From: jfmezei.spamnot@vaxination.ca   
      
   On 13-01-31 09:00, Some Guy wrote:   
      
   > Do you doubt that GardaLey can also do that?   
      
   For all I know, GardaLey can simply be monitory the list of peers that   
   the tracker feeds it.  Yes, this means it can see % of file that a peer   
   currently has. And yes, it can likely see whether a peer is accepting   
   connections or if it is "chocked".   
      
   If GardaLey advertises itself as being unchocked and accepting   
   connections, then downloaders can connect to it. Will GardaLey accept   
   the connection ? Will it actually upload content ? Does it log how much   
   content is uploaded to each pear ?   
      
   More importantly, does GardaLey actually request content from peers who   
   accept connections ? And does it download 100% of a file from 1 peer to   
   prove that peer has uploaded 100% ?   
      
   Since software can fake how much of a file it has, does the mere fact   
   that you see an IP with 100% of file in a tracker PROVE that this IP   
   address actually has 100% of the file ?   
      
   I suspect GardaLey itself fakes the percentage of the file that it has   
   in order to not show it has 100% (otherwise, it woudln't be able to   
   download from unsuspecting folks).   
      
   > For one thing, someone at Canipre would be incredibly stupid to lie to   
   > the court about how their software operates.   
      
   This isnt a question of liying. It is a question that their affidavit   
   omitted much information in order to (possibly) not show that their   
   collection technique is flawed and does not represent proof of download.   
      
   again, seeing a person in a bar/tracker is no proof that that person   
   performed an illegal act such as hiring a hooker.   
      
      
   > And another reason why their software *probably* doesn't fake the data   
   > is because the real data is easily obtainable without needing to fake   
   > it.   
      
   Finding over 4000 IP addresses who downloaded unknown/undesirable movies   
   in a period of a month ?   
      
   sorry, but before an ISP hands over the information, that ISP should   
   make sure that the data wasn't generated randomly and get some sort of   
   evidence some proper software was REALLY used to perform this   
   investigation. Perhps even ask the Canipre guy to demostrate to the   
   judge how GardaLey actually works.   
      
   If you don't make sure that the data was aceuired intelligently, then   
   nothing prevents you or me from generating some random list of IPs and   
   going after ISPs.   
      
      
   > Poisoning a torrent has immediate effect in accomplishing relief against   
   > financial loss without the costs associated with court proceedings.   
      
   You can't poison an existing torrent because there are checksums in each   
   part.  But you can create a new torrent that pretends to be "Hurt   
   Locker" but ends up being some gay movie or just 2 hours of ads for the   
   Shamwow guy.   
      
   > I continue to note that you absolutely do not in any way want to talk   
   > about the Tek Savvy side of this equation - the reliability or accuracy   
   > of their IP logs and how (or if) that issue would come up in court.   
      
   From a DSL point of view, they should have a fair degree of accuracy,   
   although they did make 42 errors in their initial pass. Every PPPoE ogin   
   is logged and authemticated to a specific customer. However, they do not   
   know if that PPPoE was made from the customer's home or not. (But bell   
   would know).   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)