Forum Replies Created

Viewing 14 posts - 181 through 194 (of 194 total)

  • RE: Design

    sjimmo (9/24/2010)


    Hugo Kornelis (9/24/2010)

    --------------------------------------------------------------------------------

    if the table is used in replication, ..., SQL Server requires that a surrogate key with the UNIQUEIDENTIFIER data type is also present in the table.

    I...

  • RE: Design

    Hugo Kornelis (9/24/2010)


    The more I think about this, the more I disagree with the notion that multiple identifier columns is a design flaw.

    Agreed. And the SQL standards agree too, if...

  • RE: Design

    Hugo Kornelis (9/24/2010)


    See the reply by Toreador to see why the distinction between zero-length string and NULL can be very important.

    As to the GUI - though I'd never call the...

  • RE: Design

    Toreador (9/24/2010)


    Rune Bivrin (9/24/2010)


    However, looking the table in the question I'd probably remove the NULL-ability on the cell phone numbers. Nullable VARCHAR often doesn't make sense. Particularly when they are...

  • RE: Design

    Hugo Kornelis (9/23/2010)


    Third bullet:

    "A table should avoid nullable columns."

    No. A table should avoid unnecessary nullable columns. When values may be missing in a column, making the column nullable is actually...

  • RE: Deprecated Features

    While I agree that errors aren't good, I didn't really notice them in this question. I looked up "deprecated features" in BOL, and found a reference to sys.dm_os_performance_counters. Matching that...

  • RE: Predict the outcome

    I couldn't find any mention of the restriction on increment either. My gut reaction was to answer "error" given that an zero increment makes no sense, but as I couldn't...

  • RE: Rowcount and multiple assignment

    Yep, I was wrong about that. And I even ran a test to verify my thought before I posted, but I managed to read the results of that test wrong....

  • RE: Rowcount and multiple assignment

    It's important to note that SET @result = cast (@@rowcount as varchar) does NOT change @@ROWCOUNT. The first 1 comes from the last INSERT #n VALUES(3).

    This is one of the...

  • RE: Cast vs. Convert

    sknox (3/8/2010)


    Rune Bivrin (3/8/2010)


    Good question, but which alternative is correct depends on how you define platform. If you define "platform" as "versions of SQL Server", then CONVERT works in every...

  • RE: Cast vs. Convert

    Good question, but which alternative is correct depends on how you define platform. If you define "platform" as "versions of SQL Server", then CONVERT works in every version whereas CAST...

  • RE: SELECT setter

    Oh, I know that. From the perspective of someone who started with SQL Server way back in 91 it's completely clear. But for the less experienced it can certainly appear...

  • RE: SELECT setter

    I got it right because I knew it, but I understand why that might not be obvious, and the behaviour is slightly iffy.

    If you do

    SELECT @var = MAX(object_id)

    FROM...

  • RE: Query cost

    This used to be true, and I have often used this knowledge to optimise slow-running queries. But since SQL 2005 it no longer matters from a performance point of view....

Viewing 14 posts - 181 through 194 (of 194 total)