Re: Chemistry call on December 12 at 16 Important information UTC

Elaine,

Thank you very much for your comments. Some responses inline...

On Mon, Dec 9, 2019 at 2:54 AM E.A.Moore <e.a.moore@open.ac.uk> wrote:

> Comments on MATHML semantics document
>
>    1. Units. I was taught and we teach our students that variables should
>    be italic, but constants and units should be upright.
>
> Very true. MathML 3 lacks an official (normative) way to express units,
although there is a note that suggests using a class to indicate it is a
unit. As you point out, they are not only semantically different, but are
rendered in a upright, non-italic font. Because a class is used, one can
style with CSS that they are always in an upright font. Also, although
single characters are rendered with italics by default in MathML, multiple
characters are not. So for units such as "l", "m", and "g", one either
needs to use the class in MathML 3 and style it or use <mi>s attr:
mathvarient="normal".

>
>    1. I remembered I had been to a lecture in the 1990s on a chemistry
>    mark up language. Apparently this is still in existence, though I have
>    never used it. See https://www.xml-cml.org/ . They have a dictionary
>    and a complex list indicating whether the item is an atom, molecule,
>    chemical structure.... You may know about this?  The authors are Henry
>    Rzepa, Imperial College, London and Peter Murray-Rust, University of
>    Cambridge.
>
> CML (it's abbreviation) is very powerful and can represent molecules,
reactions, solid-state, computation and spectroscopy. I am told (but have
not verified) that it says to use MathML to write chemical formulas. CML is
much more detailed in some sense

>
>    1. If sodium is written Na, I would recommend saying N a, even though
>    we would sometimes say sodium.
>
> MathPlayer tries to intuit when something is Chemistry markup. The subject
attr, if used, will remove the need to guess. MathPlayer presents *users* the
option of having chemicals spoken as their full name ("Sodium") or their
abbreviation. There is a way for a page author to override an user
preference for testing purposes. "subject" and "mathrole" are meant to
convey some semantics, not force specific speech. Having said that, if
people think that forcing specific speech is important in some contexts,
then we (the MathML refresh group) need to know that so we design MathML 4
to accommodate that requirement. I encourage anyone with opinions on this
to chime in.

>
>    1. The symbols for sulfite and phosphite should have the charge as a
>    superscript. There is a IUPAC naming system for these ions in which sulfite
>    for example, becomes Sulfate(IV) pronounced sulfate 4; the IV indicating
>    the oxidation state of  sulfur. IUPAC also has a project Standard XML Data
>    Dictionaries for Chemistry,
>    https://iupac.org/projects/project-details/?project_nr=2002-022-1-024
>
> I'm not clear how the IUPAC work fits in with the MathML effort. I do see
it being useful for actually speaking though. Based on your mention of
sulfite and phosphite,  I have added a few more examples to the google doc
of notations that need to be supported.

Thanks for your feedback. It is very helpful,

    Neil