Re: UNSAID - two test cases (dawg:unbound, issues#useMentionOp)

On Thu, Dec 23, 2004 at 10:15:24AM +0000, Steve Harris wrote:
> 
> On Wed, Dec 22, 2004 at 02:53:47 -0800, Pat Hayes wrote:
> > >On Wed, Dec 22, 2004 at 05:26:01PM +0000, Andy Seaborne wrote:
> > >> >Yes, but SQL for eg. has tri-value logic (true, false and NULL), so you
> > >> >can meaningfully apply operators and functions to unbound values (NULL).
> > >>
> > >> It doesn't quite work out that simply.  It's fine for operators and
> > >> functions but pattern matching isn't so straight forward.
> > >>
> > >> OPTIONAL (<x> ?p ?o)
> > >>  (?o ?q <y>)
> > >>
> > >> so ?o may be NULL then we have the (?o ?q <y>) and it needs to handle
> > >> ?o = NULL differently.  NULL is different.
> > >
> > >Yes, bun in RDF you cant have a triple like (NULL ?q <y>), so that match
> > >will always fail. Unless I'm missing something.
> > 
> > Er...seems to me that you are missing a use/mention confusion. NULL 
> > here is supposed to be a value (right?), not a piece of syntax. So 
> > what does it even mean to put NULL into a triple? Or are you 
> > suggesting that 'NULL' - that is an actual piece of syntax, a 
> > four-letter character string - is used as a kind of dummy binding to 
> > stand for the binding in cases where a variable doesn't have a 
> > binding? If so, I think this really is a very bad idea. All sorts of 
> > problems will arise right away: eg two different unbound variables 
> > will be the same. It would be much better to simply leave the 
> > variable in the pattern in cases like this.
> 
> I was doing a subsitution, as in:
> 
> ?x is unbound, (?x ?q <y>)
> 
> We dont have a symbol for unbound, so I co-opted NULL. Old habbits die
> hard.
> 
> BTW, in your proposed expressions involving unbound values are false
> scheme, would !(?x == ?y) be true or false, if ?x is unbound. I'm not
> clear on how far up the expression tree unboundedness propagates.
>  
> > > > Talking about NULLs, with all it special cases for matching and function
> > >> handling, like NULL != NULL, is no different to talking about unbound
> > >> variables.  Both need special handling.
> > >
> > >There are no special cases. Any arithemtic operation involving NULL is
> > >NULL, so NULL == NULL is NULL, NULL > 3 is NULL, ...
> > 
> > Does NULL match NULL? If ?x and ?y are both bound to NULL, are the 
> > bindings the same? And so on.
> 
> Not in SQL. I not holding that up as a shining example of rightness
> though.

In algae2, NULL != NULL .

Algae2 preserves query order when performing a query. To that end,
optionals either bind the affected variables to either a matched value
or NULL. Affected variables are those mentioned in the pattern
(including nested patterns) that are not yet bound. All the test cases
that I dreamed up bahaved intuitively when comparing NULLed variables.

What's the difference between a syntactically differentiated NULL
binding and a binding which is conscientiously omitted from the
results?
-- 
-eric

office: +81.466.49.1170 W3C, Keio Research Institute at SFC,
                        Shonan Fujisawa Campus, Keio University,
                        5322 Endo, Fujisawa, Kanagawa 252-8520
                        JAPAN
        +1.617.258.5741 NE43-344, MIT, Cambridge, MA 02144 USA
cell:   +81.90.6533.3882

(eric@w3.org)
Feel free to forward this message to any list for any purpose other than
email address distribution.