] The objective of including some 'hurdle' that new DNAs must get over
] before they can play in the game seems to be an attempt to ensure
] that organisations taking on the task of a DNA give some thought to
] what resources they will need and what quality of service they can
] deliver.

I think a fundamental question is whether ADNA (as the entity selecting
DNAs) wants to exercise some discretion at selection time, or whether
the process should be deterministic, so that a firm of accountants (for
example) can decide whether or not an application should succeed.

If the "deterministic" approach is chosen, then the IAHC/iPOC/gTLD/CORE
application process would be a good model to draw from. 

It seems to me that the gTLD process has been carefully crafted so that
there is no room for discretion in selection.  Once you relax that
constraint and allow discretion, there are a number of ways you can
change the criteria so that they seem less "arbitrary" (which seems
to be something Michael Malone is not happy about, and I can see why).
In that environment, the gTLD model does not seem so appropriate.
__________________________________________________________________________
David Keegel <djk&#167;cyber.com.au>  http://www.cyber.com.au/  +61 3 9642-5997
Cybersource P/L: Unix Systems Administration and TCP/IP network management