak> at every point in time, your programm can be interupted by the scheduler.
[19:47] <d-ArkAngel> the exe issues command which the different bits of hardware do
[19:47] <Zxcvb> d-ArkAngel: and that legacy is probably why pal games are slower than ntsc
[19:47] <d-ArkAngel> lol
[19:47] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[19:47] <Zxcvb> cbreak: no scheduler, your game is the OS
[19:48] <cbreak> you can't controll the ray directly, because you don't allways have the CPU for your programm.
[19:48] <cbreak> so you still have a graphic card, right?
[19:48] <d-ArkAngel> the xbox has a scheduler, trust me. you write code for it in directx
[19:48] <Zxcvb> cbreak: consoles don't have an OS, and for a while didn't even have a framebuffer
[19:48] <d-ArkAngel> not assembler
[19:48] <Zxcvb> d-ArkAngel: the xbox isn't a console, it is a palladium-extreme enabled PC
[19:48] <cbreak> the buffer is in the GPU/RAM
[19:49] <d-ArkAngel> there is at the very least a microcode OS on a console
[19:49] <cbreak> the X-Box runs with embeded Windows NT I think.
[19:49] <Zxcvb> d-ArkAngel: there wasn't on the playstation or dreamcast, most likely not on the PS2
[19:49] <Zxcvb> cbreak: right, the xbox is an sssca compliant PC not a console
[19:49] <d-ArkAngel> if there was no OS then nothing would happen when you turn it on without a CD in it
[19:49] <d-ArkAngel> but it does. like magic.
[19:50] <cbreak> rom :)
[19:50] <cbreak> older macs had a complete OS in its ROM, quite nifty.
[19:50] <Zxcvb> d-ArkAngel: those are apps on the ps1 built into rom, there is no OS
[19:50] <d-ArkAngel> because there is an underlying OS that sits on a ROM and deals with all the standard low level stuff
[19:50] <d-ArkAngel> that's what's on the damn ROM and OS
[19:50] <Zxcvb> d-ArkAngel: on the psx and dreamcast, you have to do the low level stuff yourself or use standard development libraries
[19:51] <Zxcvb> d-ArkAngel: there is no access_memcard_block_2 API on the playstation
[19:51] <d-ArkAngel> do you really think that every PS game have their own implmentation of how to address memory, and how to keep the screen timeings, and how to access the game controlers, and they all write all this in assembler by hand?
[19:51] <Zxcvb> d-ArkAngel: no, they use a standard implementation, which is compiled in
[19:51] <d-ArkAngel> and their own implementation of the TCP/IP stack so they can play networked, and their own hardware conrol code?
[19:52] <Zxcvb> d-ArkAngel: but all that has to be on the cd/cartrige, it is not stored on the console
[19:52] * pupok (~r00t@81-178-76-180.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[19:52] <Zxcvb> d-ArkAngel: think back to an old 8086 with a floppy drive, the app was it's own OS
[19:52] * pupok (~r00t@81-178-76-180.dsl.pipex.com) has joined #freenet
[19:53] <cbreak> it had basic functionality in the computers BIOS/ROM/Whatever
[19:53] <d-ArkAngel> and yet there was still a bios that had functionality....
[19:53] <cbreak> It had a memory mapped interface to the hardware.
[19:53] <Zxcvb> d-ArkAngel: nintendo and sony provide code that you can cut and paste to handle low level stuff
[19:53] <d-ArkAngel> you know anyone who develops PS games?
[19:55] <Zxcvb> d-ArkAngel: I have checked out the source to a few, and none of them just call an API to save, they all have the serial code
[19:55] <d-ArkAngel> because I do. and I am gaurenteeing you right here right now, that there is no insturction in a PS game, antwhere on the CD that says "flip the blue bit on for the gun"
[19:55] <Zxcvb> d-ArkAngel: I still don't see anything like you are talking about
[19:56] <d-ArkAngel> the graphics processing hardware on a PS has come a little way on since the NES
[19:56] <d-ArkAngel> lots of the stuff is implemented in hardware
[19:56] <Zxcvb> d-ArkAngel: so why does imbnes not just call a function to get memory block 2 on slot 1, why does it need serial port like code to read the memcard data itself
[19:57] <d-ArkAngel> there game will be making calls to the hardware to do 32 transformations.
[19:57] <d-ArkAngel> how to do the transforms will NOT be in the code on the CD
[19:57] <d-ArkAngel> just the instructions to do those transforms
[19:57] <Zxcvb> d-ArkAngel: it should be as simple as fileopen(memcard1,block2)
[19:58] <d-ArkAngel> what has that got to do with the thing having a ramdac?
[19:58] <Zxcvb> d-ArkAngel: it has to do with the psx having an OS to abstract all that stuff
[19:58] <d-ArkAngel> you know what the last game was that was written in assembler?
[19:58] <d-ArkAngel> Frontier Elite
[19:59] <d-ArkAngel> that was the last comercial game written totally in assembler.
[19:59] <cbreak> I have written a game in ASM last year :)
[19:59] <d-ArkAngel> (wasn't comercial ;-) )
[19:59] <cbreak> definately not :)
[19:59] <Zxcvb> d-ArkAngel: ff8 mabye?
[19:59] <d-ArkAngel> nope
[19:59] <Zxcvb> d-ArkAngel: how else can it get ps2 like graphics on a ps1
[20:00] <cbreak> because the devs use abstractions
[20:00] <d-ArkAngel> trust me, it wasn't written in assembler
[20:00] <Zxcvb> d-ArkAngel: you have to use assembler for direct hardware access, don't you?
[20:00] <d-ArkAngel> cbreak, you've written a game in assembler. how long do you estimate it would take to write ff8 in assembler?
[20:01] <cbreak> asm is not usable for larger projects. No function calls, not even loops.
[20:01] <d-ArkAngel> about a hundered million man years seem reasonable?
[20:01] <cbreak> I think they might be ready to compete with Duke Nukem Forever.
[20:02] <Zxcvb> cbreak: how else do you directly access the hardware if not by assembler?
[20:02] <d-ArkAngel> they're called compilers
[20:02] <cbreak> with a driver, or memory mapped io.
[20:02] <Zxcvb> cbreak: the idea is to bypass sony's driver code
[20:02] <d-ArkAngel> why?
[20:02] <cbreak> I get a pointer to the address where the graphic card memory is mirrored in the mem space and write there.
[20:03] <Zxcvb> d-ArkAngel: so you can get fancy ps2 type graphics on the ps1, even though the ps1 hardware isn't supposed to support such high quality graphics
[20:04] <cbreak> There are only two things a programm (as Code running on a CPU) can access: Interrupts and Memory.
[20:04] <d-ArkAngel> so you're saying that a console with no OS, can have more functionality added to it just using code?
[20:04] <Zxcvb> d-ArkAngel: yes, if your code is better than the code in sony's devkit
[20:05] <d-ArkAngel> could you get someone to write me a code version of a ATI X800 please?
[20:05] <cbreak> (I read about a cpu with soft core)
[20:05] <d-ArkAngel> yeah, transmeta make one
[20:06] <d-ArkAngel> but that still has a core that sits inside that executes the microcode insrutctions that the "cpu" is programed with
[20:06] <d-ArkAngel> if you open up a PS how much stuff is inside it?
[20:06] <cbreak> yes. I doubt they make it like a Programable Logic Chip :)
[20:07] <d-ArkAngel> by my reconning you've accounted for about 1 CPU chip, and a bunch of RAM chips
[20:07] <Zxcvb> d-ArkAngel: you get more functionality because your code is special purpose, instead of general purpose like the libraries/drivers sony supplies
[20:07] <d-ArkAngel> so what exactly are all the rest doing?
[20:08] <Zxcvb> d-ArkAngel: ever wonder why Oracle uses it's own filesystem instead of the standard ext2/xfs/ntfs filesystem driver provided by the OS?
[20:08] <cbreak> There have to be bridging chips, IO chips, a GPU, voltage regulators, ...
[20:08] <d-ArkAngel> so I can code my Amstrad 464 to run Doom III if I bypass the code that Amstrad shipped with it?
[20:08] <Zxcvb> d-ArkAngel: no, but you can make games with much better graphics
[20:08] <d-ArkAngel> as good as Doom III?
[20:08] <cbreak> Zxcvb: Oracle still uses the os drivers.
[20:09] <Zxcvb> d-ArkAngel: not sure about that
[20:09] <d-ArkAngel> yes you are
[20:09] <cbreak> Doom III calculates A LOT.
[20:10] <d-ArkAngel> the same way you're sure that you couldn't write Doom III to work on a VooDoo 1 graphics card
[20:10] <cbreak> you can not calculate that much on something so old I don't know it :)
[20:10] <Zxcvb> d-ArkAngel: but I have seen some amazing games/demos that are well beyond what should be possible
[20:10] <d-ArkAngel> because the hardware does a whole lot more than just tell the CRT beams when to come on and off.
[20:11] <Zxcvb> cbreak: yet oracle uses it's own filesystem via access to /dev/hd? or /dev/sd?, it doesn't just use a file on the normal filesystem
[20:11] <cbreak> Zxcvb: Demos still use the hardware and are limited by it. they use tricks to make better use, but they are limited.
[20:11] <d-ArkAngel> if it's "well beyond possible" then you can't have seen it.
[20:11] <d-ArkAngel> since it's not possible...
[20:11] <Zxcvb> d-ArkAngel: beyond what should be possible according to most people that worked with it
[20:12] <d-ArkAngel> so why are these things just demos?
[20:12] <Zxcvb> d-ArkAngel: but it just required all sorts of clever tricks to make it work
[20:12] <Zxcvb> d-ArkAngel: they aren't
[20:12] <Zxcvb> d-ArkAngel: check out the graphics on ff8
[20:12] <d-ArkAngel> and the people doing it are not making games that are that good and getting apyed a mint
[20:12] <Zxcvb> d-ArkAngel: not to mention many of the games that use it wouldn't get licensed because they use ulgy tricks
[20:13] <d-ArkAngel> lol
[20:13] <d-ArkAngel> I very much doubt that licencing is based on what the code "looks" like
[20:13] <Zxcvb> d-ArkAngel: check out ultima 7 for an example
[20:14] <Zxcvb> d-ArkAngel: it relies on a bug to get 32 bit real mode, and is incompatible with emm386/qemm/windows/os2/etc
[20:14] <cbreak> Zxcvb: Stupid Devs.
[20:14] <Zxcvb> d-ArkAngel: such tricks can only work when you can be sure the platform will never change, even consoles often change (10 playstation 2 versions, for example)
[20:15] <d-ArkAngel> look, you really have no clue what you're talking about, and you can go on saying "this looks really cool I just don't understand how" till the cows come home.
[20:15] <d-ArkAngel> I'm gaurenteeing now that it's no PS code can by direct instruction turn on or off the RGB beams in your TV
[20:16] <d-ArkAngel> there is a RAMDAC, and a FrameBuffer, and a GPU between your code and the beams, and you can't bypass them
[20:16] <d-ArkAngel> and if you did then you would get WOSER perfromance not better
[20:16] <Zxcvb> d-ArkAngel: mabye not, but there are other tricks that sony's development kit won't let you do
[20:17] <d-ArkAngel> yeah, and Freenet uses custom code instead of some parts of the java SDK
[20:17] <cbreak> so what? :)
[20:17] <d-ArkAngel> but it doesn't mean that it's not running on top of an OS
[20:20] <d-ArkAngel> besides there is deffinatly and OS on a PS, because there's more than one version of the PS.
[20:20] <d-ArkAngel> they brought out hardware revisions to beat the people and their mod chips
[20:20] <Zxcvb> is it a real OS though?
[20:21] <cbreak> No, its purely in software...
