W3C home > Mailing lists > Public > www-math@w3.org > May 2000

Comments on MathML Last Call Draft (cont'd)

From: Pankaj Kamthan <kamthan@cs.concordia.ca>
Date: Sun, 7 May 2000 17:57:55 -0400
To: www-math@w3.org
Message-ID: <3915AEA3.21790.A08A74@localhost>
Here are some comments on Appendix C of the MathML 2.0 Draft.
Hope they are useful.

Pankaj Kamthan
--

GENERAL

1. Differences in case. Some with <reference> and others with <Reference>. 
Examples: Compare C.2.3.2 <reference> M. Abramowitz and I. Stegun, 
Handbook of Mathematical Functions, [4.2]</reference> and C.2.5.1 
<Reference> M.&nbsp;Abramowitz and I.&nbsp;Stegun, Handbook of 
Mathematical Functions, [4.1] </Reference>. Same with use of <name> and 
<Name>. This should perhaps be consistent.

2. Some with &nbsp; and others not. See 1. This should perhaps be 
consistent.

3. In several MMLdefinition's, "periods" and "commas" are missing from 
appropriate places. 

4. In several MMLdefinition's, both name and description is given (C.2.2.10); 
in others none of them are given (C.2.3.6). This should perhaps be 
consistent.

5. According to C.1.2, the "property" and "example" elements use MathML 
syntax. Why is "reln" continue to be used when it is deprecated in MathML 
2.0? Example: C.2.1.1 and C.2.2.4, respectively.

6. Compare
C.2.2.3 ... <MMLattribute> <name>definitionURL</name> <value> URL 
</value> 
<default> none </default> </MMLattribute> ...
and
C.2.3.2 ... <MMLattribute> <attname>definitionURL</attname> <attvalue> 
CDATA </attvalue> <attdefault> none </attdefault> </MMLattribute> 
<MMLattribute> <name>type</name> <values> any MathML Type </values> 
<default>real</default> </MMLattribute> ...

a. Why do we have (different) names like <attname>, <attvalue> and 
<attdefault> in the second case? This should perhaps be consistent.

b. How come their values are different? First case: URL; second case: 
CDATA.

c. <values> -> <value>

These problems seem to be generic (occurring in more than one sections).

SPECIFIC

C.1.2
... as a child of the MMLDefinition at ... -> ... as a child of the 
MMLdefinition at ...

C.2.2.5
definitionURL="www.w3c.org/MathML/Content/arcsin" -> 
definitionURL="http://www.w3c.org/MathML/Content/arcsin"

C.2.5.2
The part "Also called the natural logarithm. The inverse of the 
exponential function." is incorrect. (It applies to ln(x).)

C.2.2.8
1. delcared -> declared
2. ... <MMLattribute> <name>definitionURL</definition> <value> Any valid 
URL </value> </MMLattribute> ... Missing end tag for "name".

C.2.2.9
"Lambda is a binary function, where the first argument is the variable 
and the second argument is a the expression." ->
"Lambda is a binary function, where the first argument is the variable 
and the second argument is the expression."

Why does the functorclass say "Nary"?

C.2.2.10
1. etc. . -> etc.
2. <forall> -> <forall/> 

C.2.3.1
<reln/> -> <reln> and need the corresponding end tag.

C.2.3.3
integerl -> integer

C.2.3.23
compex -> complex
<property>???</property> 

C.2.3.24
<ci>a</cn> Tag mismatch.
<ci>&epsilon</cn> Tag mismatch and ";" missing from the general entity. 
<ci><mrow><msup><mi>a</mi><mi>b</mi><mrow></cn> Tag mismatch for 
ci. End 
tag for msup missing.
Why are C.2.4.3, C.2.4.4, C.2.4.5, C.2.4.6 given a commutative property? 
gt, lt, geq, leq aren't commutative. (C.2.4.5 rightfully questions that.)

C.2.5.2
Why isn't 
<property> log(1) = 0 </property> 
provided as a property (like C.2.5.1)?

C.2.5.3
areguments - > arguments
"<signature> (algebraic,bvar,interval) -> algebraic </signature>"
Why only interval? We can have (presumably measurable) sets. See 4.4.5.1.

C.2.5.5
(Binary) -> Binary

C.2.5.8
quantifiy -> quantify 

C.2.5.9
"<property> ... </property>"
If there is nothing in it, this line could be deleted. Same for C.2.6.8, 
C.2.6.9 C.2.6.11, C.2.8.7, and several others. The same idea for 
"<signature> ... </signature>"
in C.2.7.1, C.2.7.2.

C.2.6.3
"<signature> (set*) -> set </signature>"
Shouldn't it be (set+ set) -> set? (I am assuming: * := 0 or more; + := 1 
or more.) The same for C.2.6.4.

C.2.6.7
Why is it (set*) -> boolean here and (set, set) -> boolean in C.2.6.8? 
Shouldn't they be same (to be (set+ set) -> boolean)? Same argument for 
C.2.6.9 and C.2.6.10.

C.2.8.1
The line 
"<Reference> ditto, [4.3.27] </Reference>"
could either be removed or should be added to other trigonometric functions 
C.2.8.2, ...) as well. What "ditto" means could be filled then.

C.2.9.1
dimenions -> dimensions 
<values> random_variable | continuous_random_variable | data </value>: 
Tag mismatch.

Same problems with C.2.9.2, C.2.9.3, C.2.9.6.

C.2.9.4
discription -> description 

C.2.10.6
<signature>(matrix)->scalar* </property>: Tag mismatch.
Received on Sunday, 7 May 2000 17:58:52 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:49 GMT