There
was an error in the parsing of the home domains list used to generate my
results published 5/2/01. Although the domains ".da.uu.net" and ".dsl.gtei.net"
appeared in the home domains list, they were ignored during logfile processing.
This led to the number of clients from home domains being *under* reported
in my analysis of 5/2/01 (not surprising, since these two domains did account
for a reasonable number of clients).
On 5/13/01 I re-ran the analysis, including .da.uu.net and .dsl.gtei.net as supposed 'home domains', and updated the rough results on my webpage. The new results increased the estimated percentage of home users noticably. The tables below show the original results from 5/2/01 and the revised results from 5/13/01 respectively.
Original (From 5/2/01)
|
|
|
Home clients as % of all clients |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Revised (From 5/13/01)
|
|
|
Home clients as % of all clients |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Grenville Armitage, 5/13/01