[20:21] <d-ArkAngel> what do you mean by "real OS"
[20:21] <cbreak> like in "real men use <insert here>" :)
[20:21] <Zxcvb> the purpose of an OS is to abstract away the hardware
[20:21] <d-ArkAngel> is it an amount of code that runs on the hardware and services the requests of the programs that are running, then the answer is clearly yes
[20:21] <Zxcvb> if you still need your own code to access memory cards and other hardware, it isn't an OS
[20:22] <d-ArkAngel> and because I've just mentioned that there are a number of different hardware revisions of the PS, ie that not every PS has the same stuff in the box
[20:22] <d-ArkAngel> then CLEARLY there's a software layer of abstraction
[20:22] <cbreak> it doesn't provide everything
[20:22] <cbreak> -> the PS2 can run PS One Games, because of some kind of abstraction.
[20:23] <Zxcvb> cbreak: then the PS2 shouldn't run any games that rely on direct hardware access, much like ultrahle can only ever run games that rely on nintendo's devkit
[20:24] <Zxcvb> cbreak: that is, games that implement their own drivers
[20:24] <d-ArkAngel> you can run the PS2 OS on a number of other pieces of hardware for developing games. they don't write code, compile it, then stick it on a CD rom and put it in their PS just to find there is a missing semi colon.
[20:24] <d-ArkAngel> they have versions of the box with totally different hardware.
[20:24] <d-ArkAngel> that has a version of the PS OS on it.
[20:25] <d-ArkAngel> it might be all in ROM, and it Might be none upgradable, but it's still there
[20:25] <cbreak> A PS2 can run all PS One Games I ever heard of.
[20:25] <Zxcvb> d-ArkAngel: the whole point of an OS is to take care of all of the hardware access
[20:25] <d-ArkAngel> Is windows an OS?
[20:25] <Zxcvb> d-ArkAngel: yes
[20:26] * lostlogic (~lostlogic@node-4024215a.mdw.onnet.us.uu.net) has joined #freenet
[20:26] <Zxcvb> d-ArkAngel: DOS is not, you still have to use your own code to access serial ports and such
[20:26] <d-ArkAngel> and yet windows programs still have low level access to the hardware if they want
[20:26] <Zxcvb> d-ArkAngel: no they don't
[20:26] <d-ArkAngel> ummm, what does DOS stand for?
[20:26] <d-ArkAngel> let me give you a clue
[20:26] <Zxcvb> d-ArkAngel: supposed to be disk operating system, but it doesn't actually provide an OS
[20:26] <d-ArkAngel> D_____ Operating System
[20:27] <d-ArkAngel> umm, yeah it does
[20:27] <cbreak> Zxcvb: On a linux, you can access files on a Hard Drive. But you can also acces the bits directly on a /dev entry. Different layers of abstractions.
[20:27] <Zxcvb> d-ArkAngel: the OS is supposed to provide a standard interface to hardware using drivers, so that apps don't need to code their own version
[20:27] <d-ArkAngel> what kind of definition is that?
[20:27] <d-ArkAngel> other than one you just made up to try and win this argument
[20:27] <Zxcvb> d-ArkAngel: in DOS, when you bought a new printer, you had to make sure all your word processors could support it
[20:28] <d-ArkAngel> and?
[20:28] <d-ArkAngel> it's the same in linux
[20:28] <d-ArkAngel> is linux not an OS?
[20:28] <Zxcvb> d-ArkAngel: no it isn't, you just have to make sure there is a ghostscript driver, you don't have to check if openoffice, koffice, etc all support it
[20:28] <Zxcvb> d-ArkAngel: in DOS, you have to go directly to the hardware
[20:29] <d-ArkAngel> and yet the code I use to access my webcam is all using direct low level access to the hardware because there are no drivers...
[20:29] <d-ArkAngel> and yet it's running on linux, which is an OS
[20:29] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) has joined #freenet
[20:29] <Zxcvb> d-ArkAngel: exactly, which you aren't supposed to do, you are supposed to have a driver do it for the app
[20:29] <toad_> hi
[20:30] <d-ArkAngel> ahhh, right, so you can have an OS and yet still do things low level?
[20:30] <d-ArkAngel> hi toad
[20:30] <Zxcvb> d-ArkAngel: and in windows, you aren't allowed to do at all
[20:30] <d-ArkAngel> yes you are
[20:30] <cbreak> You can, If you are priviledged enough.
[20:30] <toad_> oooh fun bashing the metal
[20:30] <Zxcvb> d-ArkAngel: nope, you have to write a driver to relay it
[20:30] <d-ArkAngel> nope
[20:30] <Zxcvb> d-ArkAngel: the HAL prevents low level access by normal apps
[20:31] <d-ArkAngel> false
[20:31] <d-ArkAngel> heard of viruses?
[20:31] <d-ArkAngel> heard the term buffer overflow?
[20:31] <Zxcvb> d-ArkAngel: if it didn't then the sound card's hardware soundblaster emulation could be used, and hacks like vdmsound wouldn't be required
[20:31] <cbreak> SQL Slammer, my favorite :)
[20:32] <cbreak> but CIH may be more appropriate to be mentioned in this context.
[20:32] <d-ArkAngel> the SB emoulation is provided in software by the drivers not hardware
[20:32] <Zxcvb> cbreak: CIH will not run on modern windows OSes because direct hardware access by apps is forbidden
[20:32] <d-ArkAngel> false
[20:32] <d-ArkAngel> CIH will not run on modern OS's because it exploits holes in an older version of the OS
[20:32] <cbreak> so toad_, how's simulating progressing?
[20:33] <d-ArkAngel> you heard about the virus you can catch looking at a jpeg?
[20:33] <toad_> moderately well
[20:33] <d-ArkAngel> you know how that works?
[20:34] <Zxcvb> d-ArkAngel: it still doesn't communicate with the drive controller directly, or the video card directly
[20:34] <d-ArkAngel> it causes your computer to write too much into a buffer and floods code over other areas of memory so they get executed blow the level of the OS
[20:34] <Zxcvb> d-ArkAngel: otherwise it could overclock the video card in software
[20:34] <d-ArkAngel> no reason it couldn't
[20:34] <cbreak> my CPU is overclockable in software.
[20:34] <Zxcvb> d-ArkAngel: it would have to install a driver, just like the current apps do
[20:34] <cbreak> as are some GPUs on x86.
[20:34] <d-ArkAngel> what utter rubish
[20:35] <Zxcvb> d-ArkAngel: why do you think the dlportio _driver_ is required to write to the parallel port
[20:35] <d-ArkAngel> if you get your code executed on the CPU you can do anything
[20:35] <toad_> sure, but not necessarily directly
[20:35] <Zxcvb> d-ArkAngel: in modern windows, you have to be in ring 0
[20:35] <d-ArkAngel> an OS trys to stop you doing it by checking what you're going to do before it does it.
[20:35] <toad_> you may have to jump through hoops to get a driver in
[20:35] <toad_> modern windows is protected mode
[20:35] <toad_> you know what that means? :)
[20:35] <d-ArkAngel> but if you can bypass the checks via a bug you can do what ever the hell you want
[20:35] <cbreak> use a hole to get in super user mode, and you are god.
[20:35] <toad_> it's protected ;)
[20:35] <toad_> d-ArkAngel: indeed
[20:36] <d-ArkAngel> yeah and now there' NX protection too
[20:36] <Zxcvb> toad_: right, the buffer overrun code could install a driver, but not directly do it
[20:36] <d-ArkAngel> but this is all in the contect of a Playstation
[20:36] <Zxcvb> d-ArkAngel: of course, once you have admin, you can install a driver anyway so it's not much harder
[20:36] <cbreak> http://dhcp-32-094.via-eth.ch/~cbreak/freenet/htl.scale.pdf
[20:36] <d-ArkAngel> NO
[20:36] <cbreak> looks promissing, a bit.
[20:36] <d-ArkAngel> it's nothing to do with having ADMIn rights inside the OS
[20:37] <d-ArkAngel> it has to do with getting arbiraty abounts of ASM run on the CPU bypassing the OS checks
[20:37] <cbreak> Zxcvb: If you can get the CPU to be in super user mode and execute your code, you can do anything the os, including its drivers, can do.
[20:38] <toad_> cbreak: what's that?
[20:38] <d-ArkAngel> you can bypass any amount of OS, but still have an OS there. But no matter how much stuff you bypass you can't bypass the RAMDAC on a PS
[20:38] <d-ArkAngel> end of line
[20:38] <cbreak> a size vs htl plot with the four sample points you mentioned yesterday.
[20:39] <toad_> cbreak: hmmm
[20:39] <toad_> use log scale
[20:39] * d-ArkAngel breaths in and goes back to codeing in java that runs on an OS
[20:39] <toad_> and a VM
[20:40] * d-ArkAngel and is still unable to access the RGB bits inside the beam control hardware in his CRT
[20:40] <toad_> <spaetz> toad: you still need the simulation boxes? No hurry, just curious when I can shut it down again... - probably
[20:40] <cbreak> d-ArkAngel: Use a Magnet
[20:40] <toad_> lol
[20:40] <d-ArkAngel> :-)
[20:41] <cbreak> You can even bend the beam itself. looks psycho :)
[20:41] * d-ArkAngel wonders if it's possibly to work out the orientation of all the transistors on a memory chip and deliberatly switch them to cause a magnetic pulse....
[20:42] <toad_> d-ArkAngel: re adaptive HTL... we're getting rid of HTL eventually...
[20:42] <cbreak> do memory chips realy use transistors?
[20:42] <toad_> d-ArkAngel: in any case, i wouldn't have the max htl vary dynamically; maybe the decrement probability or something
[20:42] <d-ArkAngel> yeah, more or less sets of 3 per bit I think
[20:42] <toad_> cbreak: yup
[20:42] <toad_> and capacitors iirc
[20:43] <d-ArkAngel> yeah that's the thought I was just having.
[20:43] <toad_> <d-ArkAngel> but network load is going to increase in situtaions where huge numbers of nodes join, but converstly hustge amounts of new resources become avaliable too (since there are all these new nodes arround)
[20:43] <toad_> that's exactly the problem
[20:43] <d-ArkAngel> you could decrement by the average probability that you can correctly route something on your current node
[20:43] <toad_> increasing the HTL will only make the load WORSE
[20:44] <d-ArkAngel> but if you've got the choise of requests with more HTL that succeed, or more requests because some are failing and being retreid....
[20:44] <cbreak> when you finaly get rid of HTL, you could change the hop probability with load.
[20:44] <toad_> <d-ArkAngel> one thing that this might help with is increasing the number of people in your RT (improve the speed a node can get new peers when it joins the network) - eh? how?
[20:44] <toad_> cbreak: sure
[20:45] <toad_> if we did, there'd be a greater probability of dropping on higher load
[20:45] <toad_> not a smaller one
[20:45] <cbreak> of course :)
[20:45] <d-ArkAngel> yes, but if your requests are all eventually successful, then you meet more data soruces don't you?
[20:45] <toad_> <cbreak> For Mac OS X it's 128.
[20:45] <toad_> <d-ArkAngel> 200
[20:45] <toad_> <d-ArkAngel> for linux
[20:45] <toad_> <d-ArkAngel> (mines set to 200, and I've not specifically got a max connections line in my .conf file
[20:45] <toad_> <cbreak> and how many are used?
[20:45] <toad_> <cbreak> My node only uses about 64 Connections.
[20:45] <toad_> hmmm why do we use half the limit normally?
[20:45] <d-ArkAngel> within sanity limits of course
[20:46] <toad_> d-ArkAngel: eh?
[20:46] <d-ArkAngel> never mind my brains probably foaming at the mouth or something :-)
[20:47] <d-ArkAngel> (personal rule if you can't describe something to someone the second time arround it's either two complex, or just plain wrong)
[20:47] <d-ArkAngel> too complex even
[20:47] <toad_> <Zxcvb> it was shut down recently, they are running on backups but it is likely the backup servers will be siezed as well - WHAT?!
