Wow, interesting thread, glad I missed most of it.
Some thoughts though....
1) Mac Pro is a Server class machine, comparing it to Desktops is silly.
Not even close. It's a skimpy workstation.
Look at any "server grade" MB and the differences will be glaringly obvious. Namely Number of SAS and SATA connection, Number of RAM slots, and etc. The top of the line 2009 Mac Pro is configured like a home PC with a medium to good grade WS MB. Nothing close to server is offered in the MP line.
2) A box of carefully chosen consumer grade parts from NewEgg does not make a workstation. (sure, nice desktop, maybe, but NOT a Server grade machine) It also requires you to violate the Mac OS X license to use those parts as a Mac.
Which is it? Workstation or server? Do you actually know the difference? It doesn't sound like it from this.
So let me get your statement right: If I buy all the parts for a server from NewEGG and put them together it won't actually be a server because they came from NewEGG? LOL Someone needs to sue those guys!
3) 2009 MacPro is certainly less value for the money than prior models. The 2008 Octo was certainly a great value.
Yeah the 2008 machines were "fair". Not "great", not "fantastic", just fair. 2009 machines are not.
4) Above all, Apple has to make a profit, it is why they exist. Get over it.
Then they're stupid. I would be willing to bet they're taking a real beating on MPs in 2009 because they've priced themselves right out of a very large section of the market.
5) ECC: HoooBoy... Let me don my flame retardant suit....
Yeah, and you better make sure that suit fits tight with all the misinformation you've added here.
Those of you with access to racks of densely packed server systems, pop into your management software and take a look at the number of ECC corrections that have taken place in the past month, it will surprise you.
Modern memory is very densely packed, has insanely small transistors, and operates at miniscule voltages. It takes a *much* smaller event to toggle a bit than it did even 10 years ago. It may have taken a "super-nova generated alpha particle" to flip a bit in a dual rail 5v/12v 1Kbit DRAM chip from 1979, but in 2009 it takes a lot less.
If you reboot your PC every night/week, you may not have an issue, but when you are dealing with servers that run 24/7 with many many weeks of uptime, those bit flips become possible, and problematic. The insidious part is that you may not know it has taken place until the machine crashes, or generates an incorrect result.
In a server environment where reliability, accuracy and uptime are crucial, ECC is a requirement. For your desktop, not so much, but going back to point#1, a Mac Pro was not intended to be a desktop, hence it uses ECC.
This is really laughable. Everything presented here is wrong or the opposite of being correct.
Modern memory is MORE prone to ECC not less.
Servers get ECC hits because of overheating. They're basically false positives of actual soft errors. You can get the same thing to happen on your Mac Pro by removing the DIMM heat sinks or by unplugging the fan. In this case it can be argued that having ECC capable RAM is better but this is still pilot error by all reasonable definitions and in such cases your machine is very likely to crash or shut itself down anyway. It's not an actual ECC soft error. The same thing happens to people over-clocking their RAM on over-clockable systems. Here read this:
http://www.anandtech.com/mac/showdoc.aspx?i=2832&p=2
"We had no problems running all of our benchmarks with the standard Crucial FB-DIMMs; however, if we ran a memory stress test for even just a short period of time the modules quickly reported correctable ECC errors. Apple's original modules did not generate any ECC errors, so it looks like the additional cooling is necessary under the most extreme situations."
And this suggests the SAME thing:
http://diglloyd.com/diglloyd/free/MacPro/memory.html "
Verifying ECC status
Then verify that there are no ECC memory errors, by choosing “More Info…” and clicking on “Memory” (as shown below) . If you see ECC memory errors, return the memory to the vendor—ECC errors should not occur under normal operating conditions."
as just two examples.

The RAM here in the first example is essentially getting so hot it's "losing it's mind" so to speak. The same thing is common in data-center sized servers.
You can see ECC errors if any have happened during your up-time in the Mac Pro. Just pop open the System Profiler and click on Memory. If any have happened since the machine was powered up they will be listed. I challenge anyone with a properly cooled system to show me one. If all your RAM has finned heat-sinks and the fans are working there will be no RAM malfunction and therefor no ECC corrections will be listed.
If ECC is a requirement, crucial for accuracy and uptime then there are millions of systems in the world currently defying your premiss.
And finally in the almost impossible event of an actual ECC error you're machine
will not produce an inaccurate result or crash as you incorrectly state. ECC doesn't just check for errors, it corrects them. Bad memory modules that cannot be corrected my cause such troubles but that a tiger of a different tale.
oh yes, almost forgot...
6) Apple needs to sell a consumer level desktop machine.
They already do. The MacPro Quad is one. It's just priced like it's not.
