Re: An invented language (or a hoax?)

[Jorge Stolfi <stolfi@xxxxxxxxxxxxxx>]

    > However, I have a suggestion for Jorge (I think this is right
    > down his line of research). The writing of the Codex is
    > extremely regular, and clear. Say I scan it. Jorge, can you
    > figure out an optical character-recognition algorithm that would
    > turn that into an ascii transcription? I strongly suspect the
    > problem is not so very difficult.
    
Unfortunately I have no experience with optical character-recognition.
But our VMS fellow Andras Kornai (hello, are you still there?) has
published a paper or two on the subject, which I got from his
homepage; and his methods seem quite appropriate to semi-cursive
scripts like "Serafinese".
    
    > I had given much thought to it for the Easter Island
    > hieroglyphs, which are so astonishingly modular, and, in my
    > ignorance, I came to believe it could be done. I think it would
    > have very general applications, too.
    
I imagine that Mayanists could use that technology too,
although they don't seem to be at that stage yet:

  Maya Hieroglyphic Database Project
  http://outreach.ucdavis.edu/programs/maya2.htm
  
  Mayan Epigraphic Database (MED) Project
  http://jefferson.village.virginia.edu/med/glyph_catalog.html

By the way, Mayanists don't know the whole alphabet yet, so they have
the same transcription problems we have (Are these two glyphs the
same? Is this one glyph or two?) Only that theirs is much worse...

Plus they have the problem of figuring out the reading order, which we
fortunately don't have.

All the best,

--stolfi