#2: Rules should be easy to figure out and not rely on a listing of specific models.
I actually think that a whitelist of specific models is much easier to figure out. Do you really want to have to test everyone's blasters before a war? Time wasted for travel and setup is one of the top complaints at community wars. That, and I really don't want to introduce a need for yet more gear to transport. Venues like Sycamore Island, Beacon Terminal, Northgate, etc. are not gear-friendly. Most of us test our guns to different standards and use different methods for output, shot time, even range. Ben and I use output methodology that requires the use of a computer to review the footage, iSoaker uses a stopwatch. Ben and iSoaker end up with x, SSC Ben ends up with average, I end up with average minus drop off. I no longer allow for more than 3 MPH of wind or less than 10 full shots when testing range. Ben allows AP drop off to 75% of max range, iSoaker allows to 80%, I allow to 90%. Both Bens measure range in feet and inches, rounded to the nearest foot. I do the same, but rounded to the nearest inch. iSoaker measures in metric, then rounds to the nearest half meter and converts.
Just to check everyone's range, output, and capacity would require up to a tape measure, angle measuring device, camera, computer with internet, gallon jug, measuring cup set, shipping scale, refilling source, somewhere flat with a hard surface, and a lot of patience. Obviously, these things would be better done way before a war, but newbies won't test their guns. We also have differing levels of credibility with our data. I test my outliers rigorously (thrifted CPS 1200 MK1, CPS 2000 #5) to make sure the results weren't a fluke. That poor 2000 has been range tested 46 times just because I couldn't believe that it really shot in excess of 54 ft. I also retest everything when I change standard, like the move from 6 shots to 10 shots. I didn't need to retest for the reduction in wind limit from 5 MPH to 3 MPH, as none of my original data was collected in wind greater than 3.0. It's not an arbitrary number, either - I tried to test some AP guns, but found that any more than ~3 MPH ruins the test. A 5 MPH crosswind can blow a 3x stream right out of the testing area.
If we set parameters objectively, then we're not just making people use the same light primary rules, we are also making them use the same testing standard. Getting everyone to agree on one standard is like getting everyone to agree on speaking one regional dialect of English.
A whitelist may be more confusing for newbies and needs to be kept current, but it's great for hosts. If it's not on the list, it's not a light primary, it's so simple to manage round setup when you know exactly what is allowed. You can just eyeball what everyone is using in a few seconds, if something is modded/homemade, it can receive a 2nd look, but most of the work is done already. They can all share an objective set of parameters, but the host is most likely going to end up relying on a list anyway. If we agree on a common method like Stephen's, the first thing I'd do is figure out all the stock models that make the cut and write them down for reference, aka a whitelist.