The State of Custom Software Development in Laboratory Informatics
Written by Gloria Metrick on . Published as written. It may not reflect how these systems work today.
When Paul Keister left a comment in my blog, the other day, I happened to go read his blog, in return. His latest posting in his “Common Knowledge” blog was “Is Custom Development OK, Again?” ( http://traf-o-data.blogspot.com/2010/04/is-custom-development-ok-again.html ) in which he points to an article from Dr. Dobbs (software development).
It got me thinking about the state of custom development in our own industry. For all the years talking about COTS (Commercial Off-The-Shelf) software and of pretending that all the programming we do for them is just a type of configuration (rather than programming or customization, for example), we still seem to have a lot of custom development going on in various parts of the industry.
There appears to be what might seem to some of us to be a counter-culture of software development, but I sometimes wonder if their segment is larger than we think. Maybe it’s those of us working with COTS and pretending we’re not programming who are the counter-culture rather than the majority.
LIMS and ELN projects tend to be relatively large software projects and, like other software projects, so prone to failure and overruns. Then, are COTS merely a type of sales promise that create a warm and fuzzy feeling upon purchase, but which don’t live up to the benefit over custom development? Or, is it the fault of those who implement these systems, such as myself? Then again, it could be that customer expectations continue to be too high for these projects to trend up in their success? Looking at the continuing bleak and high failure rates among these COTS systems, one has to wonder.
Gloria Metrick
GeoMetrick Enterprises
http://www.GeoMetrick.com/