RE: server .mets - links ono Slyck

From: Neil Saxby <nsaxby_at_mediadefender.com>
Date: Sun, 17 Jun 2007 15:09:06 -0700

Cheers Ben,

 

Obviously, I still have a lot to learn about all the techniques we use.

 

Speak to you soon,

Thanks,

Neil

 

From: Ben Grodsky
Sent: Friday, June 15, 2007 11:23 PM
To: Neil Saxby
Cc: Octavio Herrera; Randy Saaf; Tabish Hasan; Jay Mairs
Subject: RE: server .mets - links ono Slyck

 

Neil,

 

Totally cool and understandable about the "counter counters." :)

 

Whether our servers are in a .met or not isn't the only way we would have a "presence." The easiest way for us to protect is when users connect to our servers and everyday hundreds of thousands of users are on our servers successfully getting nothing. :) But we do protect on the eDonkey network as a whole. We're constantly working on more solutions to get better effectiveness on servers that aren't ours -- bandwidth contracts, IP rotations, dynamic IP alternatives, and other software initiatives to get IP sources.

 

Thanks,

Ben

   _____

From: Neil Saxby
Sent: Fri 15-Jun-07 13:34
To: Ben Grodsky; qa
Subject: RE: server .mets - links ono Slyck

Thanks Ben,

 

Makes perfect sense and Iím fully aware of who MD generally targets. The only reason I asked in the first place is because Iím working with a few clients running their own internal tests. When I receive a .met from a client it would be nice to turn around and say we still have presence. Failing that, I would use the reasoning you have outlined below and explain that the majority of eMule users wouldnít know what a .met is, let alone where to find one.

 

I also thought this information might help us understand how many people out there have identified our servers. If sharing this info with you guys is counterproductive please let me know and Iíll stop.

 

Btw: this is where the IFPI always obtain their .met lists: HYPERLINK "http://ed2k.2x4u.de/index.html"http://ed2k.2x4u.de/index.html, they usually obtain the full server list and itís fairly safe to assume that most European record companies would use the same list when testing internally. However, if they use another vendor (i.e. Advestigo) it could be in that vendorís interest to undermine our effectiveness (using a filtered list) and present their technologies as a more suitable alternativeÖ obviously ;o)

 

Just trying to help out a bit with those Ďcounter countersí and maybe obtain some slightly different answers for those disgruntled clientsÖ hope thatís ok.

 

Cheers,

Neil

 

From: Ben Grodsky
Sent: Friday, June 15, 2007 8:36 PM
To: Neil Saxby; qa
Subject: RE: server .mets - links ono Slyck

 

Neil,

 

The default user is who MediaDefender targets, not an expert who delves into researching how-to pirating forums, such as slyck.com. With that said, many threads on slyck.com refer to our eMule spoofing and decoying as an annoyance -- veritable testimonials of our effectiveness, but not quantified to be particularly meaningful to our clients. As the article says, "In the eMule client, there's already a list of servers in the server window." The article goes on to say, "The servers featured are not verified safe servers. Unsafe servers usually return purposely faulty results in searches or cause other difficulties with the eMule client." Sure there are ways to minimize and essentially avoid most of our protection techniques, given enough stubbornness and perseverance, but that isn't what we target.

 

We target general masses of people that use p2p networks. This is one of the reasons that it's important to come up with a testing methodology with our customers that we can reliably replicate in-house with our QA team. Amongst the data that we have access to is what typical p2p user behavior is, the problem we have with these user behavior data is we've never come up with any way to present it to our clients -- we can see it with database queries, but charts and graphs don't work well to express these notions and we end up resorting to anecdotal-sounding remarks. One common statement is "the average user clicks on the file with the most sources." This is absolutely true, but the % of users clicking on the file with the most sources is different from network to network and fluctuates a bit depending on how well the top-sourced file is named, or the file format (compare user searching for video but top result is audio, or vice versa).

 

Let me know if that makes sense.

 

Thanks,

Ben

   _____

From: Neil Saxby
Sent: Fri 15-Jun-07 05:28
To: qa
Subject: server .mets - links ono Slyck

Hi guys,

 

I got these 2 .met files today (links from this article on slyck: HYPERLINK "http://www.slyck.com/story1484.html)"http://www.slyck.com/story1484.html)... I guess youíve seen them already, but I was wondering if anyone could let me know if we have any presence on these lists and if we do, how strong it is.

 

Many thanks,

Neil

 

No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.472 / Virus Database: 269.8.16/849 - Release Date: 6/14/2007 12:44 PM

 

No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.472 / Virus Database: 269.8.16/849 - Release Date: 6/14/2007 12:44 PM

 

No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.472 / Virus Database: 269.8.16/849 - Release Date: 6/14/2007 12:44 PM

 

No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.472 / Virus Database: 269.8.16/849 - Release Date: 6/14/2007 12:44 PM

No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.472 / Virus Database: 269.9.0/851 - Release Date: 6/16/2007 12:50 PM
 
Received on Fri Sep 14 2007 - 10:56:05 BST

This archive was generated by hypermail 2.2.0 : Sun Sep 16 2007 - 22:19:48 BST