Bad news: f116v was substantially mangled by Retracer

[Jorge_Stolfi]

There is a very strange glyph that look like an overlay of two different shapes. Both of them appear to be roughly the same color, so I'd say this mess could have been created in one sitting. I see no obvious signs of past writing, previous faint strokes, etc.

And I agree.  That entire word appears to have been retraced over the all but invisible bits of the original text.

The glyphs [in worm-scrapped patch 1] are visibly damaged, and to me it looks like even the darkest strokes were disrupted by this damage, which suggest to me that the worms happened after the writing was finished and there was no attempt to write over these glyphs after the damage was done.

The sequence of damage and retracing is debatable, but I cannot make sense of the ink patterns on that page without assuming at least one round of retracing. 

Consider the leg of the initial "six" on line 3, for example.  Could that be the worst level of fading that happened?  Statistically, we must assume that the worst was worse than that.  Many original glyphs must have faded even more than that leg -- that is, they are totally invisible now.

In fact, it is the claim "there was no retracing" that needs to be proved.  Hundreds of years ago large parts of the text would have already faded away to the point of almost being lost forever.  Why wouldn't the owner try to salvage the document by having a skilled scribe carefully retrace what could still be seen?

And bad retracing would explain why the handwriting is so stilted, and why not even expert paleographers can make sense of that text...

Maybe the rationale for having both lights on was to separate dark spots created by pigmentation from dark spots created by shadows on the rough surface. Similar to the effect of a circular light source.

Since the two lights are quite oblique (can't tell for sure, but roughly 45° elevation), that does not eliminate the shadow effects on the rough surface.  "Hill tops" that get ~45° light from both lamps will be ~40% brighter than "hill slopes" that get 90° light from only one lamp; and the bottom of sufficiently deep valleys will get no light at all,

But my point, again, was that the image with both lights on should be the simple average of the images with each light separately.  Thus they did not have to actually take the '0' images.  They could have obtained the same result from post-processing the 1 and 2 ones.

Instead, it would have been more useful to take a set of images either with light coming from East or West (to enable full 3D reconstruction) or with an all-around illumination (to better reduce the effects of roughness).

All the best, --jorge