[IETF-IDRM] Re: [IDRM] draft-irtf-idrm-handle-system-00.txt

Thomas Hardjono thardjono@mediaone.net
Mon, 28 May 2001 21:56:25 -0400


Yes, Mark, I realize that it won't work :) .. I was rather soliciting ideas 
about solutions.

I think identifiers/names/URIs can be guaranteed to be unique and
persistent forever (literally).  Just think about ISBN numbers.
An ISBN number will not be "recycled", in the sense that the number
will be associated with the book long after the author is gone.
As far as I understand, each country has a body/authority to which
an ISBN number can be requested.

However, in the digital world, the location of the work on
the Internet may change.  Perhaps the notion of unique naming & 
naming-authority should be decoupled from the system that manages the Records
pointing to locations.

If a naming-authority is "acquired" by another naming-authority,
then perhaps only the legal ownership should change.
The NA numbers stay the same, only physically managed by a different
company/organization.
(The same analogy applies to one CA acquiring another.  Another analogy
is when a company who owns a block of IP addresses
acquires another company with a different block of address.)

cheers,

thomas
------

At 5/26/01||09:05 AM, Mark Baugher wrote:
>Hi Thomas,
>
>At 01:53 AM 5/26/2001 -0400, Thomas Hardjono wrote:
>
>>In the case of moving NAs, Does it make sense to simply concatenate NA 
>>numbers
>>as sub-NAs (parsing from right-to-left), creating a path upward to the
>>root NA (ie. similar to CAs and jurisdictions)?
>
>This won't work for a persistent name.
>
>Mark
>
>
>>cheers,
>>
>>thomas
>>------