What about a broader definition of what can constitute a paleolithic diet beyond individual lists of allowed and banned foods. I quite like Loren Cordain's ideas from some of his papers on this based on surviving hunter-gatherer groups. No doubt people will disagree about details. High, moderate and low are relative to the modern western diet.
Protein - High (19-35%)
Carbohydrate - Moderate (22-40%)
Total fat - Moderate (28-47%)
Total fibre - High
Vegetables and fruit - High
Nuts - Moderate
Salt - Low
Refined sugars - Low
Glycemic load - Low
Feel free to argue about the details but what do you think about the concept of a definition like this? How would you define it?
A definition should not focus on macro-nutrient content. If you want a very broad definition just go with "Eat real food, avoid things with poison in 'em."
Then you can argue about what exactly is real food and what contains poison ^^
Combine Michael Pollan's rules (Eat real food; 5 or less ingredients; etc), with "don't eat things that cause allergic reactions in people - even if YOU don't have the reaction" (peanuts, wheat, dairy, etc), and add in a dash of "don't eat things you wouldn't eat raw" (grains, beans, etc).