Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sdd.hp.com!elroy.jpl.nasa.gov!ncar!gatech!purdue!haven!mimsy!tove.cs.umd.edu!cml From: cml@tove.cs.umd.edu (Christopher Lott) Newsgroups: comp.software-eng Subject: Re: measuring software complexity (was: Counting semicolons) Summary: plan carefully Message-ID: <33118@mimsy.umd.edu> Date: 17 Apr 91 16:46:23 GMT References: <36650004@hpopd.pwd.hp.com> Sender: news@mimsy.umd.edu Reply-To: cml@tove.cs.umd.edu (Christopher Lott) Organization: The University of Maryland Dept of Computer Science Lines: 33 In article <36650004@hpopd.pwd.hp.com> daves@hpopd.pwd.hp.com (Dave Straker) writes: >On measuring complexity... > >Before you measure anything, you should ask: Why am I doing this? >What am I going to do with the information? What questions will >it help answer? What decisions will it help me make? u This seems to be a good spot to put in a plug for the Goal/Question/Metric paradigm, developed by Dr. Victor Basili and others here at U Maryland. When doing measurement, our group has found that the bottom-up approach leads to FAILURE. (lessee, here's a metric that's cheap to compute, let's see what it tells us, yeah, that's it.) A vastly better approach is to start from the top. First you state your Goals. These may be to characterize, to evaluate, etc. Then you write some Questions that will allow you to achieve your goals. E.g., what was the cost of X, how many Y were found in phase Z. Finally, you define and look for Metrics that will answer your Questions. Questions may apply to different Goals, and similarly Metrics may participate in answering different Questions. This is a specialization of the scientific method. State problem, write hypothesis, experiment, evaluate, repeat. For more information, email me and I'll dig more to find tech reports which I can send out; they're free, but we're suffering from a NASTY budget crunch, so a preaddressed mailer or some such would be Most Helpful. For even more information, have your organization send you for a 6-month or longer sojourn here at Maryland CS. chris... -- Christopher Lott \/ Dept of Comp Sci, Univ of Maryland, College Park, MD 20742 cml@cs.umd.edu /\ 4122 AV Williams Bldg 301 405-2721