In answer to some points raised on here, yes it was originally proposed that NERC worked totally independantly of NAS. It was what the ATC project team members wanted.
However, things are not as simple as getting what you want because it makes sense.
When you are in any phase of a project as big as NERC you cannot simply provide a single solution to the money me. When you prpose a system, or a development, or a change etc. you must not just justify your requirement, but you must give options. Generally the method used is known as an AOD (Analysis, Object, Design).
So you must give a detailed requirement of what you want, how you want it to work and why. But its not enough to simply do that, you have to give a range of options; As an example say you wanted to save more data from the workstations for later examination. You have to detail how you could do it:
1) Send it on the existing LAN (which might overload it, so details required).
2) Build an extra LAN to cope with the greater volume of data flowing.
3) Change the speed at which the processor handles other tasks, to allow your data to be sent.
4) Change the processors for faster ones.
Etc. Etc.
Its not a simple process and there are various stages (hoops!) to go through. At the end of the day you propose your solution, the one you really want, but its not your decision to make. Those that control the purse strings make the final decision based on the AOD they've seen.
Hence, despite it being the project staff's wish not to be lumbered with NAS it was out of their hands.
No idea why NAS flopped again, last time it was an incorrect route amendment which had been input. What the solution to NAS is I don't know but a stable and more reliable replacement is required urgently. However, as we all know, there's no money available for that! Sigh!!