[arin-ppml] How hard is it to transition to IPv6?

[Martin Hannigan]

On Wed, Apr 1, 2009 at 1:50 AM, Joel Jaeggli <joelja at bogus.com> wrote:

> Martin Hannigan wrote:
>
[ clip ]


>
> > That's a costly trick. Try a capex program to replace all CPE so that
> > every customers has chip sets that support 3.0 and DOCSIS. It all sounds
> > nice on paper, but it isn't.
>
> It doesn't sound nice on paper. It sounds messy, imperfect, likely to
> move at different rates in different providers/regions/customers. It
> will be subject to postive and negative network effects. The technology
> lifecycle of some of the newer components may be shorter than we'd like,
> while the older one's may in fact last longer than we think they should.
> This is normal.



It wont be subject to technology life cycle at all. Accumulated depreciation
across balance sheets should prove that out. It will be subject to cost. Not
everyone has the same cash flow  and finance-ability that a NYSE listed tier
1 in the US has. While I expect to see convergence finally really happen and
sustain some level of net growth due to cost reductions for customers of
same products, I also expect little growth in v6. Cost, costs, and
costs-s-s-z.


Best,

Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20090401/ad13cc31/attachment.htm>