[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

What Should an Engineer Address when 'Selling' IPv6 to Executives?

On Tue, Mar 5, 2013 at 11:49 PM, Mukom Akong T. <mukom.tamon at gmail.com> wrote:
> On Wed, Mar 6, 2013 at 12:09 AM, William Herrin <bill at herrin.us> wrote:
>> 1. What is the real dollar cost of doing the project (including both
>> up-front and currently indefinite ongoing costs of dual stack. And
>> don't forget to price out risk!).
> Now in the post, I mention cost elements. At a point when you are still
> trying to convince execs about v6, is it possible to have an accurate value
> for this cost. Wouldn't cost elements and ball-park amounts be sufficient?
> Please could you shed some more light on 'Pricing out Risk'? any tools and
> techniques to do that would be highly appreciated.


For that, you need the help of a real cost analyst. That's what
they're for; they help organizations figure out a solid idea what
something will really cost before they start spending money. If your
organization is large, you may even have one on staff somewhere.

You can also consider pitching an IPv6 pilot project where you get
your IP addresses and bring IPv6 in from your ISPs but then stop just
on your side of the border rather than thoroughly deploying it. That
strongly bounds the cost, including the cost from risk. One of the
elements of such a pilot project would be contracting a cost analyst
to help you collect figure out what data to collect during the pilot
and then produce a cost model from the data to figure out what it'll
really cost for a full deployment.

>> 2. What is the real dollar cost of not doing the project. (i.e.
>> customers you expect to lose because you didn't do it. Don't suggest
>> that IPv6 will allow you to avoid acquiring more IPv4. That's not yet
>> true and if you say, "It will be in 5 years" the exec will respond,
>> "great, come see me in 5 years.")
> IPv6 has elements of a disruptive technology (right now it really only
> addresses the needs of a fringe segment of the market and also is perceived
> as worse with respect to feature set). The inherent problem with such
> technologies is that no one knows the real dollar cost of NOT taking action
> (when concrete data becomes available to support that, it would mean it has
> already seen market success and so if you still don't have it, you'd be too
> late.)

Practically speaking, there's no point in projecting the cost of never
taking action. You only have to project the cost of not intiating
action *this year*, or within two years or whatever the budgeting
cycle is that allows you to get started on the proposed project.

>> Implicitly they'll also be looking for the answer to a fourth
>> question: Do you know WTF you're talking about? If you assure them
>> it's all peaches and cream with tiny costs and no opportunity cost,
>> the answer is, "no."
> I believe if anyone who can phrase the "IPv4 Exhaustion Problem + IPv6
> Solution" in very specific terms of the business model of the company will
> implicitly inspire confidence in execs that they know what they are talking
> about.

Your first paragraph loses the argument: the day has past when IPv6
could become a credible solution to the IPv4 exhaustion problem. Like
it or lump it, NAT was the solution to the IPv4 exhaustion problem.
Which the exec will learn when he chats up his computer literate buddy
before making his decision.

If IPv6 approaches ubiquity, it'll get another crack at solving that
problem. Until then, the business case for IPv6 deployment is about
making your products compatible with emerging standards and to what
(if any) degree your customers will penalize you if you do not do so
during your projection window.

If you're an ISP or you make network software, this is a
straightforward case to make. There are public sources of IPv6
deployment rate data. You can presume that a similar rate holds among
your customers and that the customers who deploy IPv6 will disqualify
your product if the product doesn't work with IPv6.

If your business isn't networks, you have a much harder case to make.
As another poster noted, the end of IPv4 is not on the radar yet. A
statistically insignificant number of people will change banks this
year over their bank's web site IPv6 reachability.

Bill Herrin

William D. Herrin ................ herrin at dirtside.com  bill at herrin.us
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004