__/ [ B Gruff ] on Thursday 27 July 2006 17:24 \__
> On Thursday 27 July 2006 16:55 Roy Schestowitz wrote:
>
>> __/ [ B Gruff ] on Thursday 27 July 2006 16:50 \__
>>
>>> On Thursday 27 July 2006 15:12 Roy Schestowitz wrote:
>>>
>>>> More than 95% of e-mail is 'junk'
>>>>
>>>> ,----[ Quote ]
>>>> | More than 95% of e-mail is junk, be it spam, error messages or
>>>> | viruses, report mail monitoring firms.
>>>> |
>>>> | [...]
>>>> |
>>>> | Further work has shown that most of this junk mail is originating
>>>> | on hijacked home computers.
>>>> |
>>>> | E-mail security firm Return Path said 99% of the computers it monitors
>>>> | that send mail have been taken over by spammers or virus writers.
>>>> `----
>>>>
>>>> http://news.bbc.co.uk/1/hi/technology/5219554.stm
>>>>
>>>> Microsoft Windows MUST be illegalised.
>>>
>>> Damnation - that's *another* area where the US leads the UK.
>>> Between them, they account for 25% of all spam, but they are streets
>>> ahead of us, 23.2% to 1.8% - and to make matters worse, they are #1, and
>>> we are
>>> #10 :-(
>>
>> USA is number 1.... for SPAM export. It's actually one of my .sig's.
>>
>> There's a misleading aspect to this statement though... a snag if you
>> like. While it might be true that most spammers (or most prolific
>> spammers) are in the States, the figures only account for machines that
>> are hijacked by spammers, not the spammers themselves. They could use
>> America as a zombie nation but live elsewhere.
>
> True. However, it could also be true that American spammers use the
> machines of other nations? I don't see why you prefer one scenario over
> the other?
You argue that there's a point of balance here... a sort of
global equilibrium where one takes machines from the pool,
at random. If that were the case, come to consider how many
among the world's machines are located in the States. It's
only natural to assume that a lot of SPAM will originate
from there, assuming worldwide O/S monoculture. A SPAM per
capita (where capita is a machine, or one inhibitant) would
be a more meaningful thing for this survey. They could
normalise all the figures by population size. Whilst you
could easily do this by looking at Wikipedia, finding out
how many machines have not yet retired and are also
connected is hard. And then you need to normalise by on-line
time, bandwidth, etc. It makes it a high-dimensional
problem, assuming you seek meaningful statistics from which
to conclude country X is most spam(mer)-friendly.
Best wishes,
Roy
--
Roy S. Schestowitz | Free 3-D Reversi: http://othellomaster.com
http://Schestowitz.com | SuSE Linux | PGP-Key: 0x74572E8E
5:20pm up 7 days 5:35, 9 users, load average: 1.01, 0.91, 0.85
http://iuron.com - Open Source knowledge engine project
|
|
