Re. Writing by provocation

[Geoff Hart <geoff-h -at- MTL -dot- FERIC -dot- CA>]

     As several posters noted, getting information by inserting errors into
     a document for the authors to catch (or inserting comments and
     questions for them to answer) is a dangerous strategy. I agree fully
     that some of the spurious information does make it into the final
     printed document, and some is ignored or simply missed by the author
     (and thus, questions remain unanswered), but there's another serious
     risk: loss of credibility.

     The best way to work with an author is to establish a relationship of
     mutual respect and consideration, ideally with questions delivered
     in-person if you're working at the same site. One overwhelming
     advantage of the personalized approach is that you can watch for signs
     that you've overstayed your welcome, that you've offended someone, and
     so on, and adjust your questions and comments accordingly. (The
     personal touch won't always work, because some authors are simply
     difficult and annoying. So are some of us! Adapt this approach as
     necessary.) Acting unprofessionally in handling a document, as
     elsewhere, undermines your credibility ("he's not taking this
     seriously", "she misses obvious errors", "they don't want to talk to
     us about the problem"). You have to learn when to joke around, when to
     stop joking and accept informality, and when to be extremely formal
     and delicate in your approach.

     In defense of the "writing by provocation" approach, I used to use it
     long ago, in only occasionally judicious doses. Early in my career, I
     was responsible for producing a monthly report to program managers on
     the status of the 100+ publications that were floating around at any
     given time. Initially, this took me most of a day to complete until I
     circumvented the old system and developed my own tracking system.
     (Based on a WordPerfect mail merge document since I had no access to
     database software... details available from me if anyone's
     interested.) This trimmed the computer aspect of preparing a report
     down to about half an hour, but photocopying and delivering the
     reports took much longer... no network! After a while, I got the
     impression that no one was actually reading these things, so I started
     putting in obvious questions to the managers; when none of these were
     answered, I tried not delivering the report (but keeping a copy
     available for photocopying just in case!). When no one noticed after
     three months of nondelivery, I simply stopped sending the reports. In
     retrospect, I was rather fortunate not to be called on the carpet, and
     I wouldn't try that particular trick again. But the moral is clear:
     the approach doesn't work well in many situations.

     --Geoff Hart    = -at- 8^{)}   <-- devil's advocate, not cuckold!
     geoff-h -at- mtl -dot- feric -dot- ca

     Disclaimer: If I didn't commit it in print in one of
     our reports, it don't represent FERIC's opinion.