I agree broadly with his conclusion - but I'm not quite sure what point he is trying to make by posting those comments on a metal detecting blog. Excavation, as opposed to non-intrusive field surveys, is an intrinsically destructive process but circumstances vary widely, depending on such factors as the aim of the project, whether it is in an urban or rural environment, the type of landscape, whether the project is research or developer-led (rescue or salvage), and so on. It should also be borne in mind that excavation is only one facet of archaeology. The removal of topsoil in the very limited area of an excavation (one trench is typically only 20 square metres) on one site does not detract from the fact that the topsoil may be of vital importance at another location, the fact that noting features such as pottery scatters in the initial fieldwalking may have alerted archaeologists to the site in the first place, or the fact that field surveys are often highly significant without the need for any excavation at all.
It seems a necessary "evil", the pragmatic solution in certain cases of at least conducting some archaeology rather than none at all within the constrained parameters, is being used to justify a far greater and unnecessary "evil", an unhindered and selective grabfest of metal items in the archaeological record elsewhere.
"Steve" has said nothing about responsibility and moderation, nor has he mentioned the drawback of the selective nature of a detector or its widespread abuse. Metal detecting may be of great value in the right circumstances but to promote it without any qualifier by using the bare information that topsoil is often removed in the very limited area of archaeological excavations is disingenuous and misleading.