Tuesday, November 17, 2009

Andy Gibbons: Grand Mistakes in ISD

Mosiah 4:29- "I cannot tell you all the ways whereby you may commit sin..."
  • We can't discuss all the ways that you can make your project die. But there are many.
Andy Gibbons (website), "Grand Mistakes"
  • Guided Missle System - commuted from San Diego. Learned that all SME's aren't. "It turns out taht even the people who were the senior maintenance people on the system didn't have a full understanding of how the system worked." Be bold. Our tendency was to change the project to fit what they knew. We asked for someone new. The second person was not a real SME. Second SME was not and SME either. Then they found Mr. Cooper!! Project was supposed to be over in 6-8 mo. Created the trouble-shooting text ''ultra-fast." On-time, below budget, took materials to DARPA. Mis-perceived what they wanted; they wanted a simulation, not a training. How little people communicate when they think they are communicating, in a design world. Sometimes client doesn't know what they want, but then they describe something and when you create it, it's different than what they wanted. Make sure you understand your client thoroughly, and make sure you stay with your client. Make sure your SMEs are SMEs, and when you find that they're not, get rid of them early. We discovered on the job that verbal text is best way to instruct. Have to consider nature of content separately from natrure of anything and design "out of a box." Representation is different from the controls, strategy, etc.
  • Flight Evaluator - Scope Creep. NW Airlines, study for Aerospace Studies. North Dakota- all they have is sky, so they had plenty of space for teaching people to fly! They wanted some cool things to show off. We had a vision that out-did our software capabilities. The program became so big that it would only fit on one computer, and there was no way of transferring it. We didn't understand how to temper our design tools. Financially taxing. Two weeks later we came up with a prototype. Show something sooner. You have to know in advance the scope within which you are designing, what the edge of your resources are with the "gee whiz." You can build a house with a hammer, or you can build it with an air gun, and the guy who uses the air gun will probably get it done faster. We were able to bring the cost down by an order of magnitude each time.
  • MD-80 Flight Guidance System. Another system for TWA. The environment in which we were working changed. Video disk wiht grahical, new operating system, new graphipng system. Otherwise, everything else was stable. We were overtaken with versioning within our own company. During a project, you can get overtaken easily by a new version of software. Have to understand everyone on your team. One of the reasons that you want to be knowledgeable in advance is to understand where your team is. Programmers, Artists- the moment you start talking with them, they start thinking of what they want and how they would be doing it. Sometimes they don't know what the instructional method is, and if they knew, they'd do something much differently. Many people do not understand.
  • SMEs and scheduling conflicts. Companies do not send the best workers to move to a place. Used pet shop- What they'll send you is the used pets. Learn as much as you can about them, and learn early when you need to make a change.
  • Get to the heart of the other person's subject matter as quickly as possible is a skill to be learned. There are different kinds of subject matter. You can do a task analysis when you know the subject matter. You have to develop, as an INstructional Designer, a sense of knowledge that's different from other concepts of knowledge. YOu have to understand knowledge structurally.
  • Structuring Subject Matter:
  • Task Analysis (several varieties)- Cognitive, Psotka
  • Joe Anderson, Kittornger (Act * - If, Then rules; WMEs)
  • David Johnson - task analysis methodologies
  • We have been so flat in our dealing with knowledge, it's a scandal. We continue not to be interested in what kinds of knowledge are there.
  • Subject Matter Experts, keeping on schedule: There is a feeling among designers and people. There will develop within your team a feeling that there is one right way to do this, and everyone will start fighting for their own way. You have to avoid that. In design, there are always a multitude of right answers. Optimization is a dream, a fantasy. What you're looking for as a designer is a design that satisfies the customer. Debate about how to refill a paper machine, held up progress for three weeks. You have to watch people thinking they're right getting in the way. You have to be open to lots of ideas and develop a spirit among your team like IDEO. You have to develop that kind of culture. (*This is what I want to cultivate in my team, in my design, and in my home.*)

No comments: