The "problem" with different setups is there if line lenghts are measured by same method for low and high Y's.
Backlines are always straight, so these should define the nominal line lenght. In that respect I see Sandras' calculations a bit misleading, even though differencies between the lenghts of different Y's are apparently correct.
Frontlines are in practice straight if Y is very low. Then normal line check method is always valid (trimmer out, bar fully sheeted in = all lines are equal lenght). But if Y is high and same method is used, it gives different trim (when adjusted all lines equal) because now front lines are not straight when connected to kite, but go via Y route. Then frontlines "are shorter" compared to backlines, because they have that bend.
This difference is of course countered by the bridle of the kite for which high Y is originally designed for. So normal line lenght check method works still, but only for that kite and that bar combination, or similar high Y designed kite. This "design" work is just simple bridle line lenght paring to high Y bar.
So, there will be trim problems if high Y and low Y bars are cross used.
But like Peter Frank wrote it is usually quite easy to correct these normally not so big differencies when tuning a kite. There is nothing strange here, but some less experienced kiters may become puzzled why different bars - "correctly" checked - will work differently if cross used.
Pinching is very theoretical problem. LEI's have very low projected AR's, meaning wing span is small. Then line attachment points are close to each other, and there is also some stiffness in LE strut. So these kites tolerate very well high Y's, because "pinching angles" are low even then.
However low Y is universal and it has some other advantages, so there is no reason not to use it. It works of course in high Y kites, but then trim must be corrected by ....... backlines. (better leave this unanswered
Big high AR race foil kites which can have nearly 10m wingspan and not much canopy curve will not tolerate very high Y's, or at least as low as possible Y is better.