[split] Jorge's Retracer Thread: darker ink, retracing of text and drawings

[Jorge_Stolfi]

I think I spent quite a while looking at the manuscript, I'm quite familiar with the script.

I spent a lot of time myself doing that (not sure it was a "good" time  ).  I helped Rene and Gabriel to edit their interlinear transcription, checking and transcribing a good part of it.  That was based on older images, mostly B&W.  Since the BL 2014 scans became available, I have re-transcribed You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view..  

And I have been writing detailed descriptions of each page.  You are not allowed to view links. Register or Login to view. is still very messy and very incomplete but should give and idea.  Search for @f57v for a recent entry.   These notes may not be worth reading, but writing them forced me to look closely at every flower and nymph and star...

I had long noticed that there were late additions to the figures, like the "Habsburg" crown and showercaps.  Probably 25 years ago, no longer sure.  But it was only very recently that I realized that the original text and drawings are by now extremely faded, so that all the sharp dark ink is a later restoration.

All arguments to "mis-shaped", "redundant", "invalid", "nonsensical" presuppose that we somehow know the right correct shapes and designs. I don't think we do, and all these arguments are subjective based on one's own ideas about what the script should look like. For example, when there was a claim that \) character doesn't exist in the MS, I easily You are not allowed to view links. Register or Login to view. just because I remembered seeing it.

And I have my own long You are not allowed to view links. Register or Login to view. too.   And a booklet of sketches like these
   
But now I recognize that most of those -- even when they occur 2-3 times -- are accidental malformations by the original scribe, or mistakes by the Retracers.

Even if some of those are legit symbols (like those in the You are not allowed to view links. Register or Login to view. sequence, or "&" in modern English), assuming that they are errors until proved otherwise should not hamper any statistical studies or decipherment efforts.  Those studies must always take into account that there are many errors in the transcription files, including errors by the original Scribe.  Thus, for those studies, the MRT makes at worst a quantitative difference, by raising the expected fraction of errors in the input from ?.??% to ?.? times that...

All the best, --stolfi