Risk Management -- 5 minute course

What is risk management? It is planning for things that might go wrong.

Many software projects do not do this. They just hope that all will go well.

Step 1 -- Create a 9-part grid, where X axis is "likelihood of happening" and Y axis is "impact on project". (Give some examples of items in each box.)

Step 2 -- Make a list of things that can go wrong on this project.

Step 3 -- Put the items you listed into the grid.

Step 4 -- Take action now to mitigate problems from any bad-and-likely items.

Step 5 -- Make plans (or take some actions) for other items in moderate-danger boxes.

Web-Time Development -- 10 minute course

Demand for speed is nothing new. My friend who worked at Prudential in early 80s... Never enough time, not enough control, always demand for new features.

"Rapid development" and "RAD" are not development methodologies. They are goals. But you need a concrete way to achieve that goal.

If you hear a company say, "We do rapid development here" or "We use RAD" be cautious. This is often code for "We work at a frantic, burnout pace here." You should ask them "How exactly do you do RAD? What methods do you use to get things done quickly?" If they have a good answer, fine.

Look at picture on RD page 9. Good to keep this in mind as you work on demanding projects.

There is a good discussion on RD pages 113-115 about whether a project really want rapid development, or something else.

Should you throw out everything you learned in this class when doing fast-paced web development?

No, you should use the techniques you learned here to do rapid development, with good quality.

1. Start simple.

Suppose your manager comes to you and says, "Our company has no web site at all now. We want you to create one that shows out product catalog, sells the products, tracks their delivery, and lets customer access our help desk."

What should you do for Version 1? (Make sure your company owns their own domain name. Then create a nice-looking one page web site with basic company info. You can do this in a day.)

2. Use rapid spiral / incremental development. Add a few small features for Version 2, get it working right, and roll it out. Then add a little more for Version 3, etc. (What would versions 2, 3 and 4 be?)

3. Keep the quality high. You (and your company) won't look good if you do something fast but it doesn't work right.