Re: Re[2]: acceptable error rates

[Sella Rush <SellaR -at- APPTECHSYS -dot- COM>]

Arlen wrote:

list of unacceptable .2% error rates... followed by

>>I could go on, but I think I've made my point.

You've actually made 2 points (for the price of 1):

1.  How statistics can be used to support virtually every position
conceivable.  Since all your examples were fail/succeed examples, we
would actually have to determine an error/fail rate at the lowest level
[i.e., number of inaccurate words--or letters (!)], which of course
would result in a much lower error rate.  If you really care you can
figure it out for yourself. ;-)

2.  The same point made by several of the posts, which is that its too
easy to interpret a typo error rate incorrectly without the full
picture.

Oh--here's another point you made, Arlen (you were really cooking!).
Technical writing really should be judged on a fail/succeed basis.
This, of course, will need to be determined on a case-by-case basis.
For documentation, someone had a good idea (which I've also heard
elsewhere) to monitor tech support lines and see where the user guides
are failing.  For my little engineering company, success meant pleasing
the client and maintaining an overall positive reputation.  If Mr. X at
Boeing can't stand a misplaced comma and is willing to pay for 3
proofreading rounds of 3000 pages to avoid seeing one, then that's what
Mr. X gets.

I can feel hot breath of an off-topic flame on my neck so that's all I
have to say on this subject.  Thanks, Arlen, I knew someone would go
after the stats and now I can breath again. {grinning}

++++++++++++++++++++++++++++++++++
Sella Rush, Tech Writer
Applied Technical Systems, Inc. (ATS)
Bremerton, Washington
Developers of the CCM Database