Re: draft response for LC comment 51 RM1 and 62 JM1

Let me be clearer about my position on this.  Alan wrote:

> Is the only problem loops in the definitions? Could it be fixed by
> another global restriction? I'm concern that "we didn't think much
> about it" might not be good enough. Better to have thought about it
> and ruled it out beforehand, or to think about it now in response to
> the comment?
>
> -Alan
My understanding is that while [1] describes this looping problem in the
end of section 3.2, it also describes a solution: a requirement that the
datatype naming axioms be acyclic.  So unless someone can make a
convincing argument that such a restriction is unworkable, I will
object to responding that the looping problem justifies dataranges
being unnamed and will object to leaving named dataranges out of
the language.

-Evan

[1] http://web.comlab.ox.ac.uk/people/Boris.Motik/pubs/mh08datatypes.pdf

Received on Tuesday, 3 March 2009 20:05:06 UTC