Pretty but flawed?

[roq]

Hi again Richard

Sorry to bombard you, but just thought I'd mention problem #30869. The solution is a very pretty 5 move mate that would probably win you a brilliancy prize if you spotted it over the board. Black is a pawn up so if his king could escape he would probably win. Unfortunately though there is an ambiguos seven move mate found by Fritz in brackets in my line...
 
1.b3+ [1.Rc7 Nc6 2.b3+ Kb5 3.c4+ Kb4 4.Ra1 b5 5.Rb7 Nd4+ 6.Bxd4 exd4 7.Rxb5#] 1...Kb5 2.c4+ Kb4 3.Ra1 b5 4.Rb7 a6 5.Ra4# Line

So why did I give it ***** anyway? Well no normal human would find that seven move mate, I think, whilst the actual solution is hard but logical.     

[richard]

Hi Roq,

That is quite pretty.

I'm actually not as concerned about the longer mate issue as I am about the "winning material versus hard to see mate". One of the strong arguments against preferring the hard mate over the material gain is "why risk getting the calculations on the N move mate wrong when I can just pluck this queen in 2 moves? Therefore for a human the queen take is the 'best move' ".  I believe this argument can also be used for the long mates, the short mate is better as it requires less calculation (of course this is not true in all cases as some longer mates may have less branching points than shorter mates in the same position, but in most cases it probably holds).

Of course it is annoying for a user who does calculate the longer mate to find they are "wrong". "How can mate be 'wrong'?" is a natural reaction.  This is part of the reason I'm looking at removing them from the problem set, but the bigger reason (for me) is that it makes it easier for me to have a go at reducing the number of material wins versus hard to find mates in the problem set.

Regards,
Richard.