On Wed, 30 Jan 2002 15:26:14  
 Jim Elwell wrote:
...
>Now, if I am buying a drill, I would care a lot more about how many holes 
>it could drill per minute than it's dimensional characteristics. So what is 
>wrong with "units" that are useful in the real world, even if they are not 
>somehow tied to fundamental units of measure?
>
I guess you're missing the point, Jim.  What I think Adrian is saying is that we 
already have well established ways of expressing *performance issues*!  I'm not 
against "useful units", but I must concur with Adrian that IF there is already a 
perfectly legitimate way to convey an important piece of info, why come up with silly 
elocubrations?  If people are familiar with such "standard ways" of expressing 
performance, why go out of your way to avoid them?  Some examples, below:

>Take your toilet example. Switch from "gallons per flush" to "liters per 
>flush" and you have a useful measure with a metric component. No, it is not 
>entirely metric but it provides information the person specifying it needs 
>to know. He may need this info a lot more than (for example) "fills at 50 
>mL/s."
>
And what would be wrong with 50 mL/s???  Do you honestly feel the person would not 
understand how much water the toilet would be "dragging" if it were expressed that 
way?  Besides, it stands to reason that xx mL/s would be more useful a piece of info 
than on some "flush" stuff.  Why?  Who is to say that there is some "cap" involved 
when someone presses that button?  Or what if you hold it for a while longer (because 
maybe your poop is too big and it really needs to be flushed out well?  It may sound 
silly, but it does occur, doesn't it?).  In summary, I'd like to believe that water 
flow would be more... "accurate" than some "flush"...

>Another example: laser printers are generally rated in "pages per minute." 
>A rather useful figure, and pretty much unrelated to metric or colloquial 
>systems.
>...
Why not m/s?  If everyone adopted it everyone would still be able to make comparisons 
on how fast printers are.  If someone is really interested in finding out how that 
relates to how many pages per minute that would "translate" to, fine, it shouldn't be 
that difficult to figure that out.  However, such figures are quite imprecise since 
they'd depend on how many characters there are on a page, etc, etc, etc.  In this 
sense, maybe m/s may not even be the best performance ratio to use, I don't know.  
Adrian's point remains though, that one can find perfectly legitimate standard ways of 
expressing performance values.

Marcus


Is your boss reading your email? ....Probably
Keep your messages private by using Lycos Mail.
Sign up today at http://mail.lycos.com

Reply via email to