[20:47] <d-ArkAngel> they had a run in with the freedom of speech police (the FBI)
[20:48] <d-ArkAngel> presumably because they sourced storys with details that the FBI regarded as vital to domestic security
[20:49] <d-ArkAngel> the seating plan at a government function, or the shoe size of the first lady or soemthing vital
[20:49] <toad_> well we're not ready for them sadly
[20:49] <d-ArkAngel> indeed
[20:49] <d-ArkAngel> that was the general concesus
[20:49] <d-ArkAngel> made me think some fun things about my doing a freesite myself tho :-)
[20:50] <d-ArkAngel> not that I'll do it before i finish this sim, I promise! :-)
[20:50] <d-ArkAngel> I found and interesting perfromance booster. there's a better datastructure to implment a LRUQueue with
[20:51] <d-ArkAngel> http://java.sun.com/j2se/1.4.2/docs/api/java/util/LinkedHashMap.html
[20:51] <toad_> you won't by any chance have access to serious hardware?
[20:51] <toad_> i.e. more than 2 way SMP?
[20:51] <toad_> d-ArkAngel: that's a slight memory gain, right?
[20:52] <d-ArkAngel> and it removes one layer of function calls, and it's not synchronised (and it's not using a synced data structure inside it either... double gain there ;-) )
[20:52] <Zxcvb> toad_: if you want to remove certain information, then you have to get it all
[20:52] <Zxcvb> toad_: what good is siezing the original servers when the backups still have the same data
[20:53] * Redb3ard (~oylerj@c-24-125-12-101.va.client2.attbi.com) Quit (Read error: 113 (No route to host))
[20:53] <toad_> The FBI's latest anti-free-press actions began at the beginning of October when they visited Indymedia's ISP demanding the removal of identifying information from photographs of undercover police officers that was posted on the Nantes Indymedia website. When asked what the US government was doing requesting the removal of information from a French-run website that contained information about Swiss police actions, the FBI stated th
[20:53] <toad_> "courtesy" to the Swiss government. The FBI agents stated that no laws had been broken, and no crimes had been committed. However, because no identifying information was posted on the website in question, it was unclear what actions the FBI was requesting.
[20:53] <toad_> LOL
[20:53] <d-ArkAngel> :-)
[20:53] <toad_> Zxcvb: i thought there was some problem with the backups?
[20:53] <Zxcvb> toad_: exactly, so how can the FBI comply with that request without getting the backups
[20:54] <Zxcvb> toad_: there is no problem with the backups _yet_
[20:54] <d-ArkAngel> it's the shell game
[20:54] <d-ArkAngel> which server is the next backup uploaded on....
[20:55] <d-ArkAngel> you'll never find the queen ;-)
[20:56] <d-ArkAngel> toad: I got the LRU queue code from here http://javaalmanac.com/egs/java.util/coll_Cache.html
[20:56] <d-ArkAngel> looks like they've got some good examples of java snippets
[20:57] * Redb3ard (~oylerj@c-24-125-12-101.va.client2.attbi.com) has joined #freenet
[20:57] <Zxcvb> bbl
[20:58] <toad_> btw
[20:58] <toad_> any yanks here?
[20:58] <toad_> well anyone really
[20:58] <d-ArkAngel> I'm an "anyone" ;-)
[20:58] <toad_> has there been much mainstream news coverage of that US war crime caught on camera today?
[20:58] <toad_> TV news? or whatever?
[20:59] <d-ArkAngel> that's the first I've heard about it.
[20:59] <toad_> where are you?
[20:59] <d-ArkAngel> Middlesborugh
[20:59] * Zxcvb (~somebody@client-216-176-88-97.consolidated.net) Quit (Read error: 104 (Connection reset by peer))
[20:59] <d-ArkAngel> ;-)
[21:00] <toad_> d-ArkAngel: heh
[21:00] <toad_> any yanks here?
[21:00] <d-ArkAngel> want me to ask one?
[21:00] <d-ArkAngel> at random
[21:01] <toad_> d-ArkAngel: american commander ordering an air strike on a crowd of peaceful albeit rapidly moving protesters, caught on gun camera with sound track, one iraqi eyewitness
[21:01] <d-ArkAngel> nice
[21:02] <d-ArkAngel> hell, I'd order an airstike if I though I'd get away with it... but that's probably why they won't let me have one of their encrypted radio sets....
[21:02] <toad_> US military totally denying everything
[21:02] <toad_> which IMHO makes them accomplices
[21:02] <cbreak> http://dhcp-32-094.via-eth.ch/~cbreak/freenet/htl.log.pdf
[21:02] <toad_> the only way there will be justice is if this gets major press
[21:02] <cbreak> hmm... doubt that will be of use :)
[21:03] <toad_> sadly i didn't catch it on tape
[21:03] * Zxcvb (~somebody@client-216-176-88-210.consolidated.net) has joined #freenet
[21:03] * toad_ hopes to see it on US election campaign adverts :)
[21:04] <Zxcvb> what war crime?
[21:04] <Zxcvb> and no, I rarely if ever watch news on TV
[21:04] <toad_> <toad_> d-ArkAngel: american commander ordering an air strike on a crowd of peaceful albeit rapidly moving protesters, caught on gun camera with sound track, one iraqi eyewitness
[21:04] <Zxcvb> sounds like that vietnam massacre all over again
[21:04] <toad_> well there was some sort of firing going on in a building
[21:04] <Zxcvb> are they sure it was the gun camera?
[21:05] <toad_> the commander orders a hit on the building involved
[21:05] <toad_> when the air support arrives, big crowd runs round from a side street
[21:05] <toad_> no attempt at getting under cover, totally idiotic tactics if they were armed
[21:05] <toad_> and there was an iraqi witness who said they were unarmed
[21:05] <toad_> pilot asks commander what to do about the people
[21:05] <Zxcvb> toad_: in nam, US soldiers would bayonet babies to death, because they had orders to kill every man, woman, and child
[21:06] <toad_> commander tells him to splash them (not his exact words)
[21:06] <toad_> bang
[21:06] <toad_> Zxcvb: indeed
[21:06] <Zxcvb> toad_: so it wasn't over the radio?
[21:06] <Zxcvb> toad_: if it was, then mabye someone was pretending to be the commander
[21:06] <toad_> Zxcvb: video plus audio
[21:06] <toad_> gun camera footage with soundtrack
[21:07] <Zxcvb> toad_: no, was the commander who gave the order physically there, or was it by radio
[21:07] <toad_> well the commander was there, he could see what was going on, he was talking to the pilot by radio
[21:08] <Zxcvb> toad_: but are they absolutely sure it was the commander, not just someone on the radio that sounded like it
[21:08] <toad_> i don't know, they certainly didn't suggest that theory
[21:08] <Zxcvb> toad_: mabye someone in an awacs wanted to "kill them all" and faked it
[21:08] <toad_> party line is the civilians were armed
[21:08] <Zxcvb> toad_: or insurgents
[21:08] <Zxcvb> toad_: or anyone who would want a massacre
[21:08] <Zxcvb> toad_: _everyone_ is armed
[21:09] <Zxcvb> toad_: if a few of the civilians were wearing belts, technically they were armed
[21:09] <toad_> yeah, my first thought was that it was staged by the militants - but there was a witness too - and who's going to volunteer to run out in front of a B-52?
[21:09] <cbreak> last year the us bombed a wedding because of fireworks
[21:10] <toad_> Zxcvb: not very effective against rifles though
[21:10] <Zxcvb> toad_: the militants might have forged the order
[21:10] <Zxcvb> toad_: I doubt they have that kind of hardware
[21:10] <toad_> if that was the case i'd expect them to have claimed that - but yes, it's possible
[21:10] <Zxcvb> toad_: mabye they don't want people to know the militants have that kind of hardware
[21:10] <toad_> Zxcvb: AK47s, whatever the american equivalent is
[21:11] <Zxcvb> toad_: M16A2, I think
[21:11] * toad_ nods
[21:11] <Zxcvb> toad_: they might be up to A3
[21:11] <toad_> so what did you say above? that you doubt the militants have the hardware to forge an airstrike order?
[21:11] <Zxcvb> toad_: yes
[21:12] <toad_> if they have, they're being really smart about it
[21:12] <toad_> i mean use that much and it'll be noticed
[21:12] <toad_> you can do a lot of damage but people will figure it out
[21:12] <Zxcvb> toad_: though if militants could forge an order, the military might cover it up by making an "they were armed" excuse
[21:13] <Zxcvb> toad_: exactly, compare normal viruses to cruise viruses
[21:13] <toad_> yeah, i see what you're driving at
[21:13] <toad_> cruise viruses?
[21:13] <Zxcvb> toad_: a normal virus like klez spreads rapidly
[21:13] <toad_> ah yeah
[21:13] <toad_> coventry
[21:14] <Zxcvb> toad_: a cruise virus will only spread to specific machines within an organization and only when necessary to accomplish the mission (like stealing data, etc)
[21:14] <toad_> still there's a lot of troops out there - it's possible they'd get hold of something - but aren't they all coded with the owner's ID?
[21:14] <d-ArkAngel> unless it's faulty ;-)
[21:14] <Zxcvb> toad_: that can be bypassed
[21:14] <toad_> i.e. you can't just steal one?
[21:14] <Zxcvb> toad_: or changed
[21:15] <toad_> no PKI? great
[21:15] <d-ArkAngel> problem with milatary hardware is that it needs to be simple enough to be reliable.
[21:15] <toad_> yeah
[21:15] <Zxcvb> toad_: the only reason a nuke can't be fired without the launch codes is because anyone who tried to modify the nuke to launch without it would be shot before they could bypass the system
[21:16] <d-ArkAngel> you don't want the Java VM on it throwing exceptions mid flight ;-)
[21:16] <toad_> d-ArkAngel: sure
[21:16] <Zxcvb> the military uses symmetric encryption whenever it uses crypto
[21:16] <Zxcvb> DES/AES/etc
[21:16] <d-ArkAngel> I bet it uses asymetric as well.
[21:16] <cbreak> stupid military...
[21:16] <toad_> you could use signatures to prevent forgery of owner of a radio set to prevent them being stolen
[21:16] <toad_> I assumed they did
[21:16] <toad_> but of course it takes 20 years to procure these things
[21:17] <d-ArkAngel> that's to big a generalisation to just say they only use symetric
[21:17] <Zxcvb> there was a time where if you had a C-band dish, you could easily get live footage from UAV cameras in afghanistan
[21:17] <Zxcvb> they could encrypt it, but they left it unencrypted "for the benefit of our nato allies"
[21:17] <toad_> lol
[21:17] <Zxcvb> so anyone with a big dish could trivially get live UAV/UCAV footage of the war
[21:18] <d-ArkAngel> and because it was cheaper than encrypting it.
[21:18] <Zxcvb> and yes, that includes bin laden
[21:18] <toad_> nice
[21:18] <cbreak> as if he cared about afghanistan...
[21:18] <d-ArkAngel> except bin laden wasn't in the states with a C-band dish ;-)
[21:18] <d-ArkAngel> *probably*
[21:19] <Zxcvb> and now, we caused iraq to give all their chemical/biological weapons to syria/lybia/etc
[21:19] <d-ArkAngel> assuming that they had any...
[21:19] <Zxcvb> if we hadn't invaded, they would have stayed buried
[21:19] <cbreak> didn't the inspectors anounce they had none?
[21:19] <toad_> on a lighter note, http://amphibian.dyndns.org/simulations2-epass-0.025-800x25.png is updated, and the blue line looks really promising
[21:19] <Zxcvb> they probably did, considering the US gave it to them
[21:19] <toad_> well, I'm skeptical that iraq got rid of ALL of it
[21:19] <toad_> but that does seem to be what the ISG says
[21:20] <toad_> anyway syria already had chemical weapons
[21:20] <toad_> maybe bio
[21:20] <Zxcvb> toad_: but as long as they are buried, what does it matter?
[21:20] <Zxcvb> toad_: think asbestos, sometimes removal does more damage than just leaving it alone
[21:20] <toad_> yeah
[21:21] <d-ArkAngel> oh come on. The US governemt is capable of misplaceing whole divisions of troops and forgetting that it has them. They couldn't find MWD if they were all marked with a big pink neon sign ;-)
[21:21] <cbreak> The US didn't search. The UN did.
