Hi. I'm hoping that some of you reading this will be able to offer up some
advice and recommendations.

I need a database that runs under MS Windows 3.x (or 95). My criteria are

    - User friendly (although it doesn't need to be drool-able) interface

    - Capable of handling a large number of rows. Maybe 3-4 million
      per table (where a row has 11 columns, with roughly 160 bytes of
      data per row).

    - Can be queried remotely. In particular, it should be possible to
      have Perl scripts running on various Unix hosts query this database
      (and add to it - hence the cross-post to comp.lang.perl). It would
      be a bonus if it was trivial for Mac based Excel could query it as
      well.

    - Automatically filter duplicate rows, preferably not storing them at
      all, without the need to do a "select distinct from. . ." style
      query every now and then.

NT based solutions would also be OK.

Has anyone got any real-world experiences or recommendations they could
pass on? I've been badgered by salesmen who will just repeat "Of course
it does that sir. No, we won't guarantee that it does precisely that,
but I'm sure it does. . ."

Followups set to poster to (a) ensure I don't miss replies, and
(b) hopefully minimise holy wars. I'll summarise responses (and, later
my experiences with the product of choice).

Cheers,

N
--
--+=[ Blueberry Hill                   Blueberry Design                   ]=+--
--+=[ http://www.*-*-*.com/ ;    1/9 Chelsea Harbour Design Centre, ]=+--

--+=[                      Have a banana. . .                             ]=+--