bug#11460: 24.1.50; doc string of `assert'

[Drew Adams]

> FWIW, this part of `cl-assert' is not only poorly documented, but
> it's poorly designed.  And it doesn't have the excuse that it's
> inherited from Common Lisp.
>
> So I'd rather not document that mis-feature in any more detail.

In that case, my request would be that the entire bug be recorded
and kept open for future fixing (e.g., todo list or whatever):
design + implementation + doc.

IOW, let's not use the fact that the design, in addition to the
doc, is faulty as an excuse to just drop this.  If it needs
fixing and you cannot or do not want to tackle that now, then
it should be kept open for fixing later.