If my understanding of the code definition of "outlet" is correct, all circuits supply "outlets." I think the distinguishing feature of the first proposal you posted, earlydean, is that it requires AFCIs only for circuits supplying outlets in "living areas." The 2005 code doesn't have a definition for "living area," so I don't know exactly what that term means.
It does appear that there is an existing exception for life support equipment, so the first item in my list is covered.
I note in reading through some of the proposals and comments on them, that section 700-something prohibits putting fire alarms on AFCIed circuits. It would appear that they themselves don't have enough confidence in these contraptions to allow them to be used when the rubber really hits the road.
Consider the following set of facts:
1. The NFPA themselves don't have enough confidence in the reliability of AFCIs to even allow them to be used on fire alarm circuits;
2. There is a lot of evidence that AFCIs have a high level of unreliability in the form of false tripping;
3. People on this forum have done their own tests of available AFCI that showed that they are incredibly unsuccessful at detecting actual arc faults.
These facts cause me to reiterate the main point the list in my original point, that there are many uses of electricity where a falsely-tripping AFCI will cause very bad problems. These applications should be permitted to be on circuits without AFCIs.
[This message has been edited by SolarPowered (edited 08-04-2006).]
