# ISSUE-10: Algebraic differentiation operator

## Algebraic differentiation operator

- State:
- CLOSED
- Product:
- All
- Raised by:
- Michael Kohlhase
- Opened on:
- 2007-02-08
- Description:
- In the diff operator, we used a bvar schema to specify the bound variables. But

in the new canonical MathML, we only allow bvar in bind (which carries the

expectation of alpha-equivalence). Thus we cannot make algebraic diff a binding

operator, and have to pass the differentiation variable as a regular argument

(i.e. no bvar

For seeing how things work out in the concert of things, I have added an aDiff

symbol to the content dictionary calculus_veccalc for algebraic differentiation.

Have a look, how it works.

I am not sure the naming is good, or whether we should have two CDs, one for

algebraic-, and one for function-operations. I kind of like this idea, since

there are other algebraic/function distinctions here, and we have to think about

CD grouping for K-14 anyways. - Related Actions Items:
- No related actions
- Related emails:
- ISSUE-10: Algebraic differentiation operator (from dean+cgi@w3.org on 2007-02-08)

## Related notes:

Again, it is possible to map this to different elements of the OpenMath

dictionary depending on its structure. The structures do not need to be in 1-1

correspondene. There just has to be a 1-1 mapping from some structure in

MathML to OpenMath.

Having said this, I am open to arguments that users \"EXPECT\" a different

operator for algebraic operations. If so, one should be supplied, but the

justification is different.

Also, if we introduced it, we might need to deal with legacy issues. deprecation

is a possibility.

On the whole, I think it would make the notation easier to use and understand.

(fewer surprises)

I think that we have satisfactorily dealt with this using lambda, closing this issue

Michael Kohlhase, 8 May 2008, 06:18:29Display change log