At its May 28, 2009 meeting, the Plone Foundation Board of Directors passed a resolution setting a 3,000,000 USD valuation of the Plone 3.3 code base to carry on our non-audited books.
The purpose of the resolution was to be able to give a reasonable estimate of the Plone Foundation's intellectual property. The board — after consultation with an accountant — decided that valuing the code base under well-known metrics would be the easiest way to do so.
Code lines were counted based on the shipping Plone 3.3 code. A distribution of the Python packages from the Unified Installer was used as a base. Packages that are not maintained in svn repositories controlled by the Plone Foundation (e.g., Kupu, KSS) were excluded. Only code lines were counted (no text files); blank lines and Python comments were excluded.
The basic COnstructive COst MOdel (COCOMO) model was then used to estimate the programmer years that would be required to reproduce the code. The coefficients for organic projects were used. Programmer time was priced at $55,000 a year, the value currently used in ohloh.net valuations.
Files: 3,048, Lines of Code: 215,046 Using COCOMO coefficients: a 1.05, b 2.4 Programmer years: 56.26 @ $55,000 per year = 3,094,216 USD
What does it mean?
Very little — unless you need a rough estimate of the size and value of the portions of the Plone project that are held by the Plone Foundation. The numbers may be most valuable in explaining the size and depth of the project to agencies and clients.
Is it realistic?
The board believes that this is a very conservative valuation. It does not include a number of other very valuable assets for which we have no accepted valuation methodology, including our extensive language translation files, 800+ add-on products, and thousands of pages of documentation.