"The best way to stop the distribution of a bit of software is to charge money for it"

From Nick Jenkins
Revision as of 04:35, 9 February 2006 by Nickj (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
"The best way to stop the distribution of a bit of software is to charge money for it".

This sentence bothers me.

On a very simplistic level, it's correct of course. Economics-101 tells us that if you reduce the price, this will normally increase demand. Reduce the price of getting some software to the zero (or effective-near-zero for media and download costs), and you would expect demand to increase significantly.

On a more complicated level, it's like saying "more people would own more cars if they were free", or "novels would be more widely read if they were free". Whilst true, it completely misses the point that if these things were free, very few people would be willing to make them. On this level, it's an rather "well, duh!" statement.

People like Richard Stallman say the car analogy is broken because cars are physical objects, and taking a car denies someone else the use of that car. Again, simplistic - this only looks at it from the view of consumers, but completely ignores the costs of production. Cars take resources to produce (labour, money, and capital - in the form of plant equipment). Software takes resources to produce (labour, money, and capital - in the form of office equipment). Products take real resources to build - deal with it!

The open-source software products that have really succeed thus far tend to be infrastructure products; Essentially, the things that are of use to the many (analogous to roads, sanitation, public education, secure property rights, an enforced legal framework, and so forth). In software terms, infrastructure translates into things like databases, web servers, web browsers, email servers, kernels, system tools, development tools. Each project in this category is different, but on the whole the dollar cost of development is generally either paid for by companies who want to be involved in the project for some strategic business reason, or by individuals donating their own time. The number of companies and individuals who would have enough interest in something to expend the required resources to develop it is partially determined by how many people will benefit from the software in question, and the number of people (and developers) that will be exposed to a bit of software is largely determined by how useful it is to most people; So core infrastructure with a larger pool to draw on might be expected to get more development resources.

Open-source infrastructure impacts heavily on some software companies, and very little on others. One commercial software company that has made a strategy more than any other out of capturing the core infrastructure areas is Microsoft. This goes some way to explaining why open source software is more of a threat to Microsoft than it is to the vast bulk of commercial software vendors.

Open source software has not succeeded notably however in the area of niche applications; Applications that are specific to an particular industry or market segment. This is because these applications are usually fairly complex (so they take non-trivial resources to build), whilst being useful to a very small market, who have quite specific needs. This is probably the area in which open source software will succeed the least, and there is a vast body of niche-specific applications in existence (most of which you've probably never heard of unless you work in that field).