about changing anything at all (Was: Re: OpenMath floats)

[Andreas Strotmann]

David Carlisle wrote:
 
> Any change to any part of the language will potentially break the
> existing libraries, I don't say that we should never do that (there are
> some changes I'd like to see in the XML encoding....) but I think that
> we need to have some idea of the amount of extra work any change would
> make for the library implementors, and then just consider each case on
> its merits.
 
Someone (JHD, I think) made a point during the workshop in Eindhoven that
what we should legitimately be concerned about is changes that invalidate
existing OM (encoded) *objects*, while existing *libraries* should (at
least to some extent) be fair game for changes (ideally, existing
applications would just link in a new version of the library to become
compliant with a new version of the standard).

Thus, for example, the vote for making OpenMath a full XML application
would not invalidate any existing OM object in the XML encoding -- it
would only make some other variations on the theme legal OM XML encodings.
The INRIA library's internals, however, would have to be changed to
reflect the new degrees of freedom, but other libraries we heard about had
already used an XML parser as a subroutine library in their code, and that
practice would now be legitimized (besides providing a more
well-structured piece of code).

At this point in the development of OM I would prefer what has been called
"conservative" changes: ones that keep older versions of OM-encoded 
objects valid but may add new features that old libraries may not
necessarily understand.  I would argue that adding the necessary
subroutines for handling conservative extensions to a
well-designed library should be simple.

In particular, this would mean that current encoding practices may become
"depracated" in a next version, but not "illegal" -- new software would
still need to be able to read depracated encodings, but should not produce
them.

 -- Andreas