I understand some of your reasoning, but I think you are not addressing the majority of issues for professional users .
That is a difference in the notion in what "professional" means. Professional means adapting to change over time. I'm well aware though that some use the label to cling to a rigid methodology that has been adopted over an extended period of time.
You mentioned the switch from PPC to Intel, ....
Based on my limited experience, the PPC-Intel move went very smoothly .
[/quote]
There is a distinct difference between transparently and smoothly. May folks confuse those two. Rosetta affords some transparent scaffolding but it did not enable a smooth transition. It doesn't transition/switch anything at all since still running the same old stuff.
I'm also keeping my software up to date , usually with a 6-12 months waiting period after new releases , same with hardware and OS .
Anyone following that kind of methodology couldn't have had a problem with Rosetta going away with Lion.
10.5 Leopard Oct 2007
10.7 Lion July 2011
That is approximately a 4 year gap. Even if the breath of the software didn't completely transition to fat binaries (or Intel only) for 18 months ( a period longer than that cycle) and then the user took another 18 months ( again a period longer than that cycle ), the total 3 year vendor+user upgrade cycle is still less than the almost 4 year "get your act together" window Apple provisioned. Not being done in almost 4 years is highly indicative of a " kick the can down the road" strategy leveraging Rosetta.
People who were still deeply dependent upon Rosetta when it was de-supported were NOT trying to solve the problem. The vast majority has decided to permanently depend upon that crutch. That really wasn't professional behavior. That is not following established "best practices" in software maintenance.
OS 10.7 and 10.8 introduce compatibility issues I haven't seen since to this extent since Apple went to OSX , see above .
"Save as..." vs implicit versioning support is not really a software/format compatibility issue.
There have been no dramatic changes in technology, or the OS itsself, which require loosing support for anything that worked flawlessly in Snow Leopard, and in most cases had been adjusted for current OS and hardware demands
Not really true. Pragmatically Carbon is at a deadend at 32-bit. Cocoa classes are only straightforward way into modern 64-bit OS X apps. 10.7 and 10.8 made that transition all the more apparent. In short, there was no large change for apps that had been keeping up. For those apps that had been lagging behind, ( despite several years of Apple at WWDC pointing to which APIs were the future and which ones had shifted more to a backwards compatible orientation (i.e., either are or going to deprecated) ) got caught.
With Apple's "Update every year" OS X cycle I can see some profession folks either skipping or waiting till the "last" update of any one one of those. For example, going to a even (or odd ) schedule.
10.7 10.9 10.11
or
10.8 10.10 10.12
I don't think Apple or most of the software vendors will have a problem with that. Similar to how many folks will transition to either Adobe x.0 or .x5 updates.
The other approach is wait 12 months after the OS X update comes up. That typically would be a transition to a 10.x.(>= 5) upgrade.
As for MP customers, there are no new markets for it, only existing ones .
If there are no new markets for the Mac Pro then it is dead (or at least zombie status). Mac Pro useres telling Apple they think the market is dead do not bring a new Mac Pro's to the market faster. All that promotes is Apple spending time on what plan they have for transitioning out of the market. Not spending time future developments.
A user always has to adapt, but at the same time a new workstation (and OS) needs to integrate into existing workflows without major hassle, and compatibilty with recent external hardware is mandatory .
It isn't if that external hardware is oriented to solving problems Mac Pros don't have. USB 3.0 sure. Apple not including a discrete USB 3.0 ( presuming that go with a Xeon E5 solution ) would be a huge blunder. Thunderbolt though largely solves problems that MBA and perhaps Mac Mini (and lessor extent iMac) have. Solving root cause problems the platform doesn't have is not in any way mandatory. Existing Mac Pros have PCI-e expandability. Existing Mac Pros can output to multiple (even more than just 2 ) monitors at the same time. Thunderbolt is not bring new, significant market exapnding capabilities to the table. It is in the "nice to have" category, but mandatory is an illusion (especially at the current price points which are no more competitive than the currents solutions. )
The existing "workflows" are in very simllar context. Using "save as" primarily as a versioning mechanism doesn't mean a person can't do version with the new Lion Applications. ( nevermind that this isn't an OS issue but an application level issue. ).
The MP market has lost many semi-pro and 'enthusiast' users to the iMac and MBP, as those became more and more capable;
The MP market has more than just one border.
[ more affordable, relatively less I/O ] <--> [ MP ] <--> [ less affordable, relatively higher I/O ]
It is quite myopic to think that there are only transitions between the first two above and not between the second and third. There are actually new users who effectively can come in from the above market because they were prevented form entering the market due to price restrictions. Price is a dominating factor even in "professional" market. The notion the Mac Pro price has no significant impact on the market is disconnected from reality.
Sure, the market below is much larger than the one "above" the Mac Pro. Mimicing the market below (e.g., Thunderbolt) isn't going to have much traction at all if folks are transitioning between those markets based on price. Nor is it going to have much traction trying to transition those in the above market.
In short, the Mac Pro should be focused on the same thing the Mini/iMac etc. folks are in terms of getting more users from the "more expensive boxes because now more capable". Focusing solely on the "Mac Pro is only good for Audio/Video pros" is dead end ghetto which doom the product long term. Following that strategy is exactly how the Mac Pro got to the position it is now.
it could be argued that most remaining MP users are non-customers, based on your opinion .
Not really an opinion. People who don't buy anything are not customers. A customer is someone who is
buying something. It is huge stretch to put people how haven't bought anything in 4-7 years as customers. Your business will probably have problems if spend most of the time focused on those kinds of "customers".
MP 2013 this month! 
, DON'T 
