[LINK] Ergas article in the Oz

[zeta]

On 29/06/2010, at 0:42, "Tom Koltai" <tomk at unwired.com.au> wrote:

> The copper should be allowed to die a natural consumer dictated death,
> and not a mandatory legislative mandated snip.
> 
> Budgetary considerations in paying for the NBN are not a good enough
> technical reason to disconnect the copper.
> 
> Tom

That is all ok, but having dispatched several servers to the storeroom just recently because their support costs outdid their supposed usefulness, I believe the reality is that the product vendors wil determine the end of life of the various infrastructure, just like we saw the demise of token ring and similar protocols.  

While manufacturing plants tool up for the new tech that fibre and wireless offer, along with streamlined processes and the new raw materials used, they are less likely to invest in old tech such as copper and the infrastructure that uses it.  And so such vendors  who know their product roadmaps will be reducing their support offerings for (and the keyword here is) legacy systems, will start charging ungodly amounts for their product support line items.  And govt departments in general have no choice but to accept that these support costs are a fact of life and plan their own technology refreshes with this in mind.   

So if a server or switch costs $250,000 to provision, you can bet that at least 15% of that will go on "support" for a new system.   Then as this system nears its end of life, watch the support matrix and the costs go up almost exponentially.   EMC are very good at this for example.  I guess most of the vendors are, too.   And government and the big corporates have no choice but to pay it, for compliance reasons
or be pushed onto the new roadmap, which has a cheaper support cost and so on.

My prediction is, that. copper will, for better or worse soon become a legacy protocol that vendors will find 
ways to get out of, because afterall, it's about bottom lines, not what customers want, even a govt customer... 


rachel