Double Pivot

I need some help with a double pivot problem. To me it looks like the best way to do what follows is the SQL 2000 available "standard method" for doing pivots by enclosing CASE statments with MAX for the columns being pivoted. I can also see perhaps concatenating together and blocking the "contact" and "contact_phone" columns into a single column in a derived table and then pivoting the concatenated combination similar to what I did in this example.

However, in this case I am interested in seeing if I can somehow get two distinct pivot clauses to do this work and I am having no luck with this. I have this exmple, but it looks pretty cruddy:




Code Snippet
-- --------------------------------------------------------------------------
-- Data for this problem is stored in table @support and consists of
-- (1) application_name
-- (2) support_role -- whether the contact is the primary or secondary
-- support associate
-- (3) contact -- the name of the support associate
-- (4) contact_phone -- the phone number of the support associate
--
-- The problem is to pivot the contact information and output columns:
-- (1) Application Name
-- (2) First Contact -- the name of the primary contact
-- (3) First Phone -- the phone number of the primary contact
-- (4) Second Contact -- the name of the secondary contact
-- (5) Second Phone -- the phone number of the secondary contact
--
-- This method uses two separate PIVOT clauses to pivot the data into
-- columns. This looks pretty cruddy.
-- --------------------------------------------------------------------------
declare @support table
( application_name varchar(20),
support_role char(1),
contact varchar(10),
contact_phone varchar(14)
)
insert into @support
select 'Clean', 'P', 'Rick', '(904) 555-1212' union all
select 'Buggy', 'P', 'Jim', '(217) 555-1212' union all
select 'Buggy', 'S', 'Chris', '(309) 555-1212' union all
select 'New', 'S', 'Rick', '(904) 555-1212'
--select * from @support

select application_name,
max(isnull(p,'')) as [First Contact],
max(isnull(xp,'')) as [First Phone],
max(isnull(s,'')) as [Second Contact],
max(isnull(xs,'')) as [Second Phone]
from
( select application_name,
P, S, xP, xS
from
( select application_name,
support_role,
contact,
contact_phone,
'x' + support_role as support_role2,
contact as contact2,
contact_phone as contact_phone2
from @support
) as x
pivot ( max(contact) for support_role in ([P], [S])
) as p1
pivot ( max(contact_phone2) for support_role2 in ([xP], [xS])
) as p2
) y
group by application_name

/* -------- Sample Output: --------
application_name First Contact First Phone Second Contact Second Phone
-------------------- ------------- ---------------- -------------- ----------------
Buggy Jim (217) 555-1212 Chris (309) 555-1212
Clean Rick (904) 555-1212
New Rick (904) 555-1212
*/




Is there a way to get this query to work better or am I just better off using the SQL 2000 "Standard Method"?