Re: All hail Neo!

On 26 Apr 2006 15:30:06 -0700, "Marshall Spight" <marshall.spight_at_gmail.com> wrote:

>It is worth noting that this is a *design* issue and
>not a theoretical one per se. The semantics of SQL's
>null are well-defined, if rather clunky. Likewise, my
>empty-set semantics are quite well established and
>theoretically sound. The question is, which one leads
>to the most useful software? I propose that a piece
>of software that refuses to run calculation on data you
>do have because of data you don't have is less useful
>that one that is not so pedantic. Since this is a design
>issue, the only way to validate that assertion is through
>HCI testing, which I don't expect either side to perform.

     Let us make the assumptions explicit.

     For example, C will quite happily do integer calculations that cause overflow and not indicate any sort of error. (The result is undefined.) That makes it less useful and is one reason why I do not use it much.

     In the database arena, I want to know when my data is not up to snuff, just as I prefer seeing "Division by zero error" to getting nonsense output.

>I also observe that argument-by-naive-user is not one
>I consider very strong, unless the software is designed
>specifically for naive users. (Which is clearly no the case
>for the DBMS category.) Nonetheless, it is interesting that
>excel, which is probably the math-oriented software with
>the largest penetration into the broad market, works the
>way I propose.

     Not checking assumptions is something that I expect with naive users. There are quite a few of these users. Depending on the domain, it might well include me. I would prefer to have guardrails. To me, "That operation is invalid because..." is better than a questionable answer. I can then correct the error and try again.

     How fast do you want your wrong answers?

     Personally, I can wait indefinitely for wrong answers.

Sincerely,

Gene Wirchenko Received on Thu Apr 27 2006 - 00:44:56 CEST