Posted by Shuurai on 01/04/07 20:20

--CELKO-- wrote:
> >> Ken writes in the first chapter that he will be using rows/records and columns/fields interchangeably throughout the book. <<
>
> Hopefully that will be fixed in his SQL-2005 book.

It doesn't need to be fixed - as he rightly points out, it is common
usage among those in the industry.

> Since the
> Sybase/SQL Server family is built on a file system model (i.e. static
> contigous physical storage) it is not such a difference here. But in
> Sand, Teradata, Model 204 and other SQL products they use totally
> different storage methods. For example, Sand outputs table one column
> at a time, not one row at a time. They are assembled from compressed
> bit vectors and are not contigous.

The terminology one uses in casual conversation does not need to match
the inner workings of the technology.

> >> People who use SQL use these terms interchangeably. <<
>
> No, we don't; SQL Server people and Newbies who are still thinking in
> terms of file systems do.

Yes, we do; if you spent some time with real developers and
administrators, you'd find that the terms "record" and "field" are
very, very commonly used. Personally, I prefer row and column but so
long as you know what people are saying, it's just not an issue.

> >> The people who make a big deal of it are akin to the people who correct grammar errors to win an argument. <<
>
> No, more like accountants, mathematicians or engineers who insist on
> the proper terms, computations and mindsets to solve problems.

I've worked with engineers and accountants for years. The ones who
nitpick over syntax are typically the ones that like to nitpick in
general.

> Or, to keep you analogy, more like editors and writers who who correct grammar
> errors to do their job properly.

A good editor knows that common use of the language by the intended
audience is every bit as important as formal rules of grammar.

> >> If they had something useful to say, they would. <<
>
> Gee, after seven good selling books, 800 artifcles and having a few
> standard SQL programming techniques named after me, I think I can claim
> having had something useful to say :)

Which makes one wonder why it doesn't happen more often. :)

[Back to original message]