Expectations around collation

by Michael S. Kaplan, published on 2005/11/30 04:01 -05:00, original URI: http://blogs.msdn.com/b/michkap/archive/2005/11/30/498183.aspx

Back many years ago when I first moved to Columbus, OH (the place I lived before Redmond, WA) I owned a cat named Kim (short for Kimberly Cleopatra).

Well, I didn't really own Kim; no one ever really owns a cat. I think she pretty much owned me. :-)

Anyway, she injured herself and the veternarian had to put one of those collars on her for a few days. His parting advice was to have Kim take it easy for a few days.

It was at first kind of amusing to me, thinking that he must not own a cat if he thinks such things are under the control of the "owner" of the animal.

But my thought later was wondering how an animal who sleeps 21 hours a day can actually "take it easy". Since that appears to be her job, her nature, what she just naturally does.

Kim has passed on now (she was quite a few years old when I inherited her)....

Anyway, assuming you are still reading I will explain where I am going with this.

There is an interesting side conversation going on in the comments of posts like the one I wrote about the Hungarian technical sort or about whether a comes before A. The interesting part is people talking bout the various expectations of different kinds of users, who may or may nor be in different locations or speaking different languages or performing different tasks.

I found it quite amusing as people from different viewpoints added their opinions about what was the expectation, which seemed to change depending on all of those different factors.

And I was in a meeting yesterday where I found myself amused again as people in the meeting talked about the expectations of users who may prefer English sorting versus German sorting versus Dutch sorting.

It was at first kind of amusing to me since all three of those languages use the default collation table, so it seemed funny to talk about conflicting expectations when all three would sort the same way on Windows. :-)

But my thought later was wondering how a feature that the majority of users cannot describe the rules of effectively could engender so many discussions!

It is just the natural expectations that people expect Windows to do for them.

And as I watched the uncontrollable force of the people in the meeting talking about the people who use Windows, as I thought about the readers of this blog who commented about their various (sometimes conflicting) expectations that they have around collation which are also quite uncontrollable, I can't help remembering Kimberly Cleopatra and hoping she has found her door into summer.

After all, her expectations pretty much represented "the way things are supposed to be", too.

Maybe that is why i like collation so much -- its cat-like-inspiring qualities? :-)


This post brought to you by "Ѥ" (U+0464, a.k.a. CYRILLIC CAPITAL LETTER IOTIFIED E)

# Suzanne McCarthy on 30 Nov 2005 10:39 PM:

I find all this stuff you write about collation and alphabetical order increasingly interesting. Well, I was interested to begin with, of course, but ... Forget the fact that Pinker isn't interested in it. Some people have no imagination!

# Michael S. Kaplan on 30 Nov 2005 10:49 PM:

Heh heh heh -- well, I don't know that he finds it to not be interesting. It just does not fit into the "language as an instinct" model, maybe. :-)

# Suz on 30 Nov 2005 11:03 PM:

But we all have an instinct to organize!

# Michael S. Kaplan on 30 Nov 2005 11:14 PM:

Well, yes. But that was not the sort of instinct he was describing in the book....

# Suz on 30 Nov 2005 11:33 PM:

I do see that ... I am simply commenting that alphabetical order is worth considering in greater depth ... as you do.

# Michael S. Kaplan on 1 Dec 2005 9:22 AM:

Indeed I do, Suz!

Please consider a donation to keep this archive running, maintained and free of advertising.
Donate €20 or more to receive an offline copy of the whole archive including all images.

referenced by

2008/08/26 Making SQL Server operations slower (without explicitly trying)

2006/08/21 Decimal vs. hexadecimal LCIDs, backcompat, and being weird

go to newer or older post, or back to index or month or day