[21:21] <Zxcvb> d-ArkAngel: true, but as long as the weapons were buried, there isn't a problem
[21:21] <toad_> d-ArkAngel: many of the ISG are re-employed from the UN team
[21:21] <Zxcvb> d-ArkAngel: but by attacking iraq, the weapons were shipped off to even worse people
[21:21] <d-ArkAngel> no, as long as sadam had them they were no problem
[21:22] <d-ArkAngel> now they're all forgotton burried under the sand
[21:22] <d-ArkAngel> some bastard digs them up and there's a whole world of hurt to be had
[21:22] <toad_> worse than saddam? i doubt it
[21:22] <cbreak> Asume you are an evil dictator possessing WMD and attacked by a force much stronger than you. What would you do?
[21:22] <d-ArkAngel> sadam who's goal in life was power and wealth
[21:23] * toad_ nods cbreak - he obviously didn't have any deployable WMD
[21:23] <Zxcvb> d-ArkAngel: it is also likely that saddam shipped some of it to syria
[21:23] <d-ArkAngel> and some nutter who's goal it is to "reak holy vengency on the unbeliving infideals"
[21:23] <d-ArkAngel> who do you really want to have the big read button
[21:23] <Zxcvb> cbreak: the WMDs were probably buried, then shipped out
[21:24] <toad_> http://amphibian.dyndns.org/simulations2-epass-0.025-400x25.png - updated
[21:24] <d-ArkAngel> cbreak: sell them and retire
[21:24] <Zxcvb> cbreak: as for your question, you run like hell and try to rebuild later
[21:24] <toad_> as is http://amphibian.dyndns.org/simulations2-epass-0.025-200x25.png
[21:24] <Zxcvb> cbreak: sure he is an evil, power hungry dictator, but he isn't totally nuts
[21:24] <toad_> Zxcvb: except that he didn't run
[21:24] <d-ArkAngel> yes he did
[21:24] <d-ArkAngel> unfortunatly he ran and hid in a hole
[21:25] <Zxcvb> toad_: he was probably waiting for an opprotunity to get out of the country
[21:25] <toad_> yeah, probably
[21:25] <toad_> he's not stupid
[21:25] <Zxcvb> the worst dictator of them all spent quite a while in prison
[21:25] <d-ArkAngel> he was probably watching the TV broadcasts by his information minister and got caught unaware of the invasion
[21:26] <toad_> http://amphibian.dyndns.org/simulations2-epass-0.025-100x25-long.png
[21:26] <toad_> HTL 10 clearly doesn't work for 100 nodes
[21:26] <Zxcvb> you know what I am expecting to see?
[21:26] <toad_> hmm?
[21:26] <d-ArkAngel> that's more than .95 isn't it?
[21:26] <Zxcvb> a couple of terrorists with machine guns take over a school
[21:26] <toad_> the yanks give up on democracy?
[21:27] <toad_> i mean in iraq..
[21:27] <toad_> d-ArkAngel: brown isn't
[21:27] <toad_> it's clearly falling
[21:27] <Zxcvb> the kids jump the terroists by suprise and take them down
[21:27] <d-ArkAngel> it's looking quite level after 1e07 to me
[21:27] <Zxcvb> the kids get expelled under the zero tolerance policy for violence, and are told "you should have let the terrorists kill everyone, zero tolerance is zero tolerance"
[21:28] <d-ArkAngel> ignoreing the horible randomness of corse ;-)
[21:28] <cbreak> Zxcvb: Terrorist have lately taken over a school.
[21:28] <cbreak> wasn't a happy ending at all.
[21:28] <Zxcvb> cbreak: yes, but not in a way that the kids could take it back
[21:28] <toad_> lol
[21:28] <d-ArkAngel> I love the russian milatary
[21:28] <Zxcvb> cbreak: I am refering to the stupidity
[21:29] <d-ArkAngel> Thy don't fuck about.
[21:29] <Zxcvb> cbreak: specifically, expelling students for fighting and winning against terrorists
[21:29] <cbreak> at least they didn't kill most of the hostages like they did in that opera last year.
[21:29] <d-ArkAngel> They said they'd learned from the last time and wouldn't be using force.
[21:29] <toad_> cbreak: they didn't then
[21:29] <d-ArkAngel> so their first move was to park a couple of tanks oustide the school
[21:29] <cbreak> the military did.
[21:29] <Zxcvb> actually, if I were in that school, I would be tempted to detonate the explosives myself
[21:29] <cbreak> used some poisonous gas...
[21:29] <toad_> cbreak: the theatre siege they only killed 1/6th of them or something like that
[21:30] <d-ArkAngel> toad: that's why I love them
[21:30] <Zxcvb> after all, if you are going to die, why not take the bastards with you
[21:30] <d-ArkAngel> they don't wait for things to happen on the bad guys terms.
[21:31] <d-ArkAngel> the west "doesn't negotiate with hostages" but the East don't even bloody talk to them
[21:31] <Zxcvb> d-ArkAngel: you do know that several american pilots have said they would crash the airplane themselves before allowing terrorists to enter the cockpit
[21:31] <d-ArkAngel> exactly
[21:31] <d-ArkAngel> they say that.
[21:32] <cbreak> let's hope no one makes jokes about being a terrorist...
[21:32] <d-ArkAngel> I didn't see any of the sept 11th'd planes crashing into the sea....
[21:32] <Zxcvb> d-ArkAngel: and kids in a school that is taken over like that should mess with the explosives to detonate them
[21:32] <cbreak> two crashed into towers.
[21:32] <Zxcvb> d-ArkAngel: that was said after 9/11
[21:32] <cbreak> Zxcvb: Stupid Idea.
[21:32] <Zxcvb> cbreak: why not?
[21:33] <Zxcvb> cbreak: it gives a good statement, "we won't tolerate terrorists, period"
[21:33] <cbreak> Because you lose possibilities.
[21:33] * Redb3ard (~oylerj@c-24-125-12-101.va.client2.attbi.com) Quit (Read error: 113 (No route to host))
[21:33] <d-ArkAngel> if you get taken hostage, just sit back and let the guys with guns come and kill them.
[21:33] <Zxcvb> cbreak: but the terrorists would know they have zero chance
[21:33] <toad_> d-ArkAngel: then what happened to the fourth plane?
[21:33] <cbreak> Allways act in a way to retain as many possible decisions.
[21:33] <d-ArkAngel> that was the passengers fighting back, and the terrorists losting controls, as I heard it
[21:34] <Zxcvb> cbreak: there would be no point in taking over a school because they would know they would be blown up automatically
[21:34] <toad_> Zxcvb: eh?
[21:34] <cbreak> Zxcvb: A lot of children survived.
[21:34] <toad_> a lot didn't
[21:34] <cbreak> So your statement somehow seems flawed.
[21:34] <toad_> i don't know what the proportions were
[21:34] <Zxcvb> cbreak: but what if the terrorists knew the childeren were willing to kill themselves to take out the terrorists?
[21:34] <d-ArkAngel> like in russia. they took over a Theater. And the milatary killed them all. They took over a school, the milatary played soft.... and went in and killed them all.
[21:35] <Zxcvb> d-ArkAngel: the military in russia should have used sarin in the theater
[21:35] <Zxcvb> d-ArkAngel: 100% guaranteed death
[21:35] <d-ArkAngel> why? what they did worked ok
[21:35] <cbreak> Zxcvb is obviously an idiot.
[21:35] <toad_> eh?
[21:35] <cbreak> Sarin sucks in many ways.
[21:35] <toad_> well 100% death is the main disadvantage :)
[21:36] <toad_> also they'd have had time to blow the whole place so you lose the building too :)
[21:36] <cbreak> It's just not state of the art anymore.
[21:36] <d-ArkAngel> death is always a dissadvantage ;-)
[21:36] <Zxcvb> toad_ also an advantage, as terrorists would know they can't win
[21:36] <toad_> Zxcvb: ah, so you're advocating shoot first, ask questions after, in hostage situations?
[21:36] <cbreak> Zxcvb: As if a terrorist cares about his death.
[21:36] <d-ArkAngel> in the theater they were right
[21:36] <Zxcvb> how many terrorists would take over and wire up a school if they knew the first thing a student would do is detonate the explosives
[21:36] <d-ArkAngel> the school should have been done the same way
[21:37] <toad_> d-ArkAngel: no way, not with all the kids
[21:37] <d-ArkAngel> kids is worst than old people?
[21:37] <toad_> d-ArkAngel: they'd have killed a LOT of them - and mostly the younger ones
[21:37] <d-ArkAngel> as appose to what happened?
[21:37] * jay (jcl@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[21:37] <toad_> d-ArkAngel: what proportion got out alive?
[21:37] <Zxcvb> d-ArkAngel: how many terrorists would take over and wire up a school if they knew the first thing a student would do is detonate the explosives
[21:37] <toad_> anyone got any hard facts?
[21:38] <toad_> Zxcvb: would you?
[21:38] <d-ArkAngel> someone probably has somewhere :-)
[21:38] <toad_> just to spite them?
[21:38] <toad_> I wouldn't
[21:38] <jay> uh oh we're off topic...
[21:38] <Zxcvb> there are no hard facts because those facts are a state secret
[21:38] <cbreak> jay: yes.
[21:38] <d-ArkAngel> there's pages and pages of off topic
[21:38] <toad_> I don't have a problem with my own instantaneous death really but I wouldn't pull the switch and kill over a thousand people just to spite some terrorists
[21:39] <d-ArkAngel> I chose to blame Zxcvb for it all with his conversations about PlayStations ;-)
[21:39] <toad_> nor even to deter future attacks
[21:39] <toad_> now, does anyone like my simulations? :)
[21:39] <d-ArkAngel> AMD?
[21:39] <d-ArkAngel> they get sales of their processors to run your sims on :-)
[21:40] <toad_> lol
[21:40] <toad_> most of them are on AMD processors yes :)
[21:40] <toad_> brother has just discovered that buying that server DP opteron board with no AGP slot wasn't such a great idea
[21:40] <toad_> but it's good for me since i'll get his graphics card.. and I can run simulations on it when he's not here
[21:41] <cbreak> why? AGP is old anyway :)
[21:41] <Zxcvb> AMD, as in explode into flames if the fan/heatsink fails AMD?
[21:41] <d-ArkAngel> once I get moved into my new house, depending on where I stand financialy I'll buy another Dual Oppy setup for you to play with (and an nvidia graphics card ;-) )
[21:41] <toad_> cbreak: overwhelming majority of AMD mobos don't have PCI Express yet
[21:41] <Zxcvb> nvidia sucks
[21:41] <toad_> this one only has PCI-X (64/133)
[21:41] <toad_> nvidia has better linux drivers than ati
[21:41] <toad_> unfortunately
[21:41] <d-ArkAngel> and ATI have no linux drivers.
[21:41] <Zxcvb> ati stinks too
[21:42] <toad_> Zxvcb: not with modern AMD chips and approved motherboards - even if you remove the CPU fan while it is in operation, it'll just shut down
[21:42] <d-ArkAngel> Tho I found some interesting roumours about ATI
[21:42] <toad_> I know this works because I've had a number of machines shut down spontaneously due to overheating :)
[21:42] <Zxcvb> agp, pci-x, pci-extreme all suck
[21:42] <d-ArkAngel> and P4EE's are far worse for heat generation ;-)
[21:42] <toad_> although I've never tried removing the fan on a live system
[21:42] <Zxcvb> XIO is cool
[21:42] <toad_> d-ArkAngel: "far worse" is a bit strong for a difference of maybe 12 watts
[21:43] <Zxcvb> toad_: ever hear of xio?
[21:43] <toad_> Zxcvb: vaguely
[21:43] <Zxcvb> toad_: how does using system memory as video memory with no performance hit sound?
[21:43] <toad_> Zxcvb: hypertransport always sounded rather nice but it's limited to stuff on the mobo
[21:43] <d-ArkAngel> check it out. 3D benchmarks by anandtech in linux! :-) http://www.anandtech.com/linux/
[21:43] <toad_> Zxcvb: eh?
