'Fuzzy logic ... suggests that in reality things cannot be presented in simple binary terms but rather as a set of degrees of truth similar to probabilities' (Rajala 2004).
The digitisation process accepted a number of parameters regarding the level of accuracy:
the digitised wall lines are only the approximate position of the buried walls, given shadows, lighting, and collapse/taphonomy.
the line thickness used in digitisation is a generic one, rather than trying to reflect relative thicknesses of ancient walls. It is apparent on both the satellite and the aerial photography that walls are of variable thickness, but it was considered too difficult to try and reflect this within the digitisation process. However, ideas about relative wall thickness will be documented in the database.
the collapse and erosion of the earthen architecture at Merv has created erosion slopes, which spread out from the walls (Fig. 23 and Fig. 25). Measurements for the database are taken from the top of the slope, which broadly equates to the conjectured internal wall line, thus giving a reasonably accurate internal dimension for the space.
wall lines are more or less visible in different parts of the city, owing to vegetation, lighting, and moisture – so the possibility that patterns in the digitised layers may reflect survival conditions, rather than historic structuring, needs to be recognised.
images taken at different times of the year, with different moisture levels and vegetation, showed different evidence. The digitisation is based upon using all of the available imagery.