[Rivet] Poll: deltaEta -- signed or unsigned?

[Leif Lönnblad]

On 2011-04-12 18:46, Andy Buckley wrote:
> On 12/04/11 17:31, Hendrik Hoeth wrote:
> I'm sitting beside Hendrik at the moment, but may as well publicly
> register my opinion that an unsigned delta is safer and more consistent
> with our deltaPhi functions. If someone wants a signed delta(eta), I
> don't think it's too much to ask that they say vec_ref.eta() -
> vec_i.eta(). My $0.02 :-)

It's difficult to find a use-case for a signed deltaPhi. For a signed 
deltaEta it is not that difficult (in fact I saw it used in two 
different talks in the last weeks). I guess I would anyway vote for the 
unsigned version.

/Leif