To: John Mansfield who wrote (478 ) 11/30/1997 7:50:00 AM From: John Mansfield Read Replies (2) | Respond to of 9818
Crisis management in Y2K projects - From c.s.year-2000 by Ron Kenyon Ron Kenyon posted a number of simple guidelines that sum up what to do in Y2K projects. This post can be found by searching on www.dejanews.com. Article posted 7 sept 1997. ------- With Year 2000 approaching, the true crisis is *now* (crisis being the interval in which decisive action must be taken, or future results are determined by default). A few notes on managing in crisis ... Take charge. Take stock. Develop a simple, robust plan (more will go wrong, there's time for ingenuity but not for finesse). Take action. Limit the problem perimeter. What isn't solved in time won't be solved. Usually this means solving a smaller problem, sometimes a larger one. Take big losses up front by decision, rather than later on by default. In the great urban firestorms, dynamite -- not water -- may be the tool of choice. Limit decision and communication processes. Keep It Simple Stupid. If it's overly clever, you probably don't have time to think of it, and you certainly don't have time to explain it. Limit the crisis team. Qualified or not, not everyone who wants to help *can* help. Select, form up, and close the door. Managing more people takes more time. Limit expectations. Define "success" in terms that might be achievable. Managing external relationships takes time. Appoint exactly one PR officer, and don't send somebody who wants to be liked. Limit red tape. No niceties, no formalities, no apologies. Localize. Sub-optimize. Shorten supply lines. Limit risks to dependencies over which you have control. Rough it. Limit fault-finding. Difficult, as vital crisis team players may be at fault. [In re Y2K, "solutions" orineted to litigation and corporate-style "risk management" may discourage resources, drive information underground, and capable destroy vendors.] Set a challenging pace -- not an impossible one. Instantaneous peak effectiveness greatly exceeds sustained effectiveness ... only it's not sustainable. [For Y2K, remember the real scramble starts after D-Day when production systems are breaking left and right ... you'll need fresh horses. Watch the gauges, don't hit the wall early. "Triage" is cliche, but misunderstood. Some Y2K literature describes it as "taking the most critical cases first." Exactly wrong. The most critical cases consume the most care and aftercare, with most doubtful prognosis. The same resources that can treat and save (or lose) one critical case, numerous sub-critical cases can be treated successfully. Brute-force recoding effort will not stem the flood of Y2K system failures. We must take hard decisions, sharp decisions, big decisions, and by all means, correct pragmatic decisions. Future posts will suggest some hard decisions ... A) deploying some of our best talent to disengage systems, not fix them B) shifting to more fail-soft, reconfigurable enterprise integration architectures C) abandoning favorite technologies (including some of the oldest and some of the newest) D) breaking out of the straightjacket of institutional HR -- RonKenyon@aol.com 'Roger said "I had a grandfather, a lumberman, who cut nothing but walnut trees ... He was shortsighted. Was anyone warning him? I bet there was ...' (Ackerman, "The Moon by Whale Light")