Michael J. Swart

March 16, 2014

T-SQL Tuesday #52 Round-Up: Argue Against a Popular Opinion

Filed under: SQLServerPedia Syndication — Michael J. Swart @ 1:52 pm

T-SQL Tuesday LogoThere were thirteen excellent posts this month. (See the invite post here). I loved reading every one of them. It gave a really good perspective on topics that are typically seen as cut and dry. Thank you to each one of the bloggers for putting in the time to participate this month. Here are the posts, each with a mini-review.

Michael J Swart
(@MJSwart)
I Don’t Mind SELECT * Sometimes

The opinion:
The popular opinion is that “SELECT *” should always be avoided. I argue that there are some cases where there are no downsides.
Did I buy it?
I may be a little biased, but I totally bought it. Truer words were never spoken.
Boris Hristov
(@BorisHristov)
Only My Technical Skills Matter

The opinion:
The popular opinion is that only technical skills matter. Boris argues that the most effective professionals develop other skills (like soft skills).
Did I buy it?
For sure. I would also add to his post. It’s easy to agree that soft skills matter. But I’ve recently learned that (holy crap) do they ever matter!
Tech skills are important.
Communication skills can come with enough practice
But tech skills + communication skills + wisdom are a lethal combination.
Russ Thomas
(@SQLJudo)
Eye Heart You Dee Effs

The opinion:
There are a lot of UDF-haters out there in SQL land. Not Russ, he loves them.
Did I buy it?
What say I? I say aye! Well done Russ. Totally got into the spirit of this month’s topic. BTW, he gets the Michael J Swart “Best Article Of The Month” award.
Bob Pusateri
(@SQLBob)
Why I Hate Row Compression

The opinion:
Row compression has been an extremely popular feature since its introduction. But Bob warns us of the risk of using it as spackle for sloppy data modeling.
Did I buy it?
Sloppy design causes future problems. I buy that. Row compression can enable sloppy design. I buy that. The Spock in me says row compression can cause future problems. But the Kirk in me has trouble accepting it.
Thanks for writing the most though-provoking article.
Ken Fisher
(@sqlstudent144)
Keep your head down!

The opinion:
Ken takes a look at people known for their collaborative character and he argues against that behavior with the more isolationist advice: “Keep your head down!”. The sarcasm is called out explicitly and so Ken is actually promoting more collaborative behaviors.
Did I buy it?
His post’s thesis argues with the popular trend of more collaboration. As a blogger myself, thumbs up! It’s was a really well written article, the light-hearted style makes for a good read.
Rob Farley
(@rob_farley)
Scans are better than Seeks. Really.

The opinion:
Rob and I had a great discussion in the week leading up to #tsql2sday. He wants us to believe (with no irony) that scans are better than seeks.
Did I buy it?
After reading the article, yes, I buy it. It turns out there’s a lot of subtlety when talking about seeks and scans. Especially when seeks perform RangeScans or when scans operate on filtered indexes.
Brent Ozar
(@BrentO)
Stop Tuning with Wait Stats Percentages

The opinion:
It’s all in the title. Wait statistics are a popular method of tuning SQL Server. Brent argues that you shouldn’t pay much attention to the reported percentages. Or at least not without some careful context.
Did I buy it?
Sure did. My own take is that wait stats are useful when measured against representative load, when there’s something to tune, and when you look wait rankings over percentages.
Jeremiah Peschka
(@peschkaj)
What Use is an Average?

The opinion:
I invited Jeremiah to contribute this month. Jeremiah has a writing style that doesn’t always show in his more technical blog posts. I was not disappointed. The popular opinion he seems to be arguing against is that the “average” statistic is a worthwhile measure (when applied to query durations for example). He wants us to believe it’s not always enough.
Did I buy it?
Yep, it’s not always enough. And he gets creative with his citations. It was crazy enough Jeremiah… it was crazy enough.
Steve Hood
(@SimpleSQLServer)
Why worry about CXPACKET

The opinion:
The opinion Steve Hood tackles is that the wait type “CXPACKET” is a problem and should cause worry. He argues that it just indicates parallelism which is not necessarily a problem.
Did I buy it?
Yes, In the particular case of CXPACKET, it seems that enough people are tempted to treat the symptom rather than the disease. Steve’s article addresses that.
Steve Jones
(@way0utwest)
Arguments

The opinion:
Steve Jones argues against “Never Turn on Auto Shrink”.
Really?
Really Really?
Did I buy it?
I’m a believer. He gave not one but two scenarios where I would use autoshrink as well.
Mike Fal
(@mike_fal)
Stop depending on “it depends”

The opinion:
So this is interesting. Mike Fal is asking us to get off the fence. It’s popular for DBAs to fall back on “it depends” when cornered for an answer. Mike wants us to “stop it”.
Did I buy it?
Yes, you bet. One of the reasons I picked this topic (“argue against a popular opinion”) is to explore exactly what things depend on. One of the best written articles in this list.
Jes Schultz Borland
(@grrl_geek)
Why Back Up System Databases?

The opinion:
Arguing against public opinion. Jes tells us that that could be the title of her autobiography. The opinion she argues against is that it’s super important to back up your system databases. She thinks it’s just not as important as others want us to believe.
Did I buy it?
Yes, Jes makes a good point. She raises something important. Whether you back up your system databases or script out something to help rebuild a server from scratch. It should be practiced at least once.
Tamera Clark
(@tameraclark)
Don’t Be That Guy

The opinion:
Tamera takes a different tack. Rather than an opinion, she argues against Mr. Popular asking us to not be that guy.
Did I buy it?
She gives 10 pieces of advice to a particular target audience. It’s great advice and it’s a good piece of writing. My only worry is that not enough of her target audience will recognize themselves as the target audience.

1 Comment »

  1. Thanks again for hosting. I still think this was one of the better topics in recent memory. I also appreciate “the award”.

    Comment by Russ Thomas — June 4, 2014 @ 1:36 am

RSS feed for comments on this post. TrackBack URL

Leave a comment

Powered by WordPress