Saturday, March 15, 2008

Stanford Entrepreneurial Thought Leaders

Marcelo Olivas turned me on to these podcasts from Stanford (also available via iTunes). These are regular talks & interactive discussions with entrepreneurs and industry leaders who visit Stanford University. I learn something new from every episode. The lessons are applicable across industries and company maturity.

They also have videos available from their main page.

Highly recommended.

Friday, March 07, 2008

Sausage Making

Otto von Bismarck – the “Iron Chancellor of Germany” in the 1800’s is said to have made this observation:

"Laws are like sausages, it is better not to see them being made."


The comment suggests that the making of sausage is a messy business. It’s better if you just avoid thinking about what goes into making the sausage and simply enjoy the results. Similarly, the making of laws is a messy business that can be unappetizing.

I use this reference quite a bit on projects. The extension to this in the software development world is:

"Software Releases are like sausages; it is better not to see them being made."


Software development sausage making includes our development approach (e.g. points, burnups, self-directedness) and in some cases, our technology choices. I have, at times, made the mistake of opening up the “sausage making” process for stakeholders to assess, critique, and well…watch. Sometimes this is appropriate. For example when getting into detailed conversations about cost/benefit analysis on technology purchases, or to determine whether those choices are in line with the corporate IT strategy, it makes sense to discuss them. Too often, though, I think we make the mistake of getting caught up in discussing the sausage making with stakeholders in many cases where, frankly (sausagely?), they might be better off not knowing how the sausage is made.

This is a classic mistake that technologists make. We get so enamored of our technology and our process, that we think everyone else must be interested in how we do our work.

We should focus stakeholder reviews more on the array of sausages we produce, rather than how they are made. Don't show iteration burnups, or discuss the nature of a "point" in the estimation process. Apply an adapter/interface on the information to convey only that which is appropriate to the audience. Consider this a sausage casing, if you will, that abstracts the detailed content regarding the inside of the sausage. So, rather than discuss with them that

"This sausage is almost done – it needs some fennel, and a little more pork fat, and a touch of lard."


we should be saying something like

"This sausage is almost done; we are 85% confident that it will be available for consumption in the mid-April timeframe and 98% confident that it will be ready in time for the Memorial Day picnic in May. If you want to increase the probability that it is delivered in time for April, we can eliminate one of the ten sausages we have slated for April and refocus on this one."


It is in these conversations that we provide our stakeholders with information that is suited to their digestive profile.