After a dataset has been populated, it no longer reflects changes in
database. It is the shortcoming I found in ADO.NET. In ADO, we have dynamic
recordset, keyset recordset which can reflect the changes to database done by
other users immediately. Why can't I find these useful stuff in ADO.NET?

Re: No cursor when compared with ADO by Stephany

Stephany
Sat Apr 02 03:47:54 CST 2005

It would appear that you have not done any reading abour ADO.NET. If you had
you would know that ADO.NET and ADODB are completely different animals and
cannot be compared to one another.

If you wish to do so, you are quite free to use ADODB and, if you do, you
will have all your 'legacy' functionality.

I do recommend, however, that you take the time to study ADO.NET more deeply
and I think that you will find that is is far more flexible than you seem to
think at he moment.


"Samuel1234" <Samuel1234@discussions.microsoft.com> wrote in message
news:EC3A0F7F-667F-40E5-8F6D-8EF1CB36C0C0@microsoft.com...
> After a dataset has been populated, it no longer reflects changes in
> database. It is the shortcoming I found in ADO.NET. In ADO, we have
> dynamic
> recordset, keyset recordset which can reflect the changes to database done
> by
> other users immediately. Why can't I find these useful stuff in ADO.NET?



Re: No cursor when compared with ADO by Frans

Frans
Sat Apr 02 03:49:28 CST 2005

Samuel1234 wrote:
> After a dataset has been populated, it no longer reflects changes in
> database. It is the shortcoming I found in ADO.NET. In ADO, we have dynamic
> recordset, keyset recordset which can reflect the changes to database done by
> other users immediately. Why can't I find these useful stuff in ADO.NET?

Because an open server side cursor is very bad for scalability. Say you
have a database server and a desktop application installed on 1000
desktops targeting that database server. Without serverside cursors,
that's easy. With server side cursors, the database server CAN have at a
given moment 1000 open resultsets. That's not that great for the
database server.

It's also unnecessary. After all, the data you're viewing on the screen
is a snapshot of the data at time T. It's stale the second you fetch it.

FB

--
------------------------------------------------------------------------
Get LLBLGen Pro, productive O/R mapping for .NET: http://www.llblgen.com
My .NET blog: http://weblogs.asp.net/fbouma
Microsoft MVP (C#)
------------------------------------------------------------------------