September 20, 2012 at 4:21 am
IIRC Gail has a nice example on her blog.
You could tweak it to fit in your database.
-- Gianluca Sartori
September 20, 2012 at 4:39 am
Brandie Tarvin (9/20/2012)
HowardW (9/20/2012)
Brandie Tarvin (9/19/2012)
Lynn Pettis (9/19/2012)
Do you ever feel that no matter how many times you tell someone what the problem may be that they just aren't listening?Yep.
My current drama is trying to ban WITH NOLOCK() at my workplace. Unfortunately, certain influential people like that particular hint (because it solves ALL problems), which means my ban is pretty much doomed to failure. @sigh.
Ah, well. I live to battle another day.
How about suggesting you test Read Committed Snapshot Isolation as a compromise?
It's an idea, but I may get lynched for suggesting changes in database settings. Two of the excuses I foresee are "thiings have been working fine since SQL 2000, so why change anything?" and "these systems aren't the financial systems of record, so why worry about committed data?" There are other considerations that I can't seem to convince anyone are legitimate considerations.
So what I need to do is find an example of proving NOLOCK() will screw with the data, but tailor that example to fit our systems and our data so they see the horror of what they're trying to do. If it's just sample data, it won't have the same impact. If it's our data, it should (I hope) make them reconsider.
Jason Strate has an example on his blog. Itzik Ben Gan has an example somewhere.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
September 20, 2012 at 5:54 am
Thanks, guys. I'll take a look at those.
FYI, there's a "new" IE flaw: http://redmondmag.com/articles/2012/09/18/ie-flaw-being-actively-exploited.aspx
September 20, 2012 at 5:57 am
Brandie Tarvin (9/20/2012)
Thanks, guys. I'll take a look at those.FYI, there's a "new" IE flaw: http://redmondmag.com/articles/2012/09/18/ie-flaw-being-actively-exploited.aspx
Running only on the POF now, I no longer worry about viruses or exploits because they never happen in the Mac OS... except when they do.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
September 20, 2012 at 7:16 am
Brandie Tarvin (9/20/2012)
HowardW (9/20/2012)
Brandie Tarvin (9/19/2012)
Lynn Pettis (9/19/2012)
Do you ever feel that no matter how many times you tell someone what the problem may be that they just aren't listening?Yep.
My current drama is trying to ban WITH NOLOCK() at my workplace. Unfortunately, certain influential people like that particular hint (because it solves ALL problems), which means my ban is pretty much doomed to failure. @sigh.
Ah, well. I live to battle another day.
How about suggesting you test Read Committed Snapshot Isolation as a compromise?
It's an idea, but I may get lynched for suggesting changes in database settings. Two of the excuses I foresee are "thiings have been working fine since SQL 2000, so why change anything?" and "these systems aren't the financial systems of record, so why worry about committed data?" There are other considerations that I can't seem to convince anyone are legitimate considerations.
So what I need to do is find an example of proving NOLOCK() will screw with the data, but tailor that example to fit our systems and our data so they see the horror of what they're trying to do. If it's just sample data, it won't have the same impact. If it's our data, it should (I hope) make them reconsider.
Not long ago, I worked for a company that thought 70% accuracy was fine. All kinds of weirdness in this world.
--------------------------------------
When you encounter a problem, if the solution isn't readily evident go back to the start and check your assumptions.
--------------------------------------
It’s unpleasantly like being drunk.
What’s so unpleasant about being drunk?
You ask a glass of water. -- Douglas Adams
September 21, 2012 at 9:03 am
ok im breaking out my "Realy? WTF" card on the behaviour twards the end of http://www.sqlservercentral.com/Forums/Topic1357081-392-3.aspx. did we really hit Godwin's law?
For performance Issues see how we like them posted here: How to Post Performance Problems - Gail Shaw[/url]
Need to Split some strings? Jeff Moden's DelimitedSplit8K[/url]
Jeff Moden's Cross tab and Pivots Part 1[/url]
Jeff Moden's Cross tab and Pivots Part 2[/url]
September 21, 2012 at 9:37 am
capnhector (9/21/2012)
ok im breaking out my "Realy? WTF" card on the behaviour twards the end of http://www.sqlservercentral.com/Forums/Topic1357081-392-3.aspx. did we really hit Godwin's law?
Yes, Craig mentioned a certain person (naughty Kraig, that really was evil :hehe::hehe:) so technically we hit Godwin's law - but I think Craig's comment was valid - he was merely pointing out that all sorts of people have wikipedia pages.
ScottP produced a purely political rant no SQL relevant content whatever, which I think is far more fatal to a topic than an accidental hit on Godwin's law.
