Re: Stop the bling, fergawdssake!

On 14/11/2011 9:54 PM, Ken Springer wrote:
> On 11/14/11 4:21 PM, Ron Hunter wrote:
>> On 11/14/2011 10:15 AM, Ken Springer wrote:
>>> On 11/13/11 4:58 PM, Art Kocsis wrote:
>>>> The layout of the touch-tone phone is completely OPPOSITE to the
>>>> layout of
>>>> a ten-key adding machine! The ten-key layout had been standard for
>>>> decades
>>>> and many people had committed it to muscle memory. The arbitrary
>>>> reversal
>>>> (just to be different?) has caused no end of pain and confusion. The
>>>> IBM
>>>> keyboard is another classic example of misguided change for changes'
>>>> sake.
>>>
>>> This may or may not be true, but somewhere in my history, I read or was
>>> told the change was intentional.
>>>
>>> In those days, many phone switches were mechanical. Someone with good 10
>>> key skills could tie those switches up in knots, as the switches could
>>> not keep up with the speed 10 key users could input a phone number.
>>>
>>> Having watched one of those old switches in motion, it isn't hard for me
>>> to see someone inputting 3 or 4 digits for a phone number before the
>>> switch could process the first digit.
>>>
>> Sounds like another stupid choice made for mechanical limitations, like
>> the Scholes keyboard.
>
> In retrospect, the QWERTY keyboard seems to make little sense, in light
> of the fact I've seen studies where typists who master the DVORAK
> keyboard can outpace QWERTY keyboard users. And the QWERTY keyboard
> seems to have come about to prevent jamming of the typewriter keys by
> fast typists.

Arguments about the relative efficiency of the QWERTY vs the Dvorak 
layouts miss the point, which is: the difference applies only to typing 
English. (1)

IOW, you would need different layouts for different languages. Duh!

So having a so-so standard that came about for mechanical reasons is 
just fine (2). A standard that's slightly suboptimal for each is optimal 
for all, because optimisjng it for some users would make unacceptably 
suboptimal for others.

The same principle holds for components of a system: Optimising one 
comes at the expense of seriously suboptimising one or more others. That 
would make the system as whole suboptimal. Better to have all elements 
slightly (=acceptably) sub-optimal. This is of course a non-intuitive, 
paradoxical insight, hence the unwillingness of  component optimisers to 
compromise. It explains why you can't design a perfect UI by improving 
its components one at a time; you are likely to make it worse rather 
than better.

(1) The difference in typing speed is about 10% for English, IIRC, and 
only for the most expert typists when typing continuously. In actual 
practice, typing is interrupted at frequent intervals, which eats up the 
supposed advantage of the Dvorak layout. Hence change is not worth the 
cost or effort.

(2) The reference to key-jamming as the motivation for the QWERTY layout 
is correct. Even so, it's possible to jam a QWERTY keyboard when the 
typewriter becomes clogged with dirt, oxidised grease, etc. I know that 
from experience. ;-)

[...]

HTH
Wolf K.