Date: Sat, 15 May 2010 14:11:57 +0100
From: "lsi" <stuart@...erdelix.net>
To: full-disclosure@...ts.grok.org.uk
Subject: Windows' future (reprise)

Hi All!

Just a followup from my posting of 9 months ago (which can be found 
here):

http://www.mail-archive.com/full-disclosure@lists.grok.org.uk/msg37173.html

Symantec have released "Internet Security Threat Report: Volume XV: 
April 2010".  My posting from last year was based on the previous 
"Internet Security Threat Report: Volume XIV: April 2009".  So I 
thought it would be interesting to check my numbers.  The new edition 
of the Threat Report is here:

http://www4.symantec.com/Vrt/wl?tu_id=SUKX1271711282503126202

You may recall that last year, the average annual growth rate of new 
threats (as defined by Symantec) was 243%.  This enabled me to 
predict that the number of new threats in this year's Symantec Threat 
Report would be 243% of last years; eg. I predicted 9 months ago the 
number of new threats in this year's Symantec Threat Report would be 
243% * 1656227, or 3840485.87.

The actual number of new threats in this year's Symantec Threat 
Report is 2895802, an error on my part of 24.6%.

This is quite a chunk, however it is not that far off.  My excuses:

- my number was based on averages, so it will never be exact.  There 
will be a natural variance in the growth rate, caused by many 
factors.

- in the new edition, Symantec have altered the raw data a little - 
the number of new threats for 2009, 2008, 2007 etc is slightly 
different to those same years, as listed in the previous version of 
the report.  I have not updated my projection to allow for this.

- Symantec note that "The slight decline in the rate of growth should 
not discount the significant number of new signatures created in 
2009. Signature-based detection is lagging behind the creation of 
malicious threats..." (page 48).

Am I retreating from my position?  Absolutely not.  I am now 
expecting the number of new threats in next years' report to be 
7036798.86. This is 2895802 * 243%.  This includes the error 
introduced by Symantec's changes to the raw data.  I don't think it 
matters much.

As this flood of new threats will soon overpower AV companies' 
ability to catalogue them (by 2015, at 243% growth, there will be 
2.739 MILLION new threats PER DAY (over 1900 new threats per 
minute)), and as Symantec admits above that "signature-based 
detection is lagging", and as Microsoft are not likely to produce a 
secure version of anything anytime soon, I am not at all hopeful of a 
clean resolution to this problem.

I continue to advise that users should, where possible, deploy 
alternatives; that they should, if they have not already, create and 
action a migration strategy; and that they should avoid like the 
plague, any software which locks them into a Microsoft platform.  
Business .NET applications, I'm lookin' at you.

Those failing to migrate will discover their hardware runs slower and 
slower, while doing the same job as it did previously.  They will 
need to take this productivity hit, OR buy a new computer, which will 
also eventually surcumb to the same increasing slowness.  They will 
need to buy new machines more and more frequently.  Eventually, they 
will run out of money - or, for the especially deep-pocketed, they 
will find they cannot deploy the new machines fast enough, before 
they are already too slow to use.  The only alternative to this 
treadmill is to dump Windows.  The sooner it is dumped, the less 
money is wasted buying new hardware, simply to keep up with security-
induced slowness.

Why spend all that time and money on a series of new Windows 
machines, without fixing the actual problem, which is the inherent 
insecurity of Windows?  People can spend the same time and money 
replacing Windows, and then they won't need to worry about the 
problem any more.  The difference is that sticking with Windows 
incurs ongoing and increasing costs, while a migration incurs a one-
off cost.

I don't think it takes a genius to see which approach will cost less.

Notes:
- see page 10 of the Volume XIV (2009) edition, and page 48 of Volume 
XV (2010) edition, for the relevant stats

- since my post of last year, I have also noticed a similar 
exponential curve in the number of threats detected by Spybot Search 
and Destroy (a popular anti-spyware tool). This curve can be seen 
here:

http://www.safer-networking.org/en/updatehistory/index.html

 - my projection of growth rates up to 2016 (written last year) is 
here:

http://www.cyberdelix.net/files/malware_mutation_projection.pdf

Comments welcome..

Stu

---
Stuart Udall
stuart at@...erdelix.dot net - http://www.cyberdelix.net/

--- 
 * Origin: lsi: revolution through evolution (192:168/0.2)

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

Powered by blists - more mailing lists