[arin-ppml] CGN multiplier was: RE: Input on an article by Geoff Huston (potentially/myopically off-topic addendum)

[Chris Engel]

> Hi Chris,
> 
> Actually I agree entirely -- and that's precisely why I have been tossing this
> same argument up for 4-5 years now. Diversity is a great but ultimately fragile
> thing in systems -- like the Internet as well as monetary/banking sector --

> that are absolutely critical to the everyday affairs of lots of non-industry

<snip>

> Where would you draw the line?
> 
> Regards,
> 
> TV

Tom,

Simply put, I wouldn't.  Nature does that for us. It's called Natural Selection. It can be a bit messy in the short term but in the long term it's quite effective. Things that don't work well for their environment, if left to their own ends,  simply don't last and are replaced by things that do. (IMO) Excessive attempts to interfere with that process and impose some top down plan more often cause deeper seated systemic problems then any benefits they might provide.

Using the most recent financial crisis as example, the greatest failure there (IMO) was the government(s) stepping in to "soften the blow" and provide some sort of safety net. (As an aside, my understanding is that no small amount of the problem itself can be attributed to government efforts to provide loans to people who weren't really qualified for them in the first place..... i.e government interference was a strong contributory factor in causing the problem in the first place) Yes that provides some short term relief to the many who would have been adversely effected....but it also almost guarantees that same sort of greedy, short-sighted and irresponsible thinking that led to this crisis will be repeated again in the near future....perhaps not in the same exact way, if regulations prevent it but it will find a way to express itself. As a banking executive, why wouldn't you engage in risky but profitable gambles if you knew the government would step in to save you if things fell through?

Which leads us back to the topic at hand.  If IPv6 really is such a clearly superior solution, ultimately we'll get there.  If NOT and CGN or some other solution proves perfectly viable....then they will persist. Things may get a bit bumpy for awhile....but ultimately they'll sort themselves out and we'll end up with solutions that work well for their environment. Bottom line, people are, on average, better judges of their own situations then they are of others. I believe that to be an axiom....which is why I shouldn't be trying to tell you what you should run on your networks...and you shouldn't be trying to tell me what to run on mine.

If you ask me to predict what will happen....I'll freely admit that I haven't a clue. I suspect for the next decade or so, we are likely to have a pretty heterogeneous internet....with individual players running different things on their networks and having to figure out mechanisms to transition traffic coming across their boundaries. I'm ok with that. It'll be messy, but there really is no other way to figure out what REALLY is going to work and what won't without a variety of solutions undergoing a trial by fire. If they work decently, they'll last...if not, they won't. YMMV.



Christopher Engel