This small text is the first in a list about the problems of hieroglyphic encoding in practice (actually, it could qualify as the second post on the subject, as my previous post about sign D396D396 could qualify too).
I was recently asked a question about the A90 A90 sign in the JSesh library...
I'm currently having discussions about a cleanup of the fonts, and the question of L16 has (re)appeared.
Back in 2008-2009, when M. Thomas developped most of the JSesh fonts, we had a lot of discussions about the Manuel de Codage and the like. Comparing the 1988 MdC with the Winglyph fonts, I wondered a lot about the way L16 looked in both of them.
As everyone (well, probably almost everyone) interested in Natural Language Processing, I have started playing with deep learning. Deep learning (which is basically a new name for neural networks allows a computer to automatically learn to perform a task, given a (usually huge) number of examples.
The sign D396 D396 is a nice example of the problems we might have with sign lists.
The sign is not in the original "Manuel de Codage", nor in the original (black lead) IFAO fonts. In the Hieroglyphica (and in JSesh current fonts), the sign is made of the eye sign on the "qd" sign (Aa28). Now,...