The Quality of PAS data

From the Furness Finders facebook page (5 October 2012):

Furness Finders MDC Roman Brooch found this week in Dalton, Dot, do you want to record it from this picture?  disc brooch, pin missing, traces of tinning on the front, looks to be a crude type possibly made locally? 5 October 2012 at 18:59

Tinning, eh? Is that from the use of the Regton's pocket spectroscope? And what was the FLO's reaction to that suggestion? Surprisingly (or not, depending on what you think of the PAS):

Dot Boughton Nice! What's the obverse like? I can record it no problem but this image is a bit blurry (sorry) and an image of the obverse would be nice. If you email me obverse and reverse images, not blurry, size and weight of the brooch I can record it from the images no problem! (5 October 2012 at 21:07) 

and checking that the finder had the landowner's permission to be on that land, when will you do that to avoid putting 'laundered' finds in the database? How many of the objects in the PAS database have been 'recorded' without the FLO even having the thing in her hand? How widespread is this practice and what does that mean for the reliability of the information in the PAS database in general?