[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

VMs: Re: Numbered transcription

Hi GC,

Interesting concept.  I was afraid you were going to ask me to
"hotlink" every word on a page!  The VGBT-Beta pre-release font is
now in the /vgbt directory at www.voynich.info/vgbt/VGBT.TTF , and
I've placed a couple sample pages in a protected directory
(samo-samo to get in) that demonstrate my preferred page display
format.  These are HTML's, which require this font for display,
but can these forms be set to include proofing information in both
EVA and VGBT, at the user's selection, and display the correction
in the appropriate font? (The "old transcription" field can draw
its font from the VGBT set since it is not modified, but the user
should have the option of choosing their most comfortable entry

That could be done easily, as long as the information is in the database generating the page (you wouldn't want to do it by hand, believe me). :-)

If you could define a consistent mapping between VGBT and EVA (more on this below), it would be reasonably easy for the JavaScript to perform it - all the code could be placed in a .js file separate from the pages, which could be updated as VGBT extends.

BTW: The page I put up was simply meant as a general indication of how such a process could be executed in JavaScript - for all the talk about web-based collaboration, there are precious few actual examples of it around (WikiWiki aside). :-/

I guess my problem with that is that we still need to come up with
a method of notation that joins and disjoins EVA characters.  My
{C} is an EVA <cc>, and if you write <cc> in EVA, there is no
operator present that states whether or not you agree with the
joining of the two strokes.  Well, that's a subject for a
different email.

Joins could simply be handled by adding an underscore "_" <join> character to EVA (but with a very narrow width font), and inserting that as you like. Then EVA-to-and-from-VGBT translators can be written trivially, right?

Then, the EVA text for display could simply have all the underscores filtered out (not difficult), but the EVA text for editing could have them left in place.

As a more complex version: if you also add a <disjoin> character (I'd suggest a single quote) plus two symmetric set of remapping rules, EVA-to-and-from-VGBT can be done nearly automatically. I'm thinking specifically of this type of rule:-

ch <--> c_h

A disjoint ch would then look like this:-


However, my guess is that this is probably more complex than is really necessary here: I'd suggest sticking to a single <join> character.

ISTM that, apart from the various looped <ch>'s, the "joinedness" ("ligaturation"?) between strokes is the main thing you're interested in transcribing, so that some effort can be made to systematically comprehend the body of the VMS as glyphs rather than strokes. Is that fair?

The core question then is: would your purposes be better achieved by adding a single <join> character to EVA (and mapping to your own set of glyphs in software) or by defining a glyph font which is forever a moving target?

Cheers, .....Nick Pelling.....