Re: VMs: Colored paints, touch-ups, and the michiton text

[Gabriel Landini <G.Landini@xxxxxxxxxx>]

On Saturday 03 July 2004 20:29, Jorge Stolfi wrote:
> As you recall, the language of those labels still has not been
> identified with certainty. However, the spelling of "April" with "b"
> instead of "p" points to Spanish or Portuguese (where the month is
> spelled "Abril"), or Arabic (which has no "p" sound, and uses "b"
> instead in foreign borrowings). Moreover, the epenthetic "i" ("Abiril"
> rather than "Abril") seems to point to Arabic, which AFAIK does not
> allow "br". So the person who wrote the labels could be an Arab-speaking
> person in Europe, e.g. in Spain (which was partially under Islamic rule 
> ultil late 15th century).

Yes, this is very puzzling. The month names do sound Spanish/Portuguese. But 
June is something like Yoiin (i haven't got  the vms with me now, so I can't 
check) what could this be? 

> It seems a long stretch to call that letter a "phi". 

Of course, I agree, I was just giving it some name.

> The letter on f9v looks like a miniature and incomplete EVA "t". It 
> could be a distorted EVA "q" with upturned arm...

Hm... I still think it looks like those first-word-characters in the oladabas 
page.

> It has also occurred to me that the "phi" could be not a letter but part
> of the drawing -- a stamen, pistill, or whatever. Are there similar
> details on the other front-viewed flowers, under the blue paint?

There is, I seem to remember, a single "phi" in the lowest petal of that 
flower. Doesn't it look like a short eva <f> without the top horizontal 
stroke?

> BTW, the reason why the color separation did not work so well on f9v
> may be sample contamination from the brown ink. It appears that that
> the painting-over dissolved a little of the brown ink and created a
> faint brown smudge, running diagonally over the "phi". So perhaps the
> separation will improve if you resample the blue ink where there is no
> brown smudge.

Yes, I also think that there may these be other factors, but bear in mind that 
I have no expertise whatsoever in this area:

1. I wonder if the blue pigment when applied thinly has a slightly different 
hue. Let's not forget that the pigments are not behaving like ideal filters 
(which is what the colour deconvolution algorithm is based on).

2. When applied thickly (not really thick, but dark) then brightness is very 
low, so it approaches black which has undefined hue. I noted that when 
determining the vectors based on the darkest parts the separationt, the 
result is worse.

3. One can see in f1r where the signature is supposed to be and were WMV 
applied his chemicals, there is a blotch of a different colour. I wonder, as 
you also noted,  if the same is happening when the blue pigment was applied 
(so the underlying vellum has changed colour by either diffusion of some 
other chemical in the blue paint or deterioration/oxidation)?.

Cheers,

Gabriel
______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list