chart:equation, ver. 6 10.14.2, ver 7.01 10.15.1

[Patrick Durusau <patrick@durusau.net>]

Greetings!

I ran into this while trying to solve the attribute position problem I 
posted earlier today:

The chart:automatic-content attribute is a boolean value attribute on 
the <chart:equation> element.

We currently say:

"The |<chart:equation>| element represents the equation for the 
containing regression curve. This element can contain fixed text. When 
no fixed text is provided or |chart:automatic-content| attribute is set 
to |true| the current calculated equation is displayed automatically."

The content model (yes, from one of the schema fragments but wait and see):

<define name="chart-equation">
    <element name="chart:equation">
        <ref name="chart-equation-attlist"/>
        <optional>
            <ref name="text-p"/>
        </optional>
    </element>
</define>

So, we can have "fixed text," which I assume goes in "text:p" (if you 
expand the reference) or the results of the equation, if the attribute 
(you have to expand chart-equation-attlist, which is why I dislike help 
the editor type schema fragments) to see: chart:automatic content.

Oh, wait, err, where is the equation? We say that the <chart:equation> 
element represents the equation. Hmmm, does the equation go inside text:p?

If so, then how do we distinguish it from "fixed text?"

Please note a second issue associated with this element (and the path by 
which I got here):

We say for the chart: automatic-content attribute that:

"The |chart:automatic-content| attribute indicates whether a user 
defined text should be kept or the current content should be calculated 
automatically."

If the element represents an equation, it seems to me that the "content" 
from that equation is always: 1) automatically generated, or 2) stored 
for applications that cannot process the equation.

The notion of "user defined text" appears to be ill-considered.

My suggestion would be:

1) Define an attribute that specifies the equation for the 
chart:equation element and an attribute that stores the results of 
processing that equation. (I haven't checked to find attributes we can 
reuse for this purpose.)

2) The semantic of chart:automatic-content is as follows: true, the 
equation attribute is processed, if the application supports it, if 
false, the stored result, if available, is displayed. If the attribute 
is true and the application doesn't support processing and the stored 
result isn't available, then the application displays (result not 
available).

I think that covers all the cases but I am not sure. I think it is 
important to cover all the cases. Undefined behavior is undesirable.

3) I suspect the <text:p> element was actually meant to hold a static 
graphic specified by the user but on the other hand, text to accompany a 
chart may have also been meant. It simply isn't clear. If it does mean 
user supplied text, then it had better specify where it is rendered in 
relation to the graphic, etc.

I don't have a good suggestion for that aspect of this particular issue.

Apologies for the length.

Having said that, I do think once we get them resolved that the standard 
will be more consistent, concise and generally easier to implement. 
Well, or at least implementers won't have to wonder what we may have 
meant. They may curse us for what we require but not for being vague. ;-)

Hope everyone is at the start of a great week!

Patrick

-- 
Patrick Durusau
patrick@durusau.net
Chair, V1 - US TAG to JTC 1/SC 34
Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)