On Wed, 2005-06-29 at 15:23 +0100, Derek M Jones wrote:
If one defines numeracy as the ability to manipulate numbers
using calculators, then yes this research is applicable to
numeracy. Being old fashioned I define numeracy as the
ability to manipulate numbers without external aids.
I think that's unduly restrictive. I confidently
assert that I'm numerate, but I can't do much long division
or definite integration in my head. To me, numeracy
means that, when doing arithmetic, I have a sense for
the magnitude and sign of the answer; when looking at
a graph, I can relate it to numbers, and have a sense
of how they're affected by rates of change and trends.
My point is that this particular feature is not poor design. It
makes sense in a world where pressing a different operator
key to correct an earlier mistake is a more common operation
than the second operand of a multiply being negative.
The design flaw isn't in what the buttons do, but in
the whole approach to mapping arithmetic onto them.
This is why, for its time, RPN was a better system
for operating calculators than all the pseudo-algebraic
systems. I have a remarkable proof of this, which this
e-mail is too small to contain, but Google is bigger:
http://groups-beta.google.com/group/comp.misc/msg/7353fbc64c3f7082
But RPN critically depends on the user actually
being numerate, which is why engineers loved it while
the Man on the Clapham Omnibus hated it. When numerate
people stopped buying calculators in any significant
numbers, calculator design progress basically stopped
dead, because then they had to be sold to people who
couldn't add, or who were just learning how to.
These are the kinds of people that the Thimbleby
research seems to be trying to help, but there is
no economic process that can put that help into
practice, because calculators are now a commodified,
no-margin business. Such commodification is
essentially permanent; I know of no field of
technology where a no-margin commodity becomes
a high-margin diversity again.
It's like researching more usable designs for paper
clips or tea pots. For the research to be worth it,
it has to be taken up by the major manufacturers,
except they're not interested, because their customers
aren't informed enough to see the benefits, and don't
care enough to learn.
Isn't being good enough the aim of all product development?
No; making money is the aim of all product development,
but how it works varies dependent on what you're making.
If you're selling a commodity item (that is, one that meets
widely-agreed upon standards of functionality and quality,
and that otherwise is sold solely on price and availability),
then you strive to manufacture and distribute for the lowest
possible cost that will keep you in the market. New features
only get added where a marketing campaign might benefit
from them (for example, circular tea bags), but these rarely
persist or update the standard for all.
Whereas, if you're in a non-commodified field, you
strive to be *better* than the average, and ideally to
have unique selling points that your competitors can't
copy. These let you bump the price up, which potentially
lets you get more profit per sale.
Being good enough only really works for the first
situation, not the second. Compare oil or memory chips
(commodity items) with cars or jewellery (non-commodities).
They're bought differently, and thus made differently.
PCs and operating systems are becoming commodified right
now, despite Apple's and Microsoft's best efforts (hence
Apple's switch to Intel to cut costs, and Microsoft's
full-court press on software patents to defend against Linux.)
Calculators are almost entirely commodified now, and
aren't coming back, so research into improving them
seems wasted. *Replacing* them with something cheaper
might *possibly* fly, but I didn't see that in the research.
Under 30? Frank, I think you are turning into a grumpy old
man. How long before you start saying under 40 ;-)
[Looks at watch]
About six years. That's local time, of course.
--
Frank Wales [EMAIL PROTECTED]
--
PPIG Discuss List (discuss@ppig.org)
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/
