Strategically using Open Source Software in your projects can allow you to focus money and resources where it counts…on differentiators and not commodities.

Larger projects we work on typically involve more than one piece of software or technology. In the most basic case, we deal with a database backend and application frontend to manipulate and use the data while enforcing business rules.

Even with simple examples, it is easy to illustrate why OSS(Open Source Software) allows money and resources to be focused on the parts that count.

Commodity vs. Differentiator Technology

Consider a project broken into the following pieces:

• Hardware
• Operating System
• Database
• Application Server
• Actual Application

There is a definitive budget that needs to be spread out over these pieces of the project. In general, it is desirable to minimize cost. But at the same time, most projects are initiated to create an edge over the competition.

Looking at each piece, we can determine if it is a commodity or differentiator with respect to the end applications. The following points are not intended to be technical arguments or suggestions, they are the business cases for allocating resources and money.

*Hardware* — Clearly a Commodity

Hardware is relatively homogeneous. There are differences between IBM and Dell servers, for instance, but we will consider these differences as not contributing to large overall cost or performance metrics.

*Operating System* — The Operating System is also a Commodity

Even though our choices here are not relatively homogeneous, from a web-based application the operating system is not a primary factor. There are performance and security issues to consider, but the final application could be able to run on any modern <abbr title=”Operating System”>OS</abbr>.

*Database* — Database Systems are Now Commodities

Databases are in fact becoming relatively homogeneous. The feature sets do vary from RDMS(Relational Database Management Systems), but the existence of rich database abstraction libraries (like ODBC, Perl-DBI and PHP-DB to name a few) means an application could connect and use a wide variety of database products.

*Application Server* — The Application Server is a Commodity

The application server software is somewhat transparent to our application. Again, like the operating system argument, there are security issues and performance metrics to consider. But in the long run, applications can be built to run on many commercial and open source web servers.

*Application* — The Application Itself is *not* a Commodity

*This is the differentiator*. We are not necessarily going to gain a completive advantage having the same application ported for Win32 vs Linux,

However, there is a competitive advantage if an application has better features, is easier to use and maximizes efficiency (or whatever the project initiative is).

Paying a Premium for a Commodity is a Bad Business Move

The business approach is fairly straightforward. If you consider a piece of an application a commodity, the minimum amount of money ought to be spent on it.

Consider shopping in an intersection for automobile gasoline. Given the same octane rating, most people with look at the prices and make their decision accordingly. This is because most do not believe that paying a premium for a particular brand will result in a significant preformance advantage. Therefore paying a premium gets you little gain and is money wasted.

Looking at the hardware once again. Given our specifications are satisfied, will the brand of hardware significantly contribute to the overall project? If the answer is *no*, then there is no reason to pay a premium for a particular brand.

The same though process can be applied to the operating system. Will Windows provide us a cost-significant advantage over a supported Linux distribution? In most cases, especially with web-based applications, the answer is also *no*. Therefore it is not a good business decision to spend money on licensing for the operating system, which we consider commodity software.

However, allocating the money saved on OSS(Open Source Software) for building richer features leads to a more significant differentiator and more completive gain. Again, this is the reason we have projects in the first place (in the business world).

Open Source in Closing

Open source licenses typically eliminate the licensing cost. This drives the price down, just what we want for commodities.

There are sometimes compelling reasons to select a database with a premium cost, for example. And paying for licenses can sometimes justify a lower TCO(Total Cost of Ownership). However, many times using OSS(Open Source Software) for some (or most) of the components of a project makes sense.