Some remarks on dipole showers and the DGLAP equation

It has been argued recently that parton showers based on color dipoles conflict with collinear factorization and do not lead to the correct Dokshitzer-Gribov-Lipatov-Altarelli-Parisi (DGLAP) equation. We show that this conclusion is based on an inappropriate assumption, namely, the choice of the gluon energy as evolution variable. We further show numerically that Monte Carlo programs based on dipole showers with infrared-sensible evolution variables reproduce the DGLAP equation both in asymptotic form as well as in comparison to the leading behavior of second-order QCD matrix elements.