THE WELCH COMPANY
440 Davis Court #1602
San Francisco, CA 94111-2496
415 781 5700



Date: Tue, 15 Apr 2003 21:57:04 -0700

03 00050 61 03041501




Mr. Garold L. Johnson
garold.l.johnson@boeing.com
Modeling and Simulation
Boeing
3370 Mira Loma Avenue
PO Box 3105
Anaheim, CA 92803
..
Subject:   Accuracy SDS Design Dual Environment Supports Proof Reading

Dear Gary,

The email sent earlier today on the new tool developed yesterday for restoring SDS records in the Schedule (current) 00 directory that get out of sync with the list of tasks in the Schedule itself, has about a "million" mistakes (or a least a lot of them). I don't think this is critical in this case, but for publishing meeting notes, which you will be doing shortly for the SoS meetings begun today at Boeing, accuracy is important.
..
Creating anything of complexity, like meeting notes, will always have many mistakes under the current rubric, popularized by the recent Iraq excursion, that "Initial reports are always erroneous." For initial reports on meetings, the problem is not so much spelling, because, like email, these are caught by the system, but for things like leaving out words, using the wrong word, awkward construction and lack of alignment which omits context of authority and history.
..
A core problem is that the writer reads past problems and errors, because the mind reads what was intended. The writer can read something several times and not recognize errors, under Jeremy Campbell's explanation that the human mind sees what it already believes should be there based on experience. Steven Pinker later commented in his book "How the Mind Works" that overlooking error is a feature not a bug, because drawing on experience to fill in missing information avoids the delay of deliberation. For many situations in the environment where the mind is thought to have evolved, timeliness is more critical than accuracy, because the context of the moment resolves ambiguities. But that does not happen when reading because the "context" remains constant in the writer's mind. Since literacy is inherently a deliberative process at war with the biological drive for immediacy, on important documents experienced writers and publishers often assign a "second pair of eyes" to do the proof reading.
..
Experience using SDS shows (as a fortuitous by-product) that publishing SDS records on the Internet provides a new "context." Looking at the same material in a web browser yields up a lot of mistakes that are otherwise overlooked in initial writings. When a record of doing work is initially completed, for example on creating new tools, or attending a meeting, converting the record for publication on the Internet, and then reading the version on the local drive in the wl directory before publication, reveals a lot of mistakes and opportunities for improvement that are not evident in reading the same material in the original environment, here the SDS program. This enables effective proof reading without the burden and expense of using another person.
..
This is another factor to consider when thinking about a HTML editor or other arrangement that replaces the two-step process now being used of creating the record in one environment and then making a review in another environment that yields new insights, which are otherwise hidden in a one-step process. Undoubtedly, there are other considerations that may balance this benefit, but accuracy and comprehension are two of the biggest benefits that make transformation from information to a culture of knowledge a worthwhile effort.
..
Thanks.

Sincerely,



Rod Welch
rodwelch@pacbell.net



..