[xri] Comments on 4-March XRI Reqs.

["Veizades, John" <jveizade@visa.com>]

This is a list of comments on the XRI Requirements doc that I discussed with
Gabe privately.  Comments?

John...

2.2.1.3 We should be more explicit on why we think that the URI schemes that
are used with HTTP are not adequate for the requirements outlined in the
document.

2.2.2.1 Following from the above comment there should words about the
identification of objects that have no network endpoint associated with
them.

4.1.1 Mention work that is going on with 2396bis and how that work may
effect the XRI URI scheme

4.1.4 Need a definition of what a permanent identifier is, how long does it
live and for how long can it be resolved.  I think that a permanent
identifier should live forever and there are objects that need to be
identified and resolved for longer than the life time of the physical
object.

4.1.4.7 and 4.1.4.8 For I18N we should try to refer to another body of work
that has dealt with and resolved all the issues of encoding, display, and
all the other I18N issues.  The IRI work is probably a good start if not
just the answer (see
http://www.ietf.org/internet-drafts/draft-duerst-iri-03.txt)

4.1.5 This section is not clear and the example seems to be a ID that
resolves differently in multiple distinct name spaces.

4.1.6.2 More concrete definition of attribute - examples 

4.1.8.1 This is unclear since identifiers have an assignment authority whose
practices should protect against leak through of personally identifying
information if that is desired.

4.1.8.3 What is Veronymity, this word does not exist in my reading of the
English language.

4.2.1.3 The URI should contain either an implicit resolution mechanism
(defined in the specification) or an explicit resolution mechanism that is
part of the XRI URI.

4.2.1.6 For scalability very large usually means infinite or very near
infinite.

4.2.1.7  High availability mechanism (load balancers, geographic redundancy,
etc) must not be precluded by the specification.  The implementation of
these techniques should not be specified in the document.