I personally think that only if
1. a really huge number of CSVs (preferably
all of them
) were trained with the same methods, by the same people, who would devote more or less the same amount of time to their training
2. later the effects were measured in the same way, by the same judges in similar conditions
3. the results were objectively compared with statistical methods
we could honestly start making generalisations, comparisons and speak of working and non-working lines.
Of course the above procedure is
not feasible and thus I agree with Przemek and others....
In fact theoretically every CSV should possess SOME working potential and it's up to the owner and other circumstances and coincidences if these abilites would be brought up or not.
E.g. it's like playing the instrument: nobody knows how many people from those we pass in the street might become top musicians IF the parents bothered to send them to music schools! On the other hand many of those who play in orchestras send their kids to music schools and the profession is passed on to younger generations, although none of them will ever become a Mozart.
