[Axiom-developer] Hyperdoc

[daly]

Ralf, Bill,

> Use the tools at hand and build something that can actually compete
> with current CAS (plural). We fight with the future and haven't even
> reached the present.

This isn't a competition, at least in my mind. My view is that this
is the early development of the science of computational mathematics.
Thus I see no reason to try to compete with existing systems. Besides,
MMA and Maple clearly have better resources. Chasing them is a never
ending task. The science is bigger than they imagine.

But why should we bother thinking about an "assume" facility, which is
clearly wrong, when we can do research in provisos? Why worry about
matching their symbolic (as opposed to algebraic) abilities when there
are clear advantages to an algebraic approach? Why encourage the
construction of more dead code when it is clear that code needs
extensive documentation to live? Why not raise our standards to
struggle with proving algorithms correct rather than handwave at the
problem? Why not change the expectations so that computational
mathematics papers are expected to be literate?

The future need not be a linear projection of the past.
Henry Ford didn't breed faster horses.
We can shape our future in fundamentally new ways.
Look to the 30 year horizon and imagine.

Tim