Date: 19 July 2024
License: Creative Commons NonCommercial 4.0 International License
Author: DailyGregg Consortium
May all sentient beings experience health and tranquility!
YAGATS is a system that allows us to type romanized Gregg Shorthand outlines on an ASCII computer keyboard. This document lists the symbols used in YAGATS, briefly discusses the goals of the system design, and speculates about how a computer application might parse text written in YAGATS.
This is a reference document, not a tutorial. Learning to translate outlines into YAGATS from this document would be difficult. A friendly tutorial with examples and exercises might be created at some point in the future.
YAGATS (Yet Another Gregg-ASCII Transliteration System) uses "symbols" consisting of ASCII characters to represent the strokes and punctuation marks of Gregg Shorthand. The symbols are case-sensitive. There are approximately 73 symbols.
Some YAGATS symbols consist of two characters. These symbols are called "digraphs." Other symbols consist of a single character. For lack of a better word, such a symbol is called a "unigraph."
a the "a" stroke, the large vowel circle
e the "e" stroke, the tiny vowel circle
i the "long i" stroke, the indented circle
o the "o" hook
u the "oo" hook
s the "comma s" stroke
z the "left s" stroke
f the "f" stroke
v the "v" stroke
p the "p" stroke
b the "b" stroke
m the "m" stroke
n the "n" stroke
k the "k" stroke
g the "gay" stroke
r the "r" stroke
l the "l" stroke
t the "t" stroke
d the "d" stroke
c the "chay" stroke
j the "j" stroke
h the "h" dot
q the dot representing the suffix "-ing"
@ the dot representing the word "a" or "an"
A the dot that replaces the first 'a' as in Ahed "ahead" and Auak "awake"
w indicates that the next vowel has the "w dash" under it; kwek = "quick"
x the "x" stroke (a modified comma s)
X the other "x" stroke (a modified left s)
5 the "ses" blend (left s + comma s)
3 the other "ses" blend (comma s + left s)
y represents the -ally/-ily suffix loop
: colon indicates a disjoined stroke, for example, bro(:d = "brotherhood"
% percent sign indicates an unusual joining or special turning as in
a%d = "I had" or o% = "all" (in Simplified)
' ASCII apostrophe indicates an apostrophe in the shorthand
, ASCII comma indicates a comma in the shorthand
! ASCII exclamation point indicates an exclamation point in the shorthand
# appears at the beginning of a word which is written in longhand, is a
numeral, or is any other kind of non-YAGATS string
~ tilde indicates the "jog" as seen between m and n in "I am not"
(writing the tilde is mandatory when m/n or t/d are involved, even though
the jog is required by theory and thus is predictable)
- ASCII hyphen represents shorthand hyphen
= ASCII equal sign indicates capitalization mark
Warning: the following symbols have to be escaped in regex (ie preceded with a backslash\)
( the "over ith" stroke
) the "under ith" stroke
$ the "ish" stroke
^ caret indicates a high floating prefix e.g. o^du = "overdue"
* asterisk indicates an intersecting stroke; p*m = "p.m."
. ASCII period (full stop) represents shorthand period
? ASCII question mark represents shorthand question mark
> ASCII "greater than" sign indicates the shorthand end-of-paragraph symbol
{ ASCII left curly bracket represents shorthand left parenthesis
} ASCII right curly bracket represents shorthand right parenthesis
au the a circle + oo hook diphthong as in "now"
ea the circle with a dot inside as in "panacea"
eu the e circle + oo hook as in "youth"
io the circle within a circle as in "lion"
oe the o hook + e circle as in "joy"
ya the ya- loop as in "yarn"
ye the ye- loop as in "yell"
NT the nd/nt blend as in "rend" and "rent"
TN the tn blend as in "certain"
MD the md blend as in "named"
DM the dm blend as in "random"
MN the men blend as in "many"
TD the ted blend as in "rated"
NG the ng stroke as in "sing"
NK the ngk stroke as in "sink"
LD the ld blend as in "cold"
RD the rd blend as in "cord"
DV the def/tive blend as in "definite" and "restive" (Simplified)
JD the jent/jend/pent/pend blend with j sound as in "gentle" (Simplified)
PT the jent/jend/pent/pend blend with p sound as in "repent" (Simplified)
xz xes (beginning with modified comma s)
Xs xes (beginning with modified left s)
duu ned t TDa? a'm$ur tu rn akroz t b akseTNt DMoro. Do you need it today? I'm sure to run across it by accident tomorrow.
(1) YAGATS currently does not have a mechanism for notating the "reversing principle" used in some pre-WW2 editions of Gregg. The backslash (\) appended after the reversed vowel is under consideration, e.g., "a\".
(2) Currently there is no mechanism for representing "diacritical marks" such as the long vowel indicator often written under the o-hook when it stands for the exclamation "oh!"
(3) When a US outline has a different UK counterpart, e.g., "schedule," can the UK outline somehow be included in the same line of the lexicon file, or must a separate lexicon file for a so-called "UK dialect" be created?
The design of YAGATS is a compromise between legibility and ease of learning for human users versus simplicity of parsing for computers. A system consisting entirely of one-character symbols would make it slightly easier to write parsing applications but would require human users to memorize bizarre and arbitrary symbols.
Many of the symbol allocations were obvious choices: "b" for the b stroke, "au" for the a+oo diphthong, and so forth. After the obvious choices were made, it was necessary to assign symbols to the remaining graphemes of Gregg Shorthand.
Assume you have opened a text file containing many shorthand sentences written in YAGATS notation. To parse this material (in order to translate it to longhand, or to draw the outlines, or to calculate symbol frequencies, for example), the following steps would be taken: