Re: numbering a single step + academic research

[Johndan Johnson-Eilola <johndan -at- purdue -dot- edu>]

Although there's certainly a broad, systematic logic says you *don't* 
number single steps because "1" implies a "2", there's another logic 
that says if you have fifty or a hundred sets of steps in a 
procedures manual and all except three of those sets are numbered 
sequences of more than one step, users are going to be skimming for 
specific steps based on them having numbers in front of them. And 
that skimming pattern is going to be supported by their experience 
with technical communication in general--nearly all sets of steps 
have more than one step in them, so people get used to looking for 
"1" to start a procedure--and at that point in their skimming, they 
may not know that there's only one step, so they have no reason to 
search for "*" (bullet) instead of "1".

This is where styles and systems of logic break down. Has anyone 
actually done some user testing on this to see if users tend to skim 
for "1" to find the start of a procedure (and tend to miss the few 
procedures that deny that logic and start with a bullet)? We have to 
make sure that we don't find single systems of logic and cling to them

This is also related to the point of why academics aren't doing the 
types of research people here have called for--it's nearly impossible 
to generate results that are definitive, because users' contexts and 
purposes (as well as program specifics) change constantly. I've 
reviewed submissions to _Technical Communication_ reporting empirical 
research on things like screenshot placement, and invariably the 
authors want to make claims that aren't warranted by the data (or 
that might easily be explained by numerous other factors).

However, if you want research that breaks down complicated contexts 
and users into very elementary sections, you can find a lot of that 
in cog psych journals, usability research, interface design, etc. 
(probably because people in those fields moved to them from 
scientific worldviews that prioritized the breakdown and explanation 
of reality into small rational chunks through empirically repeatable 
testing). But most practicitioners find that those results are so 
context-specific and isolated that they are, at best, rules of thumb 
that an expert technical communicator will need to constantly 
interrogate and adjust based on experience. (I'm all in favor of 
people learning to use this data in forming rules of thumb, but--as 
with grammar rules--people tend to stick to them as if they were 
immutable truths and deny all other experience.)

- Johndan

---

Subject: numbering a single step
From: Kyle Kimberlin <Kyle -at- di -dot- com>
Date: Mon, 27 Sep 1999 10:50:16 -0700
X-Message-Number: 29

I agree with those who suggest that it is not necessary to number a single step
procedure.  If there is only one step to be taken, you are not describing a
process, you are just describing an action.  You simply write a sentence, or
paragraph, telling the reader what to do.

Only a series of  steps should be numbered.

For example:

In order to restart the machine, perform the following:

1.  Press the reset button.
2.  Toggle the circuit breaker to the ON position.

-or-

In order to restart the machine, press the RESTART button.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Kyle Kimberlin
Technical Writer
Digital Instruments, Veeco Metrology Group
Santa Barbara, CA. USA
kyle -at- di -dot- com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~