The following warnings occurred:
Warning [2] Undefined variable $can_link - Line: 6 - File: showthread.php(1658) : eval()'d code PHP 8.5.12 (Linux)
File Line Function
/inc/class_error.php 157 errorHandler->error
/showthread.php(1658) : eval()'d code 6 errorHandler->error_callback
/showthread.php 1658 eval




Writing a fake Voynich by rules: copies, habits and the link between words
#83
todihk wrote:@quimqu: Can your generator already account for this shift through its existing rules, or would an additional hand- or bifolium-dependent rule need to be introduced?

Mostly it already does, without a new rule. And the one place where it fails is interesting for your question.

There is no rule about d in the skribe. It writes glyph by glyph from tables counted per hand (and per hand and section), and the first glyph of a word depends on the last glyph of the word before. So where d goes just comes with each hand's tables.

I measured your features, real text against the skribe from #81 (paragraph text, ZL3b, average of 5 texts), % of words:

A realA skribeB realB skribe
d- at word start14.813.66.75.8
-dy7.08.125.225.1
-edy0.221.50 (FAIL)18.418.2
-ody4.04.42.52.6
-dy, no e/o before2.82.24.44.4
-ey9.110.111.911.4
words with ed0.301.98 (FAIL)22.221.8
d's that are word-initial51461715

By Davis' hands 1, 2 and 3 all 27 numbers are within 25%. Language B is fine too. Only A fails, on -edy and ed.

The failure comes from two pages, You are not allowed to view links. Register or Login to view. and f58v. Davis gives them to hand 3, Currier has them as A. On those pages the real text has -edy 0.9%, ed 1.0% and 48% of the d's at word start. That is like hand 1 (0.2%, 0.2%, 51%), not like the rest of hand 3 (15.2%, 20.1%, 12%). The skribe writes them with the hand 3 table, so it puts B-style d there (-edy 26%).

So on f58 the position of d follows the language, not the hand. That fits your idea that it is an A/B thing. But it is only 684 words on one folio, and it depends on how Davis and Currier classified f58, so I would not push it far.

To your question: no bifolio rule is needed. Hand tables do it everywhere else. To get f58 right, the table should follow the language when hand and language disagree.

On neighbouring words: the link between the last glyph of a word and the first glyph of the next is in the tables, and it passes in #81 (tests 9 to 15). I have not looked at d- after specific endings separately.

As always with the skribe, it has no meaning, so this only says that writing habits are enough to reproduce the shift, not what d is.
Reply


Messages In This Thread
RE: Writing a fake Voynich by rules: copies, habits and the link between words - by quimqu - 12-11-2026, 04:30 PM

Possibly Related Threads…
Thread Author Replies Views Last Post
  Evidence for 2 rules linking words in Herbal A->B npcompl33t 30 7,146 08-10-2026, 05:37 AM
Last Post: npcompl33t
  Could synthetic “fake Voynich” tests tell us what we’re dealing with? Phoenixjn 4 4,852 10-09-2025, 09:19 AM
Last Post: dexdex

Forum Jump:


Users browsing this thread: 9 Guest(s)