[21:44] <toad_> Zxcvb: oh I see
[21:44] <d-ArkAngel> Zxcvb: it sounds like another dream to me.
[21:44] <toad_> it depends how fast your memory is :)
[21:44] <Zxcvb> toad_: 1.6GB/sec/pipe, 4 raster managers, 2.5GB dedicated frame buffer per pipe and and 1GB dedicated texture memory per pipe, 48-bit RGBA for up to 68 billion colors,
[21:44] <d-ArkAngel> There will be a perfromance hit, because you're sharing ram with your CPU.
[21:44] <Zxcvb> d-ArkAngel: it exists now
[21:44] <Zxcvb> d-ArkAngel: http://www.sgi.com/products/visualization/onyx3000/ir4.html
[21:44] <toad_> if you connect the graphics card directly to the hypertransport bus, that'd work :)
[21:44] <Zxcvb> d-ArkAngel: too bad the video card alone costs over 100 grand
[21:45] <cbreak> Zxcvb: And you have access to the electron beams :)
[21:45] <toad_> SGI is getting out of that business
[21:45] <Zxcvb> toad_: too bad, that card kicks ass
[21:45] <toad_> and in terms of fill rate and polys, the current $500 cards probably beat it anyway
[21:45] <d-ArkAngel> it's like having RAM on each CPU, you COULD have it shared, but it's FASTER to have ram on each node.
[21:45] <toad_> of course they don't have the same CPU-card bandwidth yet
[21:45] <Zxcvb> toad_: unfortunately, it costs 100 grand and is the size of a computer motherboard
[21:46] <toad_> 6.4GB/sec main RAM is nothing to compare with 20-40+ GB/sec graphics card RAM
[21:46] <d-ArkAngel> as long as you can move stuff over to the Graphics memory as fast as you need it there.... then you need nothing more.
[21:46] <toad_> and some of the high end cards have half a gig
[21:46] <toad_> next year's ones will have a gig
[21:47] <Zxcvb> toad_: try running this off a $500 card http://www.ncsa.uiuc.edu/VR/cavernus
[21:47] <Zxcvb> SGI's cards are more of a hardware 3d rendering device than a video card
[21:47] <d-ArkAngel> Try running toads sim on a 386 ;-)
[21:47] <Zxcvb> you can't render final fantasy, the spirits within with a standard video card because a video card outputs to a monitor
[21:47] <toad_> uh, what am I looking at?
[21:48] <toad_> Zxcvb: not true
[21:48] <d-ArkAngel> there is always "More" avaliable, bit there's limits at what is reasonable.
[21:48] <Zxcvb> toad_: think holodeck (nothing is solid of course)
[21:48] <toad_> you can render to a texture
[21:48] <toad_> since many years
[21:48] <toad_> you can read that texture back if you want
[21:48] <toad_> its just really slow on AGP or PCI because they suck
[21:49] <toad_> PCI Express fixes that
[21:49] <Zxcvb> toad_: sure, you could have a device that captures each frame as it is rendered (it can take a few minutes) and play it back
[21:49] <toad_> Zxcvb: no, it's not a device
[21:49] <toad_> it's built into the card you can do it in software
[21:49] <toad_> that's how you stripe over two graphics cards
[21:49] <Zxcvb> toad_: how do you get the rendered frame/image back to the hard drive though?
[21:49] <cbreak> Zxcvb: glReadPixels()
[21:49] <toad_> Zxcvb: opengl has supported it for MANY years... opengl is fully supported by modern cards
[21:50] <toad_> i have no idea re directx, directx sucks :)
[21:50] <d-ArkAngel> Amen
[21:50] <Zxcvb> so the PC finally gets a stripped down and slower version of what SGI had for years
[21:50] <d-ArkAngel> finally?
[21:51] <Zxcvb> yeah, without pci-express, you might as well do all rendering in software than use glReadPixels
[21:51] <d-ArkAngel> umm, it's been on PC's for a number of years
[21:51] <toad_> Zxcvb: faster on many things, and much cheaper... less CPU to graphics card bandwidth of course
[21:51] <toad_> although didn't you say 1.6GB/sec/pipe?
[21:51] <toad_> that's the bandwidth to the CPU?
[21:52] <Zxcvb> that is the XIO interface bandwidth
[21:52] <toad_> AGP8X is 2GB/sec... PCI Express is 4GB/sec
[21:52] <toad_> both ways
[21:52] <d-ArkAngel> yeah, but you don't need to spend the GNP of a small country to get PCI-E
[21:52] <Zxcvb> I couldn't find out how many pipes, which is the big question
[21:52] <toad_> yeah, I'm sure it's far more scalable
[21:52] <d-ArkAngel> If you've got a couple of million to spend on a GFX card, then Toad could really use a super computer to borrow.
[21:52] <toad_> I've heard things about striping over maybe 4 graphics cards with z-buffer compositing
[21:52] <toad_> using PCI Express
[21:53] <Zxcvb> does pci express support more than one card?
[21:53] <toad_> yup
[21:53] <Zxcvb> AGP doesn't for some reason
[21:53] <toad_> you need a special motherboard though
[21:53] <d-ArkAngel> AGP supporst as many as you want
[21:53] <toad_> they're not widely available yet
[21:53] <Zxcvb> what about interrupts
[21:53] <d-ArkAngel> you just need a northbridge with two AGP controlers
[21:53] <toad_> d-ArkAngel: I don't think so, AGP comes off the GART on the CPU
[21:53] <cbreak> you can put as many cards in a pc as you have slots... :)
[21:53] <Zxcvb> irq sharing between 5-6 cards will probably be slow
[21:53] <toad_> or on the northbridge yeah
[21:54] <toad_> Zxcvb: no knowledge, sorry
[21:54] <d-ArkAngel> well I've got two CPU's :-P
[21:54] <toad_> but I know people use these things for some pretty crazy stuff
[21:54] <Zxcvb> but they have to figure out which card to use when interrupt 10 is called
[21:54] <Zxcvb> does AMD64 have extra interrupts, or is it limited to the x86 number?
[21:54] <toad_> Zxcvb: won't it get allocated one dynamically by PCI?
[21:55] <toad_> I heard pentium pro supported high interrupts
[21:55] <toad_> up to 256
[21:55] <Zxcvb> toad_: yes, and use IRQ sharing, which is bad
[21:55] <toad_> I've never actually seen that though
[21:55] <d-ArkAngel> I think that IRQ shareing hasn't been a problem for quite a long time now.
[21:55] <Zxcvb> toad_: it takes extra work
[21:55] <Zxcvb> d-ArkAngel: and you can only use one card at a time
[21:55] <d-ArkAngel> besides, if you're addressing all the cards as a single unit, why would they need to be on different interrupts?
[21:56] <toad_> well, PCI Express isn't just for graphics cards, it's for all cards, it just comes on different widths
[21:56] <toad_> but I have no clue about how the interrupts work
[21:56] <d-ArkAngel> one CPU, one time, one instruction. that's how life works
[21:56] <cbreak> You could use different graphic cards at the same time since years. whats the problem?
[21:56] <Zxcvb> toad_: will PCI express allow 10 gigabit ethernet?
[21:56] <d-ArkAngel> yes
[21:56] <toad_> Zxcvb: it has the bandwidth for it
[21:56] <Zxcvb> d-ArkAngel: it's the difference between IDE and SCSI
[21:56] <toad_> I don't know what the current controllers use
[21:57] <Zxcvb> d-ArkAngel: with SCSI, all drives can be written to, with IDE if you are using the master, the slave is disabled
[21:57] <Zxcvb> d-ArkAngel: that is why your cd burner needs an IDE channel to itself
[21:57] <d-ArkAngel> and yet you can still have more than one IDE Channel...
[21:57] <d-ArkAngel> and write to them all as and when...
[21:57] <toad_> Zxcvb: as far as I know, PCI Express treats any card the same as any other card, and it's quite possible to talk to many cards efficiently, as is done in some larger servers
[21:57] <toad_> indeed, hence serial ATA :)
[21:57] <toad_> and even SAS
[21:58] <Zxcvb> toad_: how? do you not use the interrupt based interface?
[21:58] <toad_> which I am told by unreliable sources will be standard on mid range motherboards in a couple years
[21:58] <Zxcvb> heard anything about nanotube ram?
[21:58] <toad_> Zorix: no idea, although i've seen stuff about interrupt alternatives in the kernel
[21:59] <Zxcvb> a 10TB "hard drive" that also acts as system ram will be nice
[21:59] <d-ArkAngel> yeah I think it was on the back of the memo about the PS2's without OS's ;-)
[21:59] <toad_> absolutely :)
[21:59] * toad_ is looking forward to spintronics most though
[21:59] <Zxcvb> I wonder, can OSes like linux handle systems where ram and storage are the same thing?
[21:59] <d-ArkAngel> yes
[21:59] <toad_> zero power consumption, build them in 3D, no scaling-power problems so you can just use a zillion gates
[22:00] <d-ArkAngel> since to *nix everything is a file.
[22:00] <toad_> Zxcvb: like my Zaurus?
[22:00] <toad_> ;)
[22:00] <d-ArkAngel> and mine ;-)
[22:00] <Zxcvb> toad_: no, there is a difference, the zaurus has flash and ram
[22:00] <toad_> d-ArkAngel: does yours work (with OZ) ? :)
[22:00] <toad_> Zxcvb: the flash is optional
[22:00] <Zxcvb> I mean like a palm, where everything is sram, no flash
[22:00] <d-ArkAngel> last time I used
[22:00] <d-ArkAngel> it
[22:00] <toad_> well i suppose the 32MB internal isn't, but you can't use it
[22:00] <toad_> ah
[22:01] <toad_> yeah, linux gets used on embedded stuff with wierd memory issues
[22:01] <toad_> you just partition it
[22:01] <Zxcvb> openzaurus uses flash to store the base OS and bootloader
[22:01] <toad_> i believe there's a palmpilot port
[22:01] <d-ArkAngel> there is no reason you couldn't mount the ROM as R/W and dissable the ram on the zaurus
[22:01] <Zxcvb> toad_: no, no partitions, no flash, just ram
[22:01] <cbreak> d-ArkAngel: ROM can't be RW
[22:01] <Zxcvb> cbreak: flash can
[22:01] <d-ArkAngel> it's EEPROM
[22:01] <cbreak> ROM is READ ONLY
[22:01] <toad_> Zxcvb: of course you have to partition it if you want to put filesystems on it :)
[22:02] <Zxcvb> d-ArkAngel: the zaurus uses flash, and you can write to it
[22:02] <toad_> unless you want to use tmpfs for your root :)
[22:02] <cbreak> you can erease it, but not write to it.
[22:02] <Zxcvb> d-ArkAngel: at least the 5500 does
[22:02] <d-ArkAngel> yes I've got one, I knoe
[22:02] <toad_> now would somebody please tell me how any of this relates to freenet? :)
[22:03] <d-ArkAngel> and yes there is ROM it as well as flash (where exactly do you think the code that programs the flash is?)
[22:03] <d-ArkAngel> toad: beauce we're the freenet developers, and it's us doing the talking ;-)
[22:03] <toad_> ahhh
[22:03] <d-ArkAngel> well you're the develoers, and we're the hangers on at least ;-)
[22:03] <toad_> heh
[22:03] <toad_> d-ArkAngel: you can be a dev!
[22:03] <Zxcvb> d-ArkAngel: yes, a very tiny bit of ROM to handle flashing the ram
[22:03] <toad_> just get the parallel sim working
[22:04] <d-ArkAngel> No, I keep promising to be a developer, till I make something that works, I'm still officaly a hanger on ;-)
[22:04] <toad_> heh
[22:05] <Zxcvb> the current busses will be too slow for nanotube ram anyway, it is supposed to be as fast as the sram on CPU cache
[22:05] * toad_ hasn't actually heard about it
[22:05] <toad_> you got a URL?
