Erik Darling wrote me last week with this idea:
There’s a sort of recurring company chat joke about people who flip out over rows and columns vs. fields and records, etc. And of course, there’s this outdoorsman magazine called Field & Stream.
And he suggested a parody of Field & Stream called Field & Record. That led directly to this:
Thanks to the team at Brent Ozar Unlimited for the suggestion and the article ideas. You guys are hilarious!
My Two Cents on the Debate
I’m definitely a descriptivist. Language is always changing and if a word or phrase gets adopted widely enough, it is no longer “wrong” (whatever that means).
So when I hear “Field” and “Record” they’re acceptable to me. But if I’m explaining something, I don’t want to distract from the thing I’m saying. And from that point of view, I try to use “Row” and “Column” because I don’t know anyone who blinks at those terms. In other words
- When speaking, I use “row” and “column”
- When listening, I do not correct “field” and “record”.
This also means I never use the word “whom” which is a word that has the strange quality of being distracting and correct.
Exceptions
I can think of a couple exceptions
- Kendra Little made a list of confusing words or phrases. When terms are confusing and it’s important to be precise. In that case, a correction is necessary.
- Tom Larock talks this week about similar issues. He says “I feel it necessary to remind you that there is never one right answer.” Ironically, he says so in a post called Why You’re Wrong: The Data Professional’s Guide to Contentious Issues.
I dare someone to tell me I used the word ironically wrong.
Worst I ever came across was “vertical row”.
Wish I was joking.
Comment by David Swart — April 20, 2016 @ 8:59 am
“Try/Catch and Release”
genius!
Comment by mmiike — April 20, 2016 @ 2:46 pm
Thanks!
My favorite hidden joke is the escape sequences for ascii characters. Specifically, the characters that represent field separators and record separators.
Comment by Michael J. Swart — April 20, 2016 @ 3:50 pm
I think there is a big difference between correcting someone you’re having a conversation with and correcting something written (e.g. as a moderator on a forum).
While you and I both know that there is a difference between a row and a record, for most people, this difference is irrelevant. Still, I feel like semantics are a bit more important when written down and read over and over.
Also it sounds like your whom is my whilst. 🙂
Comment by Aaron Bertrand — April 21, 2016 @ 8:48 am
I buy that Aaron,
By the way. I love your answer at What is the difference between a “record” and a “row” in SQL Server?. Whilst reading it, I thought to myself Man, I wish I included a link.
Comment by Michael J. Swart — April 21, 2016 @ 9:02 am
I’m still disappointed that my suggestion of “Can deer urine improve performance?” didn’t make the cut.
Comment by Richie Rump — April 21, 2016 @ 9:30 am
Oh man, I thought that was an off-the-wall joke. Apparently I don’t know a thing about hunting. I’m surprised that deer pee is a thing. It double-confirms that that hobby is not for me.
Comment by Michael J. Swart — April 21, 2016 @ 9:39 am
I am sort of famous for being very picky about row/record and column/field. The reason is I have to teach SQL classes and after doing it for 30 years. I found that when people use the wrong words to get the wrong mindset. In particular, they have not unlearned their old sequential file model of data. They start talking about “previous” and “next” records when they go to solve a problem.
Because they see the data is a physical materialized thing in some kind of storage media, they do not understand CTE’s, views, derived tables were computed columns. The immediately run out to the temp DB and allocate physical storage for their data. They put order by clauses on statements make no sense.
If you get a chance to see it, the BBC did an educational video named “The Story of One” which is about how we learned to count. One of the scenes in it is an Australian aboriginal talking to his grandfather. He asked him how many grandchildren he has, and the man cannot answer. Instead, he has to make marks in the dirt and name the kids and then show you the marks. This is why and abstract mindset is important.
I loved the fake cover! Can we get it as poster?
Comment by Joe Celko — April 23, 2016 @ 11:17 am
Hi Joe,
I’m going to check out the “Story Of One” it sounds fascinating. And it reminds me of this article from the NYT: From Fish To Infinity which also confirms the value of abstraction.
I’m glad you liked the poster. I had fun making it. The cover was created with the web in mind, rather than print. There’s a couple tiny things I want to change if I want it for printed. But if I get around to doing that, I’ll make a PDF of it available.
Comment by Michael J. Swart — April 25, 2016 @ 9:22 am
When speaking, I use “row” and “column”
When listening, I do not correct “field” and “record”.
As a pedant, I could not help but conclude that the whole article would have been improved had you put the analogous terms in the same order – e.g.
When speaking, I use “row” and “column”
When listening, I do not correct “record” and “field”.
Comment by Edward — April 24, 2016 @ 7:38 am
Good point Edward, You’re right.
I’m not going to update the post because everyone has a little bit of troll in them 🙂
Comment by Michael J. Swart — April 25, 2016 @ 9:04 am
When I first started with SQL Server, it was as an ETL manager supporting clients who’s extracts came from mainframe processing. I would use the word “column” when describing a “field” and they would get confused. In mainframe speak, a “column” has an integer value and represents a single place in a fixed width file, across records. I had to adjust terminology for clarity’s sake, even if it wasn’t accurate in the database world.
Comment by Steve Mangiameli — April 25, 2016 @ 3:52 pm
Hi Steve,
Some would say that’s a teachable moment. But I get it, context is important.
As another example, some colleagues of mine will say something like “they took a snapshot of the database for testing purposes”.
I have to ask. “When you say snapshot, did you mean a SQL Server database snapshot? Or a hardware vendor snapshot?”
It turns out, they meant that they copied the database to a new location via a simple backup and restore! Depending on the context, terminology is very important.
Comment by Michael J. Swart — April 26, 2016 @ 8:41 am