- plus
- minus
- times
- divide
- divisible-by
- equal
- not-equal
- smaller-than
- greater-than
- smaller-or-equal
- greater-or-equal
- power
- square-root
- root
- ellipsis
- vertical-ellipsis
- pair
- tuple
- percent
- ratio
- repeating-decimal
- point
- gcd (greatest-common-divisor)
- lcd (least-common-denominator)
- absolute-value
- limit
lim
- tends-to
- defined-as (line defined via equation, e.g.
l : 2x + 1 = 0
) - index
- circled (used around
<mn>
numbers in "successive remainder" of a derivation) - imaginary-unit
- infinity
- evaluated-at
- partial-derivative
- first-derivative
- second-derivative
- third-derivative
- derivative
- differential
- open-interval
- closed-interval
- open-closed-interval
- closed-open-interval
- integral
- indefinite-integral
- definite-integral
- summation
- factorial
- combination
- probability
- set
- intersection
- union
- sine
- cosine
- tangent
- cotangent
- secant
- cosecant
- arcsine
- arccosine
- arctangent
- arccotangent
- arcsecant
- arccosecant
- hyperbolic-sine
- hyperbolic-cosine
- hyperbolic-tangent
- hyperbolic-cotangent
- hyperbolic-secant
- hyperbolic-cosecant
- ray
- directed-line-segment
- segment
- line
- angle
- inverse
- name (do we use
_($piece1,$piece2,...$piecen)
orname($piece1,...)
?) - logarithm
- natural-logarithm
- pi
- cross-product (vector-product)
- defined-as
- vector
- magnitude
- scalar-product (dot-product)
- polar-coordinate
- determinant
- binomial-coefficient
- congruent
- triple-of-direction-cosines (just triple?)
- floor
- ceiling
- euler-number
- Q: list-of-lists? e.g.
(1,6; 6,1; 2,5; 5,2; 3,4; 4,3)
- del-operator (used for gradient, diverge and curl, wiki)
- list-separator
- time-separator
- interval-separator
- where-separator (
:
such-that?)
- some uses of
↔
and→
which had unclear terminology, used in physics relationships- would have been
equilibrium
andyields
in chemistry - or potentially
if-and-only-if
andmaps-to
in mathematics - but I am not sure what the physics nomenclature is
- would have been
-
foot
ft
-
pound
lbs
-
radian
rad
-
meter
m
-
kilometer
km
-
centimeter
cm
-
hour
hr, h
-
year
yr
-
minute
min , ′
-
second
s, ′′
-
dollar
$
-
degree
°
-
kelvin
K
-
celsius
C
-
fahrenheit
F
-
mile
mi
-
east
E
-
west
W
-
north
N
-
south
S
-
per
s^{-1}
( "per second"? discussion) -
Appendix B: table of 24 unit conversions
-
Appendix C,E: physical constants
-
Greek prefixes:
- atto, femto, pico, nano, micro, milli, centi, deci, deka, hecto, kilo, mega, giga, tera, peta
- fermi
fm
- farad
F
- microfarad
μF
- coulomb
C
- volt
V
- kilovolt
kV
- ampere
A
- watt
W
- joule
J
- newton
N
- ohm
Ω
- hertz
Hz
- siemens
S
- henry
H
- milligram
mg
- microampere
mA
- atomic-mass
{}^{227} Ac
- isotope
C^{14}
- :quotient
- :system-of-equations
- Note: there is also a curious use of a matrix with a vertical border writing only the
coefficients in each
<tr>
, the rest being understood
- Note: there is also a curious use of a matrix with a vertical border writing only the
coefficients in each
- :group
- parentheticals? Or is it intent values of paren-group, bracket-group, brace-group,...
- :charge (e.g.
+q
in physics) - :permutation
- :matrix
- :chemical-formula
- :unit
The first four terms should always work for the concept name, but the last one is poor MathML because Celsius is not an operator, and in general, units are not operators. Actually, several of them are poor MathML (see later).
The
:unit
cases are more of a judgement call as to what we want to say. MathCAT goes through a repair process. It likely would repair something with<mo intent=":unit">
to be anmi
and probably also repair the two character°C
to be the single character℃
. I suspect I would also repair the<mrow>
to be the single charmi
while I was at it. The goal of the repair is present the matching rules with a single thing to match against. I doubt other software would do the same. So the judgement call is whether we expect generating software to generate just one of the forms or whether we require the consuming software to do more work and consume all of the possible forms including this one not mentioned:and some not mentioned that "incorrectly" use
mtext
(which happens because that's a way to get a non-italic 'C'). I think it comes down to whether we want to be more strict in what is appropriate MathML when usingintent
properties or whether we want to continue the very laissez faire attitude that MathML has traditionally had.Not to nitpick, but the MathML you used isn't really correct. When
mi
is used, it should be (for example)Also, when
°
is used, it should be a superscript . So that's even more things that potentially we might require AT to recognize.