[22:05] <cbreak> has freenet benefited from the sims? Or is all we have a bunch of diagrams?
[22:05] <toad_> cbreak: yes
[22:06] <toad_> cbreak: however even if they only validated what we are doing they'd still be beneficial IMNSHO
[22:06] <cbreak> :)
[22:06] <d-ArkAngel> it's proof we can use at our trial for crimes against the RIAA
[22:07] <cbreak> It sounds important to know the project is not driving against a wall :)
[22:07] <toad_> d-ArkAngel: you're getting a bit ahead of the game. first we have to build something that works. THEN we get persecuted by the whole wide world!
[22:07] <d-ArkAngel> maybe we are, but the wall will fall over eventually, and then it'll be clear driving (at least as far as the next wall)
[22:08] <toad_> one thing that seems to be coming out from the simulations is that we need much higher HTLs for networks of a practical size...
[22:08] <toad_> and we're not even simulating load yet...
[22:08] <cbreak> so how is freenet working as well with htl 20?
[22:08] <toad_> unfortunately HTL 50 is a huge waste...
[22:08] <toad_> well the simulations target 95% success...
[22:08] <d-ArkAngel> toad: exactly, we're just proving it works, so we've got written proof that we did it for our trial. Don't want them to have to spend money collecting evidence as well as making money overpriceing things to the point people won't buy it;
[22:08] <toad_> and they do it in a fairly dynamic content environment
[22:08] <cbreak> yes, right.
[22:08] <toad_> so it's basically push/pull tests
[22:09] <Zxcvb> d-ArkAngel: at least the new induce act draft doesn't affect free software
[22:09] <toad_> Zxcvb: it doesn't?
[22:09] <Zxcvb> d-ArkAngel: now, a majority of revenues must come from covered infringement
[22:09] <cbreak> isn't induce US only?
[22:09] <toad_> cool
[22:09] <Zxcvb> toad_: nope, they have to prove a majority of revenues came from covered infringement
[22:09] <toad_> except that uhmm, we have revenues!
[22:09] <Zxcvb> toad_: it will still kill most propriatery p2p apps
[22:09] <Zxcvb> toad_: yeah, from anonymous donations
[22:10] <toad_> not particularly anonymous donations actually
[22:10] <toad_> we just don't trumpet them
[22:10] <toad_> but there are no practical anonymous payment systems at the moment as far as i know
[22:10] <Zxcvb> toad_: "prove that most people donating aren't donating because they think it is a cool app and want to prevent the world from becoming one big fascist state"
[22:10] <toad_> yodel may still be around, but that's not practical, and the anonymity is a matter of trust
[22:11] <toad_> Zxcvb: easy at the moment, but if it ever performs well, it won't be easy
[22:11] <Zxcvb> toad_: freenet revenues are not tied to usage
[22:11] <toad_> well, ian and I aren't in the US
[22:11] <toad_> the bank account is iirc
[22:11] <Zxcvb> toad_: good point
[22:11] <toad_> and the paypal account
[22:12] <Zxcvb> besides, they could always pass the sssca again
[22:12] <toad_> and the 501c3...
[22:12] <d-ArkAngel> paypal isn't technicaly a back tho ;-)
[22:12] <toad_> well yes paypal rolls over at any opportunity
[22:12] <toad_> unfortunately nobody seems to want to use anything else to send us money
[22:12] <Zxcvb> yeah, just wait until paypal decides to withdraw all the funds in the freenet account as a "fine"
[22:13] <toad_> not amazon honour system, not egold (mostly), not anything
[22:13] <Zxcvb> toad_: have two accounts, a main account and a paypal recieving account
[22:13] <toad_> Zxcvb: not sure if they do that
[22:13] <cbreak> the microsoft effect.
[22:13] <Zxcvb> toad_: yes, they do
[22:13] <d-ArkAngel> I got my paypal account taken off me.
[22:13] <d-ArkAngel> and all the money confiscated
[22:13] <Zxcvb> toad_: yes, paypal does that I mean
[22:13] <toad_> Zxcvb: in order to receive significant credit card payments we have to pay a percentage on transactions
[22:13] <Zxcvb> d-ArkAngel: at least you didn't have all the money in your checking account confiscated too
[22:13] <toad_> if we had two accounts we'd have to pay it twice :|
[22:14] <d-ArkAngel> they told me it was because I was "Selling illegal hacking information"
[22:14] <toad_> well they froze our account once
[22:14] <toad_> but we got it back
[22:14] <Zxcvb> toad_: no, you have two checking accounts
[22:14] <toad_> ah i see
[22:14] <Zxcvb> toad_: the first is for paypal only
[22:14] <toad_> well then how do we move the money?
[22:14] <Zxcvb> toad_: the second is for anything other than payment services
[22:14] <toad_> money goes into paypal. no fee to withdraw but it takes f*cking ages
[22:14] <Zxcvb> toad_: the first is for payment services only
[22:14] <d-ArkAngel> it turned out to be because someone had put a linkt to donate money to me on a website next to something I had written
[22:14] <toad_> then we have to move the money to my bank which is in another country
[22:15] <toad_> the easiest and one of the cheapest ways to do this is via paypal... :|
[22:15] <Zxcvb> toad_: paypal is linked to a bank account
[22:15] <d-ArkAngel> but the 6 weeks of automated replies it took for me to get anywhere explaining this to them, they decided was too long for me to have brough it up. and so the terminated my account and I'm permenantly blacklisted
[22:15] <toad_> d-ArkAngel: nice
[22:16] <Zxcvb> toad_: transfer from paypal to that bank account, go to the bank's website, and transfer from the paypal recieving account to the normal one
[22:16] <toad_> so how come nobody ever uses any other payment system?
[22:16] <d-ArkAngel> oh and they decided to keep the ?200 I had in my paypal account as well
[22:16] <Zxcvb> d-ArkAngel: I think paypal just does that at random to help boost their revenues
[22:16] <toad_> Zxcvb: we'd end up paying two lots of fees - why not just transfer directly from project paypal to my paypal?
[22:16] <toad_> lol
[22:16] <toad_> possibly
[22:17] <Zxcvb> toad_: lots of fees?
[22:17] <toad_> Zxcvb: also inconvenience
[22:17] <d-ArkAngel> Personally I don't see why some of the big banks don't setup their own system
[22:17] <Zxcvb> toad_: two accounts at the same bank
[22:17] <toad_> no fees for moving between bank accounts
[22:18] <Zxcvb> toad_: there is no transfer fee
[22:18] <toad_> fees for user-paypal->paypal->bank->paypal->paypal->bank
[22:18] * cbreak (~cbreak@dhcp-32-094.via-eth.ch) Quit ("...")
[22:18] <toad_> Zxcvb: yes there is
[22:18] <toad_> on paypal
[22:18] <toad_> if you have a premier account
[22:18] <toad_> which at least the project account would have to be
[22:19] <Zxcvb> toad_: no, user-paypal->freenet paypal->freenet bank/checking account->main checking account
[22:19] <toad_> anyway, right now, we don't have easy access to the bank account - or at least we didn't - last time I checked, we can draw funds but not check the balance without getting hold of Steven Starr
[22:19] <toad_> Zxcvb: then ->paypal->my_paypal->my_bank_account
[22:20] <toad_> but I don't deal directly with this stuff, ian does
[22:20] <Zxcvb> toad_: no
[22:20] <Zxcvb> toad_: both checking accounts are at the same bank
[22:20] <toad_> yes, and?
[22:20] <toad_> what about my account?
[22:21] <toad_> me personally
[22:21] <Zxcvb> toad_: the idea is to shield your account from paypal
[22:21] <toad_> as opposed to FPI
[22:21] <toad_> Zxcvb: how do you propose to move money across the pond?
[22:21] <Zxcvb> toad_: I would assume FPI writes a check
[22:21] <toad_> anyway it doesn't matter
[22:21] <toad_> if paypal suspends us our revenue stream dries up
[22:21] <toad_> publicity or no publicity, NOBODY uses anything except for paypal
[22:21] <Zxcvb> toad_: what if paypal also raids your bank accounts?
[22:21] * pupok (~r00t@81-178-76-180.dsl.pipex.com) Quit (Read error: 232 (Connection reset by peer))
[22:21] <toad_> that's our experience
[22:22] * pupok (~r00t@81-178-76-180.dsl.pipex.com) has joined #freenet
[22:22] <toad_> Zxcvb: international money transfer is a PITA and quite expensive
[22:22] <Zxcvb> toad_: so you can no longer pay any bills because paypal took all the money in your bank account
[22:22] <d-ArkAngel> what about verefied CC stuff?
[22:22] <toad_> yeah, we could get a proper CC server
[22:22] <toad_> but it'd cost significantly
[22:22] <d-ArkAngel> or are the fee's too much?
[22:22] <Zxcvb> the fees are around 2%, I think
[22:22] <toad_> paypal does not have the legal authority to raid my bank account
[22:23] <toad_> because they are not the bank and there is no legal signed contract between us allowing them to
[22:23] <d-ArkAngel> yeah, I suffer $600 a month hosting costs, so far I've managed to raise... about $40 thought means other than paypal.
[22:23] <Zxcvb> toad_: yes they do, check the EULA/TOS on fines
[22:23] <toad_> Zxcvb: no they don't
[22:23] <toad_> no contract
[22:23] <toad_> UCITA only passed in a few states
[22:23] <Zxcvb> toad_: what about the terms of service?
[22:23] <toad_> and there's no UK equivalent IIRC
[22:23] <toad_> NO CONTRACT
[22:23] <Zxcvb> toad_: no, the bnetd ruling makes it binding on yet another circuit
[22:23] <toad_> they have no right to my money unless I steal it from them
[22:24] <Zxcvb> toad_: true, it doesn't apply in the UK, but what can the UK do about it
[22:24] <toad_> Zxcvb: bnetd ruling?
[22:24] <d-ArkAngel> www.blizzhackers.com first news item ;-
[22:24] <toad_> in any case I don't see where in the T&C it says they can steal your money that is no longer in their system
[22:25] <Zxcvb> toad_: check out the part about fines
[22:25] <Zxcvb> toad_: besides, even though they don't have that right, how would a UK court punish paypal
[22:26] <d-ArkAngel> ummm, paypal have a UK corportate branch
[22:26] <toad_> okay, that's strange
[22:26] <d-ArkAngel> they fuck with the UK law, and they'll find some big ass fine eating their ass
[22:26] <toad_> Zxcvb: they don't have the right to impose arbitrary fines
[22:26] <toad_> there are controls on unreasonable contracts
[22:26] <toad_> they don't for instance have the right to reposses my parents' house
[22:27] <toad_> just because I have a paypal account
[22:27] <d-ArkAngel> yeah, like the small print can't say that they have the right to my imortal soul or anything like that
[22:27] <toad_> (and I believe mum does too)
[22:27] <Zxcvb> d-ArkAngel: ah, assuming they physically have assets in the UK that would work
[22:27] <Zxcvb> toad_: you are correct
[22:27] <d-ArkAngel> www.wewantyoursoul.com has first rights on that
[22:28] <Zxcvb> toad_: however, if paypal were to do it anyway and couldn't be sued in a US court, then in practice they can do that
[22:28] <Zxcvb> toad_: of course, if they have assets in the UK those can be siezed
[22:28] <toad_> why couldn't they be sued in a US court?
[22:29] <toad_> property law is pretty fundamental
[22:29] <toad_> even a megacorp can't just go stealing things
[22:29] <d-ArkAngel> same company, the UK can seize as much of it as they deem fair even if the UK office isn
[22:29] <d-ArkAngel> t worth as much as toad get's paid ;-)
[22:29] <toad_> sure they can impose their will, but that involves smoothing the way with cash :)
[22:29] <Zxcvb> toad_: they just claim they sent out a modified TOS to you, and since those controls on unreasonable contracts are virtually non-existant in the US, it will be hard
[22:30] <d-ArkAngel> or dead boddies if cash it too hard to come by ;-)
[22:30] <Zxcvb> toad_: and just how is the UK supposed to sieze all the US offices?
