July 8, 2014 by Kenneth Fisher
I assume that since it’s the second Tuesday of the month that it’s time for T-SQL Tuesday. And you can be certain that Dev Nambi(b/t) has decided to make Assumptions the topic of the 56th T-SQL Tuesday.
I generally like to start with definitions when discussing a word like this just to make sure everyone is on the same page. So from dictionary.com here is the definition we will be using for assumption.
1. something taken for granted; a supposition: a correct assumption. Synonyms: presupposition; hypothesis, conjecture, guess, postulate, theory.
I had a hard time coming up with something for this topic. Most of the time when I’m making an assumption it’s such a background thing I don’t really notice, I just know something is true. For example I assume my car will start when I go to work in the morning. I assume that I still have a job when I get there. Heck I assume that the sun will come up tomorrow. However, eventually my car won’t start, my company has laid off over 150 people in the last year and I could always be next, and there could be a Vogon Constructor Fleet getting ready to put in a new hyperspace bypass.
I finally remembered a tweet from @SQLSoldier (b/t) pointing out some bad advice on a forum. (Note: I’m not going to include the link. I see no reason to expose the person who gave the bad advice to any possible ridicule. It was corrected (politely) on-line and that’s enough.) I was thinking that bad advice on-line tends to be worrying. Why? Because people have a tendency to assume that what they read on the internet is correct or true. This isn’t just as DBAs. It’s that way in general. To be fair we have to make that assumption any time we learn from another person through any medium; be it written (books, internet etc) or verbal. Remember that you can put anything on-line that you want. There is no need to provide proof and proof can be faked in any event. If it weren’t for the fact that several people have actually met me in person (and you have to assume they are telling the truth) you would have no real way of knowing that I’m actually a 44 yo man instead of a 25 yo woman.
So if we can’t just assume that what we read is correct what do we do? First of all are we talking about a reputable source? Is this a TechNet article? Is it a post by Brent Ozar? If so you are probably pretty good. On a side note if you are looking at a blog post you may want to check and make sure that the post date isn’t April 1st. Next do some logical thinking. Take what you already know and see if what you are reading fits in with the rest of it. If you read something that suggests deleting your log file and re-creating it as a method of clearing up a “too large” log file then you can safely skip past that one.
So you aren’t sure if the source is reputable but it seems to fit in with everything else you know. What’s next? Do some additional research. (You might do some additional research even if you feel the resource is reputable, everyone can make mistakes.) See if you can find a couple of other sources that match the first one. If so then you are looking pretty good so far. Remember that at one time every source you could find would tell you that the world was flat so even finding additional references isn’t perfect. If you can you should also try out the answer in a test environment. You are still making an assumption until proven otherwise but at least you have done your due diligence.
Remember that in general assumptions are not a bad thing. However you need to be aware that you are making an assumption and then do everything in your power to minimize the “taking for granted” part of the assumption.