How Realistic is the Dom Perignon Test -- Part 1

Posted by Jean Ichbiah , Tue, Nov 19, 2002, 21:59:01 Reply   Forum

During the Dom Perignon II contest, in June 2000, a discussion arose about the validity of the test on comp.sys.palmtops.pilot. Below is an excerpt of an exchange between Gordon Walker and Arthur Hagen.

See also -- Part 2



From: Gordon Walker (gwalker@127.0.0.1)
Subject: Re: Dom Perignon Speed Contest - First Week Interim Results
View: Complete Thread (31 articles)
Newsgroups: alt.comp.sys.palmtops.pilot, comp.sys.palmtops.pilot
Date: 2000/06/23

On Thu, 22 Jun 2000 18:37:25 GMT, Arthur Hagen wrote:

>If I were to complain against Fitaly Stamp and not the test, it would
>have been because it requires you to look at the PDA while entering
>text, and that it's not very usable in the dark. However, I believe
>Fitaly *is* uniquely suited for entering English prose on a PDA, and a
>great help to many people who needs this functionality. However, input
>on a Palm consists of much more than this - all depending on how a user
>uses his Palm - and the TEST is not fair to the input methods more
>suited for other tasks (like a numpad replacement for those who enter
>numbers more than anything, or jot/grafitti for those who have to look
>at a manuscript while entering).

But this is unreasonable. To satisfy your requirements the test would
have to contain:
o Abbreviated, unpunctuated text for Datebook entries
o Short sentences without propositions for ToDos
o A selection of names and telephone numbers
o Email and snail-mail addresses
o Mathematical formulae (spreadsheets, MathPad etc)
o Longer passages of normal text (emails, memos, QED documents)
o Numbers (expenses etc)
and possibly more. Each of these will exhibit a different pattern of
word and character frequency, each of these will no doubt be more
favourable to one or other of the various types of input available on
the Palm. To further equalise the playing field the proportion of the
test phrase that would be taken up with each category would have to be
distributed in a manner equivalent to the amount of time the average
user would spend doing each of these. Of course such information is
not available for the simple reason that there is no single profile
for Palm use.

But let's be realistic, when you're entering a Datebook reminder or a
Todo list entry, or an email address you don't tend to do several
dozen at a time. The primary tasks which requires a Palm user to spend
prolonged times entering data is the entry of normal english text, be
it for a memo, an email, a longer document or whatever. This, then, is
the main kind of entry the average user wants speeded up.

Given this, a competition to see how fast you can enter plain text
seems completely reasonable to me.

>I still claim that the test IS biased, and the results not relevant to
>everyday use, unless your everyday use consists of entering short
>passages of plain English text.

Actually it does, I use DateBk3 a lot, all the entries are brief
english phrases describing what I'm going to do. I use Todo, each of
the entries is a brief english phrase of what needs to be done. I use
memos to scribble all kinds of bits of information, the vast majority
of which is made up of plain english. I write emails in english, the
address being selected from my address book and not usually written
manually. I use Handyshopper for lists the entries in which are
usually brand names in english. The only places for me where the
pattern breaks down is MathPad which I use a lot in work. I can also
imagine that if you used QuickSheet or the like a lot you might have a
markedly different usage pattern.

In short, yes my everyday use consists of entering short passages of
english text. Given the built-in apps and most of the common third
party ones I'd guess that holds true for most Palm users although I
stand to be corrected.

Again, as such, the test seems eminently reasonable.

[...]

--
Gordon Walker


See also -- Part 2


| Edit | Reply View All   Current page
Followups