On 03/18/2013 06:21 AM, Oliver Ruebenacker wrote:
Hello David,
On Sun, Mar 17, 2013 at 9:52 PM, David Booth <da...@dbooth.org> wrote:
On 03/17/2013 10:55 AM, Oliver Ruebenacker wrote:
Hello,
On Sat, Mar 16, 2013 at 12:30 PM, David Booth <da...@dbooth.org> wrote:
You are in good company in thinking that a URI always denotes the same
resource, because that is a widespread misconception. (I call it Myth #1
in
http://dbooth.org/2010/ambiguity/paper.html .) But it simply is not true
in
the RDF semantics.
[...]
That is precisely why it is helpful to keep different perspectives in
different graphs, as Jeremy suggested.
That's a little bit like saying, since floating-point numbers are
not perfectly precise, the 1.376 in my data may not be the same as the
1.376 in your data, and therefore the two values should be kept in
separate spaces.
I don't follow what you mean. A floating point number like 1.376 may be
used as an approximation of some real number, but it is exactly the same
floating point number in all RDF graphs.
I'm not talking about how RDF handles floating point numbers. I'm
just using floating point numbers as an example to show how flawed
your logic is. If your logic would hold, that URIs have no meaning
without context, because they are only approximate, then by the same
logic, it would follow that floating-point numbers have no meaning
without context.
I'm sorry, but I still do not understand your point. I have been
talking very narrowly about existing RDF Semantics. It sounds like you
are talking much more broadly about context, but I don't know what you
mean. Clearly 1.376 has no meaning without context. To interpret it it
correctly, you need to know that it is supposed to represent a number,
it is represented in base 10, etc. It is completely meaningless without
knowing how to interpret it. What is the analogy that you are trying to
illustrate?
David
