Cost of IPv6 for IT operations team

[Mikael Abrahamsson]

On Mon, 13 Apr 2015, Jens Link wrote:

> Back in '99 my employer send me to an Oracle training. I do not remember
> much about Oracle but I do remember one quote from the trainer:
>
> "I told my first customer about Y2k in 1980. He called last week. If
> anyone of you speaks Cobol the customer is paying $large_amount  per
> hour for fixing his problems."
>
> With IPv6 there is no fixed date but I will happen too.

Correct, there is no hard date, and the timing all depends on the 
organization and their philosophy, internal politics, skill level, and 
other factors.

Some will argue that the longer you wait, the cheaper it will get because 
others will have already done the hard work, and you can benefit from that 
work without paying. The executive might think that his bonus this year 
will be higher because expenses weren't taken this year, and he probably 
won't be around in 3-5 years anyway, so why should he care?

Some people believe things should be done that is purely customer driven, 
others say success can be had by being proactive. They're probably partly 
right both of them. In some cases, there is no right answer, for instance 
as outlined in the process of going from mechanical to electronic 
calculators around 1970 
(http://www.slideshare.net/Christiansandstrom/calculators-disrupted-presentation-780029), 
a lot of value was destroyed by this technological shift, making customers 
a lot better off (cheaper and more reliable devices) but the makers were 
just gutted.

We're most likely going to see a similar shift in the car industry in the 
next 5-30 years. Imagine whole divisions within companies that have 
expertise in gearboxes, ignition systems, combustion engines, turbos, 
their jobs will just go away when you replace this with an electric motor, 
a simpler fixed-ratio gearbox and batteries.

At least someone with IPv4 expertise can apply a lot of this for IPv6 when 
the time comes for their company to adopt, and a person trained in IPv6 in 
15 years, will most likely be able to handle IPv4 just fine, at least more 
easily than having a newbie Java programmer try to fix Y2K problem in 
Cobol.

So I still am hopeful and I keep hearing people discussing "when" IPv6 
will be adopted, I don't really here "if" anymore.

-- 
Mikael Abrahamsson    email: swmike at swm.pp.se