Home | Login | Recent Changes | Search | All Pages | Help

ExtremeIncrementalism

As we observe with wikis, even tiny contributions can improve the utility of a web site. Correcting a spelling or rephrasing a sentence to make it clearer takes any reader only a few moments, and yet every subsequent reader can benefit from the improvement. This is one good reason to prefer wikis to those forum systems which just let folks add another message under some topic.

The incremental principle is useful in many places. TomGilb has been advocating it (as EVO) in software projects since the 1970s. See http://Gilb.com and grab a copy of the PDF of his new book, "Competitive Engineering". You will end up wanting a paper copy for use as a handbook and the PDF just to find stuff not indexed but wanted anyway.

He suggests that no project phase should exceed two to five percent of the project budget. If you insist, you can almost always find a way to make progress in such small steps. I am advocating this incrementalism for very wide use.

See TrashingRequirementsEngineering for how to start a project, any project worth tracking. Once you have identified the true goals of the project, you can address what steps can be taken to forward it.

TomGilb uses Impact Estimation spreadsheets for proposed specific enhancement tasks to pick the highest value, lowest cost step as the one to attack next. The planning is not finished until the project is finished; it doesn't need to be.

DickKarpinski started this page.

Only on the hypothetical Systems as Humans Interacting Together forum -- EdmundSchweppe 2006.08.17