[22:30] <toad_> Zxcvb: well, first off, it seems rather implausible
[22:30] <toad_> secondly, we have virtually no assets in the US
[22:30] <d-ArkAngel> plus you're wrong about the update TOS
[22:30] <toad_> I can see them keeping the balance
[22:30] <d-ArkAngel> if they can't prove you've seen and aggreed to it it's not binding. the BnetD ruling even supports that position
[22:31] <toad_> which is a clear reason to not keep a major balance
[22:31] <Zxcvb> d-ArkAngel: isn't there a ruling the eula is hearsay unless on paper?
[22:31] <toad_> however, what you're suggesting means that anything I sign up to online can just email me a ToS update and take all my money
[22:32] <toad_> that would be anarchy
[22:32] <toad_> the economy would break down
[22:32] <d-ArkAngel> no, they validate clickthough aggreenemts
[22:32] <d-ArkAngel> paper isn't needed
[22:32] <Zxcvb> d-ArkAngel: yet the bnetd team didn't get it thrown out "i.e. prove that the EULA I was shown contained that term"
[22:32] <Zxcvb> d-ArkAngel: but blizzard couldn't prove the EULA actually contained that
[22:32] <d-ArkAngel> no, BnetD lost the case
[22:32] <Zxcvb> d-ArkAngel: right, because they didn't use that argument
[22:32] <toad_> d-ArkAngel: is the BnetD code on Freenet yet? ;)
[22:33] <d-ArkAngel> they argued that the EULA wansn't binding
[22:33] <d-ArkAngel> no, I'm working on that
[22:33] <Zxcvb> d-ArkAngel: microsoft used that argument, which is why they got a ruling saying they didn't have to refund because there was no proof of what the EULA says
[22:33] <toad_> d-ArkAngel: hey, don't admit it :)
[22:33] <d-ArkAngel> well I'm actually working on the sims
[22:33] <toad_> I suppose I shouldn't have asked really...
[22:33] <d-ArkAngel> and I've been thinking about working on it, but failing to actually work on it
[22:33] <Zxcvb> d-ArkAngel: right, if the bnetd team had said "the EULA I was presented with didn't say that, prove it", they would have won
[22:33] <d-ArkAngel> I'll probably mention the idea to someone else, and they'll work on it
[22:34] <toad_> there's Arch over Freenet too...
[22:34] <toad_> although that probably doesn't work well yet
[22:34] <d-ArkAngel> Zxcvb: blatent lies don't work in court
[22:34] <toad_> but anonymous development would be interesting
[22:34] <Zxcvb> d-ArkAngel: just because the code is there to display a certain EULA and the EULA.txt was dropped on the hard drive, doesn't mean that EULA was actually the one that was displayed
[22:34] <Zxcvb> d-ArkAngel: i.e. "prove there wasn't self modifying code to display a different EULA than the one blizzard says"
[22:34] <toad_> okay, I'm going to bed
[22:35] <toad_> see you all tomorrow
[22:35] <toad_> happy flamewar
[22:35] <d-ArkAngel> that's like arguing that just because there was a 3000 point DNA match doesn't mean that it really was them
[22:35] <Zxcvb> d-ArkAngel: microsoft won
[22:35] <Zxcvb> d-ArkAngel: the microsoft EULA that came with windows was ruled as hearsay
[22:35] <d-ArkAngel> there has to be a point at which you take something as proof
[22:35] <Zxcvb> d-ArkAngel: read up on the "windows refund day" case
[22:35] <toad_> eh? microsoft WANTED the EULA to be shown to be invalid? WTF?
[22:36] <d-ArkAngel> that's only because people had their coppies of windows pre-installed.
[22:36] <Zxcvb> toad_: yes, because microsoft would have had to provide a full refund for unused windows
[22:36] <toad_> ahhh
[22:36] <toad_> but doesn't that mean it's not binding on the customer?
[22:36] <Zxcvb> d-ArkAngel: if an EULA is hearsay because it is in software, then it is hearsay
[22:36] <Zxcvb> toad_: not unless the customer argues the eula is hearsay themselves
[22:36] <toad_> I mean doesn't it undermine your argument?
[22:36] <d-ArkAngel> no, it's hearsay if you can't prove that they had to have seen it to get into the current state
[22:37] <Zxcvb> toad_: in the bnetd case, what the EULA actually said wasn't challenged
[22:37] <toad_> okay so as far as paypal goes, it's binding because they can prove you have seen it? IF they make you go through a click-through?
[22:37] <d-ArkAngel> MS won with the argument that people might not have seen it because it might have been an OEM install, or that they might not have used the software
[22:37] <toad_> or is it binding just because they say "please read the T&C" with a link?
[22:37] <Zxcvb> toad_: or a 1x1 pixel gif
[22:38] <Zxcvb> toad_: make that 0x0 pixel gif
[22:38] <toad_> eh?
[22:38] <toad_> if you've never seen it it's definitely hearsay
[22:38] <Zxcvb> toad_: send you an email with a 0x0 pixel gif
[22:38] <d-ArkAngel> Zxcvb, you can use arguments like that, but it'll end up in front of a jury, and 12 randoms of the street normally know a pack of lies when they hear it.
[22:38] <toad_> there is no possible way I would have read the updated T&C if that's how they send it
[22:39] <toad_> therefore I haven't seen it
[22:39] <Zxcvb> toad_: such as 'img src=www.paypal.com/images/<hash of user info>/tracer.gif
[22:39] <Zxcvb> toad_: no, the 0x0 gif is sent along with the test of the updated T&C
[22:39] <toad_> Zxcvb: so that's to prove the email was displayed?
[22:39] <Zxcvb> toad_: yes
[22:39] <toad_> I use a text based email client
[22:40] <d-ArkAngel> and it wasn't in an e-mail, it was in the web form ;-)
[22:40] <toad_> but in any case don't I have a right to cancel and a 30 day grace period if they change the T&C outrageously?
[22:40] <Zxcvb> toad_: not sure
[22:40] <toad_> isn't that a legal requirement?
[22:40] <d-ArkAngel> the one with the checkbox labeled I aggree with the terms and conditions as shown above, or avaliable here..... etc etc
[22:40] <Zxcvb> toad_: not in the USA, I think
[22:40] <toad_> I can't see how the system would work AT ALL if it's as bad as you suggest
[22:40] <Zxcvb> toad_: there are a few specific exceptions
[22:41] <toad_> there are businesses that use paypal - very many businesses
[22:41] <d-ArkAngel> it's not leagal unless you either had to sign something, or you had to click a button labled I Agree
[22:41] <Zxcvb> toad_: so you only sieze individual accounts, and only from people you think won't complain
[22:42] <Zxcvb> d-ArkAngel: the loophole is that you agreed that updates could be sent to you via email, and that you agree to read all such emails
[22:42] <d-ArkAngel> if you click that button then you've aggreed to the terms. your lack of interest in reading them doesn't stand up as a defence
[22:42] <toad_> if paypal can steal their money that hasn't even been through paypal for no reason at any time, then a) we'd hear about it, and b) it'd be illegal
[22:42] <toad_> okay, so you are obliged to read the emails they send you containing the updated terms and conditions?
[22:42] <toad_> well that's a PITA but it's no great hardship
[22:42] <Zxcvb> toad_: you should have, they claim only for "pornography" and "gambling"
[22:43] <d-ArkAngel> you've aggreeded that they will make "all reasonable effort to keep you informed" and that your still protected by the "no unreasonable, or fundemental changes" laws
[22:43] <toad_> I can't see them sending you a customized T&C just to steal your money
[22:43] <Zxcvb> toad_: no, but they can claim to
[22:43] <Zxcvb> toad_: plausible deniability
[22:43] <Zxcvb> toad_: though when I have heard about it, paypal just steals it
[22:43] <toad_> I know they freeze people's accounts
[22:43] <Zxcvb> toad_: i.e. "come sue us, I dare you"
[22:43] <toad_> and run off with the cash
[22:44] <d-ArkAngel> and they joiked all my money :-(
[22:44] <Zxcvb> toad_: yes, and sometimes they withdraw from your account too
[22:44] <toad_> HOWEVER to take money from elsewhere i.e. bank accounts, THEY would have to sue YOU
[22:44] <Zxcvb> toad_: presumably they would avoid the UK
[22:44] <toad_> they have no property right to your money
[22:44] <toad_> they would have to use lawyers to get at it
[22:44] <toad_> that would inevitably end up in court
[22:44] <d-ArkAngel> if they took money from my CC I'd just get the CC card to reclaim it and tell them to fuck off.
[22:44] <d-ArkAngel> which they would
[22:44] <toad_> right
[22:44] <Zxcvb> toad_: no, they have the capability to take it from your account
[22:44] <d-ArkAngel> (banks don't like paypal awhole lot anyway
[22:45] <Zxcvb> toad_: to get it back, you would have to go to court
[22:45] <toad_> Zxcvb: technically yes but that would be THEFT or at least FRAUD
[22:45] <Zxcvb> toad_: or convince law enforcement to do so
[22:45] <toad_> which is a CRIMINAL not civil offence
[22:46] <d-ArkAngel> they can't take money from your account without your express permission
[22:46] <d-ArkAngel> it's that simple
[22:46] <Zxcvb> toad_: if you don't have the resources to sue, and law enforcement isn't interested, then you are out of luck
[22:46] <d-ArkAngel> they can bill you for it on a CC< but you don't have to pay that
[22:46] <toad_> and there are competent bodies to deal with this, specifically because any scam that can be done probably has been done at some point
[22:46] <toad_> and a working society depends on reasonable protection of property
[22:46] <Zxcvb> toad_: that is why some people in the US have had that problem
[22:47] <d-ArkAngel> if they take your money, you can get the legal costs back when you inevitably win
[22:47] <toad_> Zxcvb: their T&C allows them to impose unlimited fines?
[22:47] <d-ArkAngel> people have that problem because when it happens they don't complain enough
[22:47] <toad_> if not, they'd have to send you a custom T&C
[22:47] <toad_> if they do that that's clear evidence of fraud
[22:48] <Zxcvb> d-ArkAngel: they also have access to your checking account directly
[22:48] <toad_> but I dunno, maybe in the Land of the Free, only corporations have property rights
[22:48] <toad_> yes technically they have the capacity to draw cash
[22:48] <toad_> but they don't have the legal right
[22:48] <Zxcvb> toad_: previously they didn't bother with T&C
[22:48] <d-ArkAngel> Zxcvb: yeah, but they still need your permission to take the money
[22:49] <Zxcvb> toad_: law enforcement wasn't interested in prosecuting because the amount was too low, and suing requires large resources
[22:49] <toad_> Zxcvb: that which does not belong to you, does not belong to you. even if you are a corporation, you have to pay for it.
[22:49] <d-ArkAngel> and if they don't demonstratably have it then they can't have anything
[22:49] <toad_> Zxcvb: so bring a class action suit
[22:49] <d-ArkAngel> what's law enfoecment got to do with anything
[22:49] <toad_> Zxcvb: if anything was happening, it's one bored clerk somewhere in paypal sending out custom T&Cs and stealing a few bob on the side to pay for his holidays
[22:49] <toad_> but law enforcement are quite willing to prosecute a theft of $500
[22:50] <Zxcvb> toad_: in fact, if I steal $10 or even $50 from your back pocket, and you have very good proof, in the US there is little that you can do because the police won't be interested, and court fees would cost more than you win
[22:50] <d-ArkAngel> it's the banking ombitsman, and tradeing standard, and a big civil suit, for the money, and the costs, and the invonvinece, and the emotional distress....
[22:50] <Zxcvb> toad_: in the UK, a judge decides who pays lawyer fees, which is typically the winner unless someone won by technicality but should have lost based on the merits
[22:51] <toad_> Zxcvb: in a civil case that is true. in a criminal case the police have a statutory duty to deal with it. furthermore there are ombudsmen and so on.
