[Om] Accessing arguments of an n-ary operator as indexed list items (in a definition)

[Arjeh Cohen]

Dear Christoph,

There is one more advantage of using sequences that I recall:
in a text on mathematics, often mentions sequences x1, x2, ...,xn.
Again, these are distinct from lists.

Best, Arjeh



On Sat, May 09, 2009 at 06:10:47PM +0200, Christoph LANGE wrote:
> Dear Arjeh,
> 
> On Saturday 09 May 2009 16:43:02 Arjeh Cohen wrote:
> > In the good old day, I used sequence rather than list
> > for the same purpose. See
> > http://www.win.tue.nl/~amc/oz/om/cds/sequence1.xml.
> 
> Aha -- hmm, maybe that should be considered for the MathML 3 spec.  It does
> not seem to conflict with what's already said there, as the spec only mentions
> apply_to_list, but not how to construct a list out of arguments.  And the
> latter, plus the additional possibilities of how to _deal_ with the elements
> of such a list, might be useful.
> 
> > The reason for choosing this slight variation of list is that
> > I thought of a list as [x1,x2,....,xn]
> > and of a sequence as x1,x3,...,xn.
> > So, in accordance with
> > f(x1,x2,...,xn) we write f(a) where
> > a is the sequence x1,x2,...,xn rather than the correspondinglist.
> 
> Ah, so you would also recommend to think of a "sequence of x1,...,xn" when
> introducing n existentially quantified variables?
> 
> > There are operators to go back and forth from one to the other.
> >
> > I tried to see what exists of this on the openmath website, but that
> > one seems down.
> 
> You mean the whole website (is up, as I can see), or the sequence1 CD?  Right,
> I don't see the latter in the svn repository -- no idea what happened to it…
> 
> Cheers,
> 
> Christoph
> 
> -- 
> Christoph Lange, Jacobs Univ. Bremen, http://kwarc.info/clange, Skype duke4701
>