This necessarily skews their impressions. While it’s said less often, the same skewing infects every report cranked out by Keystone, Internet Traffic Monitor, Google, Alexa, Traffic Ranking, F-Secure, Symantec, Trend Micro, Panda, Kaspersky, MessageLabs, trimMail and every other outfit that tries to generalize some aspect of web activity. That’s why you see four virus surveys from different vendors at the bottom of every page in Email Battles. Each vendor serves a slightly different demographic mix, the results from which accurately reflect that firm’s experience.

Sorry, after much scrutiny, we still haven’t unearthed any inherent dishonesty in either the CacheLogic graphic or the article. But we could provide additional light with more metrics, like volume. If you have sources you would like us to include, please don’t hesitate to let us know. Heaven knows, we appreciate all the help we can get.

It is four separate layers, so the HTTP graph is _behind_ the P2P graph. This gives the impression that somewhere around 2002 (can’t quite tell when, due to 3D effects), HTTP just disappeared, when this was actually just when the estimates of the two crossed.

We’re a little more misled because this is a graph of _percentages_, but it only goes up to 70%. Visually, it looks like P2P is 100% of bandwidth, when it’s barely past half. You can see how this is augmented by the fact that HTTP, which no doubt takes up most of the other 40%, disappeared in 2002.

You didn’t give an explicit source for the data, but we see that it’s from CacheLogic by clicking on the next link in the text. This provides a bit of context: CacheLogic sells products to cut off P2P traffic, so of course they are working hard to trick viewers into thinking that P2P traffic is now 99.9% of bandwidth. (It’s been discussed to death, but shall we talk about the actual data? Have methods changed/improved/grown more biased since 1999? CacheLogic sells ‘deep packet analysis’ because just looking at the port is misleading, but do they have packets to compile stats from dating back to 1999, and if not, when did the graph switch to a more inclusive method?)

Even though you didn’t draw it yourself, please don’t lend eyeballs and credibility to misleading graphics like this. Yes, we all know P2P traffic is on the rise since ‘99; why use dishonest graphics to convey a true point?

The internet is no longer just for email and browsing.

Well the current majority who are transferring large files, those files are copyrighted and the people who invest their money are not making their investment back. The stated future of data bills will mirror my statement and anyone who holds a copyright can then make money. From big media to regular joe who files a copyright and then places his work onto P2P. Such any idea jives with why copyright was established.

This model is a no brainer, as the internet will displace many services ISPs currently provide. Cable TV and phone. Also, with this model the ISPs can offer $5 a month plans with a small amount of bandwidth per month. Those want the internet but dont want to pay through the nose would jump onto the $5 plan.

Didnt you ever take economics?

Even worse, they will attract a lot of ultra-low use customers who will not generate any revenue for them.

If ISP’s cant handle large amounts of data then they shouldnt advertise ‘unlimited’ services at prices so low that they cant afford to improve their infastructure.

This is a no brainer when ISPs are now trying to sell phone service, but with the Skype(s) of the world they’ll need to seek relief and change the way they bill monthly for data.

Peace, Chaser

You’re always managing traffic, trying to keep both sides happy.

ISPs should see p2p as a the killer app of the future instead of a threat.