Tom
September 21, 2012 at 9:45 am
capnhector (9/21/2012)
ok im breaking out my "Realy? WTF" card on the behaviour twards the end of http://www.sqlservercentral.com/Forums/Topic1357081-392-3.aspx. did we really hit Godwin's law?
Wow, I frequently disagree with Celko and do again here on SQL Server related concepts, but he's absolutely delusional if he thinks we'll be using DNA as an identifier in 5 years.
And DUNS as PKs? Madness. D&B is actively trying to downsize to the point where it can be purchased. DUNS might not survive that. Why would you put your entire design at the mercy of an external, corporate artifact?
--------------------------------------
When you encounter a problem, if the solution isn't readily evident go back to the start and check your assumptions.
--------------------------------------
It’s unpleasantly like being drunk.
What’s so unpleasant about being drunk?
You ask a glass of water. -- Douglas Adams
September 21, 2012 at 9:51 am
Stefan Krzywicki (9/21/2012)
capnhector (9/21/2012)
ok im breaking out my "Realy? WTF" card on the behaviour twards the end of http://www.sqlservercentral.com/Forums/Topic1357081-392-3.aspx. did we really hit Godwin's law?Wow, I frequently disagree with Celko and do again here on SQL Server related concepts, but he's absolutely delusional if he thinks we'll be using DNA as an identifier in 5 years.
Gives a new meaning to large primary key columns.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
September 21, 2012 at 10:03 am
GilaMonster (9/21/2012)
Stefan Krzywicki (9/21/2012)
capnhector (9/21/2012)
ok im breaking out my "Realy? WTF" card on the behaviour twards the end of http://www.sqlservercentral.com/Forums/Topic1357081-392-3.aspx. did we really hit Godwin's law?Wow, I frequently disagree with Celko and do again here on SQL Server related concepts, but he's absolutely delusional if he thinks we'll be using DNA as an identifier in 5 years.
Gives a new meaning to large primary key columns.
Makes GUIDs look easy to handle
September 21, 2012 at 10:41 am
Hmm... how would that handle identical twins, etc with identical DNA?
Wayne
Microsoft Certified Master: SQL Server 2008
Author - SQL Server T-SQL Recipes
September 21, 2012 at 10:47 am
WayneS (9/21/2012)
Hmm... how would that handle identical twins, etc with identical DNA?
Identical twins don't have identical DNA.
There are several reasons for this
--------------------------------------
When you encounter a problem, if the solution isn't readily evident go back to the start and check your assumptions.
--------------------------------------
It’s unpleasantly like being drunk.
What’s so unpleasant about being drunk?
You ask a glass of water. -- Douglas Adams
September 21, 2012 at 10:51 am
Well, the thread appears to be locked, so I can't reply, but:
Celko
First, email is so obviously not a natural key for a human being that I'm always shocked when people try to use it that way. It's fine as a login string (username), in many cases, but for anything else it's worse than using names. And the problem with using them as logins is that this helps promote security-reuse, which is one of the biggest keys to identity theft. So it's not just bad database design, it's also a security hole, and assists criminals.
That is why you add security after the email address gets you into the database. Has Amazon failed yet?
Um. Yes. http://www.emptyage.com/post/28679875595/yes-i-was-hacked-hard?6f020580
September 21, 2012 at 11:39 am
Steve Jones - SSC Editor (9/21/2012)
GilaMonster (9/21/2012)
Stefan Krzywicki (9/21/2012)
capnhector (9/21/2012)
ok im breaking out my "Realy? WTF" card on the behaviour twards the end of http://www.sqlservercentral.com/Forums/Topic1357081-392-3.aspx. did we really hit Godwin's law?Wow, I frequently disagree with Celko and do again here on SQL Server related concepts, but he's absolutely delusional if he thinks we'll be using DNA as an identifier in 5 years.
Gives a new meaning to large primary key columns.
Makes GUIDs look easy to handle
Perhaps he's a bit ambitious on the time scales but organic computing is being researched. Obviously storing DNA as data wont work but maybe as an organic compound. Blue sky thinking at its best.
September 21, 2012 at 12:02 pm
I wouldn't call it blue sky thinking. They actually are trying to figure out how to do exactly the types of things Celko is describing. I'd call 5 years something of a stretch, though. We're not that close to using it as a unique identifier. And by the time we have the tech to do such a thing, I doubt we'll be using memory sticks as the other half of the key.
Viewing 15 posts - 37,936 through 37,950 (of 66,712 total)
You must be logged in to reply to this topic. Login to reply