Ok, summary here:
Cl = contact emulsions, low speed, high UV sensitivity < excellent keeping of AZO is an example >
Br = enlarging speed, good visible light sensitivity < moderate keeping >
Cl/I = variant on Cl with more visible sensitivity < poor to mediium keeping unless heavily doctored >
Br/I = variant on Br with more visible sensitivity < moderate keeping >
Cl/Br = enlarging speed with reduced keeping
Cl/Br/I = variant on above. < poor keeping unless heavily doctored >
Example: Keeping of Cl/Br (Brovira) is poor due to inherent nature and need to use Rhodium Chloride to get good curve shape. Rhodium does the same in Br and Br/I but with less keeping problem but also with less contrast effect so G is only half right.
BIG OVERSIMPLIFICATION HERE:
Now, imagine a pure AgCl in equillibrium with the environment. If a chloride leaves to the surround it returns without changing the emulsion. The same is true for AgBr. The emulsion tends to keep well. AgBrI and AgClI tend to keep well because the Iodide is LOCKED in place. Nothing much changes the crystal. HOWEVER for an AgClBr emulsion, both Cl and Br can leave and return in equilibrium. They do NOT need to return to the same place and if they do not, they stress the crystal and cause changes!
All keeping I report for raw emulsion is from emulsion held at 4 deg C. All keeping for coatings is for coatings held at 20 C and 50% RH as much as possible.