I am trying to make sense of this:

http://support.novell.com/cgi-bin/se...i?10053626.htm (How
Novell Clients Cost Addresses - TID10053626)

Now it doesn't really explain what happens in the following example:

Client: 172.28/16
Server1: 172.28/16
Server2: 172.27/22

ie the PC and server 1 are on a "standard" Class B and Server2 is on a
"subnetted" Class B.

3. The Client calculates the subnet for the workstation address and every
server/target address in the referral list (those that are on the same Net
but different subnet). The client then compares the workstation subnet
against each server subnet in binary to determine how many CONSECUTIVE
matching bits there are. This is not that same as doing an AND logical
operation between the two subnets since it is checking for CONSECUTIVE
matching bits, not total matching bits. It will then take the total number
of subnet Id bits and subtract the number of consecutive matching bits.

Now in my example if the client uses its subnet mask ie /16 it will compare
exactly to both /16 AND /22:

11111111 11111111 (00000000 00000000)
11111111 11111111 (11111100 00000000)

The example in the TID doesn't seem to cover this sort of eventuality.

So, is the subnet sorting algorithm able to cope with the rather common
practice of sub netting outside the "normal" Class A,B,C,D boundaries or not?