Hi Bob, Brian, I recently read through
draft-ietf-ipv6-deprecate-site-local-03.txt and draft-ietf-ipv6-unique-local-addr-04.txt to catch up with what was happening on the topic of site locals / unique local addresses. It was the first time I'd read draft-ietf-ipv6-deprecate-site-local-03.txt. I found it to be a good explanation of the issues that site local addresses, and more broadly, the issues overlapping address spaces cause. One of the criticisms it points out was the disagreement about what the word "site" means. I then re-read draft-ietf-ipv6-unique-local-addr-04.txt. What struck me about this draft is that it seems to quite often use the word "site" in just as fuzzy way that draft-ietf-ipv6-deprecate-site-local-03.txt criticises. For example, I find that the use of the word "site" in the characteristics list part of the Introduction really starts to imply the limitation that these addresses can only be used for addressing geographical sites, as to me, the word "site" has a default geographical connotation. I'd like to suggest the use of the more generic term "address domain" as an alternative to the use of "site" in draft-ietf-ipv6-unique-local-addr-04.txt. As "address domain" is a more generic term, I think there would also be some value adding some text giving examples of different sizes or types of "address domains". A geographic site would be one example. A mobile Personal Area Network (eg. a bluetooth / wireless enabled mobile/cell phone) could be another. Some alternatives to "address domain" might be "address area", "address group" or "address zone". I think each of these terms can be used to describe a set of devices that are sharing a common unique local address prefix, without implying any specific range or size or number. The bluetooth enabled phone with a camera scenario is what caused me to think about this. I recently witnessed a photograph being sent between two of these phones, just using bluetooth, without any carrier or carrier addressing involved (well, I don't know all that much about bluetooth, I'm presuming that phones have bluetooth addresses assigned in the factory, rather than via the carrier). Thinking about how that could be achieved using IPv6 over bluetooth (or some other wireless technology), I'd think unique local addressing would be the solution. Using the word "site" in this context, when discussion unique local addressing and mobile phone users, would seem a bit silly to me :-) Regards, Mark. -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