[22:51] <Zxcvb> d-ArkAngel: they technically aren't a bank though
[22:51] <Zxcvb> d-ArkAngel: they exploit a loophole, so that doesn't work
[22:51] <Zxcvb> toad_: right, the police deal with it at their discretion
[22:51] <d-ArkAngel> then they're running a bank without a licence.
[22:51] <toad_> people get taken to court for trivial stuff all the time
[22:52] <Zxcvb> d-ArkAngel: they aren't a bank though
[22:52] <d-ArkAngel> there are at least two states in the US they're not allowed to operate by legal ruling.
[22:52] <Zxcvb> d-ArkAngel: they exploit a loophole in the laws regarding banks
[22:52] <KenMan> what are you all blabbering about ?
[22:52] <toad_> Zxcvb: which is?
[22:52] <Zxcvb> d-ArkAngel: some states don't have that loophole
[22:52] <toad_> KenMan: Zxcvb is trolling
[22:52] <d-ArkAngel> which is? they hold your money, and deal with transactions
[22:52] <d-ArkAngel> that's a Bank
[22:52] <toad_> he wants us to believe that paypal can take money out of our bank accounts without our permission and get away with it
[22:53] <Zxcvb> http://news.com.com/2100-1017-858264.html?legacy=cnet
[22:53] <KenMan> ahh, i think you have to piss them off first
[22:53] <d-ArkAngel> and that you can write code that turns a PS1 into a PS2
[22:53] <d-ArkAngel> I pissed them off, and they emtyed my paypal account and blacklisted me.
[22:53] <Zxcvb> the Federal Deposit Insurance Corporation said it does not consider the company to be a bank or savings association because it does not accept deposits as defined by federal law, which requires institutions to have a banking charter.
[22:53] <d-ArkAngel> but they couldn't touch my bank account ;-)
[22:54] <Zxcvb> d-ArkAngel: who knows, mabye they stopped because people were getting too pissed off
[22:54] <d-ArkAngel> and yet they've still been outlawed in two states in the US for operating a bank without a licence.
[22:54] <d-ArkAngel> and have applied for a banking licence from the EU
[22:54] <KenMan> maybe they stopped because too many *important* people got pissed off.
[22:55] <d-ArkAngel> yeah like me, I'm important ;-)
[22:55] <toad_> heh
[22:55] <d-ArkAngel> possibly because I told the 70,000 people subscribed to my site all about paypal being theiving bastards....
[22:55] <Zxcvb> d-ArkAngel: I am suprised paypal hasn't challenged it based on preemption
[22:55] <toad_> yes, I'm sure they get away with stealing your paypal balance sometimes
[22:55] <toad_> but taking money from your bank account? no way
[22:55] <d-ArkAngel> exactly
[22:55] <toad_> d-ArkAngel: LOL
[22:56] <toad_> d-ArkAngel: how do you get paid then, if you can't use paypal?
[22:56] <Zxcvb> toad_: so they use the excuse that the money was just transfered
[22:56] <toad_> there's like 20 other services, but nobody uses them for some strange reason
[22:56] <d-ArkAngel> I use a close personal friends paypal account ;-)
[22:56] <Zxcvb> toad_: so they have the right to take it back
[22:56] <toad_> I've always been of the opinion that we should take everything, but ian disagrees for marketing reasons...
[22:56] <Zxcvb> toad_: that still isn't unlimited though
[22:56] <d-ArkAngel> I setup e-gold, and all the rest.
[22:56] <toad_> d-ArkAngel: how much did you transfer over paypal in total?
[22:56] <d-ArkAngel> and got ignored
[22:57] <toad_> d-ArkAngel: did they try to take it back?
[22:57] <toad_> d-ArkAngel: yes, that's been our experience
[22:57] <toad_> I don't understand why
[22:57] <toad_> I can't see any plausible reason to trust paypal
[22:57] <toad_> I suppose it's just first-to-market
[22:57] <d-ArkAngel> none, I got ?200 ish in a week or so and logged in to xfer it out and they told me it was their and fuck off because I'm selling illegal hacking information.
[22:57] <Zxcvb> oddly enough, a judge ruled their arbitration rules were unfair
[22:58] <Zxcvb> but that was before bnetd
[22:58] <toad_> when you were in fact soliciting donations for a different purpose, and completely separately distributing illegal hacking info?
[22:58] <d-ArkAngel> what has BnetD got to do with anything?
[22:58] <d-ArkAngel> have you even read that case file?
[22:58] <toad_> anyway
[22:58] <toad_> if we get big enough for paypal to be threatened, we'll just get a proper secure server
[22:59] <toad_> people DO use THEM
[22:59] <d-ArkAngel> I was collecting donations from a website that happened to contain a forum, in which some other people were discussing hacking information hosted on other sites
[22:59] <d-ArkAngel> oh and the site had the word hacker in the title
[22:59] <toad_> yeah, we got suspended because of allegedly using or running a proxy service
[22:59] <toad_> but we had a worldwide fanbase to harass the CFO until they reinstated us :)
[22:59] <d-ArkAngel> which makes me evil
[22:59] <d-ArkAngel> does that work?
[22:59] <toad_> yup
[22:59] <Zxcvb> d-ArkAngel: yes, the bnetd case basically said "there are no unfair terms"
[23:00] <d-ArkAngel> should I get my 70,000 people to e-mail him?
[23:00] <toad_> nah
[23:00] <toad_> phone calls work better
[23:00] <d-ArkAngel> umm, no it didn't
[23:00] <Zxcvb> d-ArkAngel: that is, the bnetd case weakens "unfair terms" considerably by stating that most terms are fair
[23:00] <toad_> we had the contact details...
[23:00] <toad_> we can find them if you need them...
[23:01] <d-ArkAngel> the reason that didn't work is they said it was unreasonable for them to realise they were doing something in violation
[23:01] <Zxcvb> d-ArkAngel: I thought they claimed that the "no reverse engineering" part itself was unfair
[23:01] <d-ArkAngel> but since they had plastered warnings all over their site about it being a violation of the EULA and TOC to use ti....
[23:02] <Zxcvb> d-ArkAngel: that is, they were challenging the EULA validity itself
[23:02] <Zxcvb> d-ArkAngel: of course, it also made the dmca interoperability section useless, even if there is no EULA
[23:02] <toad_> http://cvs.sourceforge.net/viewcvs.py/freenet/website/pages/news.php?rev=1.71&view=markup
[23:02] <d-ArkAngel> they did, but that claim was found unjust because it was a reasobale exception as it's not preventing people making other games, it's prevening peole copying their games.
[23:02] <toad_> towards the bottom
[23:03] <Zxcvb> d-ArkAngel: the interoperability exemption was also removed, which is seperate
[23:03] <toad_> If you are concerned
[23:03] <toad_> about whether your account might be at risk due to your political opinions you may wish to
[23:03] <toad_> speak to their PR
[23:03] <toad_> contact Hani Durzy at (408) 376 7458. If you are an investor and you would like to see
[23:03] <toad_> what other political opinions Paypal doesn't like, you may want to speak to their
[23:03] <toad_> Investor contact Tracey Ford at
[23:03] <toad_> (408) 376 7205.
[23:03] <Zxcvb> d-ArkAngel: even if there is no EULA or contract, the DMCA interoperability exemption is narrowed to requiring permission first
[23:03] <toad_> It is fortunate that
[23:03] <toad_> Johann Gutenberg did not rely on Paypal to
[23:03] <toad_> fund his work on the printing press, which also allowed anonymous publication of
[23:03] <toad_> information, since his account would probably have been frozen too.
[23:04] <toad_> if they freeze us again we'll just get a proper secure server
[23:04] <Zxcvb> d-ArkAngel: remember, each defense is taken individually unless they are specifically combined
[23:04] <toad_> they are quite expensive though
[23:05] <Zxcvb> d-ArkAngel: I also wonder when MS will sue openoffice under that ruling
[23:06] <Zxcvb> d-ArkAngel: after all, the no RE in ms office doesn't prevent making your own office suite, only using the ms office file formats
[23:08] <d-ArkAngel> The Contract clause works because it's not Adding rights that are not granted under copyright law, and the users aggreeded to the terms and conditions.
[23:08] <d-ArkAngel> they chose to surrender their right to fair use and reverse engineering when they aggreed to the TOC/EUAL
[23:08] <toad_> Zxcvb: I thought M$ was just using patents to protect file formats now?
[23:08] <toad_> well not just
[23:08] <toad_> but regardless of the DMCA
[23:08] <d-ArkAngel> they could have returned the game for a complete refund had that been a problem for them to accept those ruleings.
[23:08] <toad_> and the EULA
[23:09] <toad_> what court was all this btw?
[23:09] <d-ArkAngel> you don't need office to understand the files format
[23:09] <d-ArkAngel> California
[23:09] <Zxcvb> d-ArkAngel: but can they prove they didn't use office
[23:09] <toad_> so just a state court?
[23:09] <d-ArkAngel> yes
[23:09] <toad_> state supreme or state circuit?
[23:10] <Zxcvb> d-ArkAngel: they just have to find one openoffice developer who uses or has access to ms office
[23:10] <toad_> well screw them
[23:10] <d-ArkAngel> proof is requred to show something did happen, not that it didn't
[23:10] <Zxcvb> d-ArkAngel: for the same reason all downloadable music is illegal
[23:10] <d-ArkAngel> what are you on aboue?
[23:10] <toad_> it's not a big surprise that california's courts are for sale really
[23:10] <Zxcvb> d-ArkAngel: they prove one person had access to ms office, then use the contamination argument
[23:11] <d-ArkAngel> no it's not
[23:11] <d-ArkAngel> they have to prove that he used that copy of offivce
[23:11] <Zxcvb> d-ArkAngel: all music is "substantially similar" because of the eight note standard
[23:11] <toad_> in any case, if it's a state ruling, it only affects california, right?
[23:11] <Zxcvb> toad_: in theory
[23:11] <d-ArkAngel> or that in some way they could not reasonable have extracted the information without the copy of office
[23:12] <d-ArkAngel> Zxcvb you're just talking bollocks now
[23:12] <Zxcvb> d-ArkAngel: check out bright tunes v. harrisongs
[23:12] <toad_> okay, have fun
[23:12] <Zxcvb> d-ArkAngel: the eight most signifigant notes are used to determine "substantially similar"
[23:12] <toad_> i really ought to go to bed now
[23:12] <d-ArkAngel> There is no law that says "all music is substantialy simmilar"
[23:12] <Zxcvb> d-ArkAngel: and access can be proven by the radio
[23:12] <Zxcvb> d-ArkAngel: no, there is precedent and math
[23:13] <Zxcvb> d-ArkAngel: find out how many possible combinations of eight notes there are, and compare to the music library size of the major labels
[23:13] <d-ArkAngel> and this has something magic to do with illigal file shareing does it?
[23:13] <d-ArkAngel> there are an infinitie number of combinations ;-)
[23:13] <Zxcvb> d-ArkAngel: basically, it means that all that "legal" music you get from indie musicians is illegal
[23:13] <d-ArkAngel> infact there are more than an ifinatie number of combinations
[23:14] <Zxcvb> d-ArkAngel: not when the standard is eight most signifigant notes
[23:14] <d-ArkAngel> look, if you're just going to make up crap like that I'm out
[23:14] <d-ArkAngel> c ya
[23:15] <Zxcvb> the odds of a random song not being similar under that standard were about 1 in 212, though the standard might change if the riaa used the precedent + math argument that they owned all music
[23:17] <d-ArkAngel> one last parting shot. if they do that they've shot themselves in the foot.
[23:18] <d-ArkAngel> pick any 212 songs that were written before they came on the sceene that they haven't inherited, and that's prior art right there.
[23:18] <d-ArkAngel> no more ownership
[23:38] * Zxcvb (~somebody@client-216-176-88-210.consolidated.net) Quit ("Leaving")
[23:54] * jay (jcl@ool-18bf6dac.dyn.optonline.net) Quit (Remote closed the connection)
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.
