This is Roger L. Cauvin's Typepad Profile.
Join Typepad and start following Roger L. Cauvin's activity
Join Now!
Already a member? Sign In
Roger L. Cauvin
Recent Activity
Joah, you chose to counter the easier straw man arguments and neglected the critical one that some transit supporters are making in opposition to Proposition 1. Rail projects cost a lot to build. That investment is well worth it if it makes for more sustainable on-going operating costs. However, a transit line that's expensive to operate drags down the entire system. Imagine introducing a line that costs as much to operate as the existing set of lines in the system. How on Earth can you afford and sustain it without discontinuing or reducing service in other lines, or magically doubling overall transit operating funds? (BTW, the federal funds are exclusively one-time funds for construction, not operations, if I'm not mistaken.) So operating costs - not capital costs - deserve the most attention. Yet Project Connect's own projections show this proposed rail line would cost much more on an on-going basis than, for example, BRT to serve the same ridership and corridor. So the argument is that adopting this line would likely degrade overall transit service. I made this point in a comment on your "perfect is the enemy of the good" post and was disappointed you neither acknowledged nor addressed it.
1 reply
Here's another possible sequence: 1. Define the problems to be solved or avoided. 2. Define, in measurable terms, the least stringent conditions that would indicate the absence of these problems. These conditions are the requirements. 3. Some of these conditions either are, or occur in the context of, functional capabilities. 4. Design and specify the user interactions and system responses. This sequence is mostly logical and not temporal. On that note, I think it's important to reconsider this statement: "With requirements in place for each capability, we can then start development. This is done incrementally and iterative using our favorite agile method." This statement implies that we apply agile methods only to development. To address the noxious effects of BUFR, we should include requirements in the iterative feedback loop.
1 reply
Yikes. Was watching Western Digital as well, but it looks like earnings ended up disappointing.
Toggle Commented Jul 21, 2010 on Earnings Surprises 7/20/10 at Covestor Live
Graham, good discussion so far of free will and determinism. I'm not sure where you are going to end up with it, but I thought I'd summarize what most philosophy says about the issue. There are basically three different stances on the issue: 1. Libertarianism - belief in free will and rejection of determinism. (This sense of "libertarianism" is not necessarily the same as the political ideology.) 2. Hard determinism - belief in determinism and rejection of free will. 3. Soft determinism (or compatibilism) - belief in both determinism and free will. Compatibilism is perhaps the most interesting. Some of its adherents not only consider determinism and free will compatible, but they actually believe free will is impossible without determinism. The argument is that, if I "will" that I'm going to do something (e.g. punch you), reliance on cause and effect is my only assurance that it will actually happen.
Toggle Commented Jun 28, 2009 on Free Will, Part 5 at Graham Glass, etc.
I think "requirements elicitation" is a good term. A product manager is like a therapist (http://cauvin.blogspot.com/2005/06/product-management-is-like-therapy.html). As a product manager interviews customers, not only does the product manager learn, but the customers learn more about their own situations and problems. Unfortunately, requirements "gatherer" describes exactly what many product managers and executives believe the role of the product manager is. How many times have you heard about product managers interviewing "the business" or "subject matter experts"? So-called subject matter experts should not be the primary source of requirements (http://cauvin.blogspot.com/2006/10/smes-not-primary-source-for.html). Nor should "the business". Prospective and existing customers (often unconsciously) are experts on their own situations and problems, but you have to play facilitator in order to draw them out.
1 reply
I think this sort of technology may be useful for assessing the usability of a product. When formulating the requirements for a product, one of the things we need to do is specify in a measurable way how usable it should be. One metric is the amount of time it takes for a user to accomplish a goal. But with what I have called "neurorequirements" http://cauvin.blogspot.com/2005/07/neurorequirements.html we may be able to more directly measure a user's frustration level.
1 reply