Perceived disadvantages of open source models

Of course, open source development models lead also to the perception of some disadvantages. However, some of them are only disadvantages if we are stick to classical (proprietary) development models, which is of course not the case with open source. Let us review some of them:

• There is no guarantee that development will happen. In other words: it is not possible to know if a project will ever reach a usable stage, and even if it reaches it, it may die later if there is not enough interest. Of course, this is also a problem with proprietary software, but it is more evident in the case of open source. Especially when a project is started without strong backing from one or more companies, there is a significant initial gap, when the source base is still immature and the development base is still being built. If it is not possible to get funding or enough programmers cooperating at this stage, the project just ``dies'', or perhaps slowly fades out. Usually, when it reaches a self-sustaining level, the user and development base is such that it can proceed by itself, without other external incentives. This issue will be revisited later on, when business models are discussed in section 5.

• There may be significant problems connected to intellectual property. This point is especially important, now that some countries are accepting software and algorithm patents. It is very difficult to know if some particular method to solve a software problem is patented, and so the community can be considered guilty of intellectual property infringement. Some open source packages are already addressing this issue with switches or patches that enable or disable patented code fragments according to the country where the code is used. In other cases, developers consider source code not as an executable device, but a mere description of how a device (the computer) executes, and therefore uphold the idea that source code is not by itself (in absence of an executable program) covered by patent law even in countries where software patents are accepted. However, this idea has still to be tested in courts, and many people do not believe in its viability. In any case, it still leaves problems for the users, who need the executable programs. Although the issue of software patents is a problem for the whole software industry, open source is probably one of the more clear cases where it can be shown how they harm the regular process of software development. The specific problems are that availability of source code simplifies the detection of patent infringements by patent holders, and that the absence of a company that holds all the rights on the software also makes it difficult to use the mechanisms in use by companies to defend from patent litigation, like cross-licensing or payment of royalties. These issues will be discussed with some detail in subsection 6.2.

• It is sometimes difficult to know that a project exist, and its current status. There is not much advertising for open source software, especially for those projects not directly backed by a company willing to invest resources in marketing campaigns. However, several `aggregation points' for open source software do exist<sup>15</sup>, although in many cases they are usable only by experts, and not by the general public. They are also in many cases very specific to some software category, such as scientific software or database systems. There are only a few `clearing houses' for open source software and projects, and in many cases they are not really up to date. However, some people see this fact as a market opportunity, and several companies with experience in Internet based information services are approaching open source software with added value services which maintain information useful for people or companies trying to locate or evaluate open source software of some given characteristics. In fact, a new business is emerging for providing consulting and searching services related to the selection of open source applications for specific needs. For the general public, some more education is still needed before the regular user can approach these services and get a solution to her software problems in terms of open source software.

It is extremely important to `see' through the various interpretations of the advantages and disadvantages of open source, and if possible try to analyze with quantitative methods if open source can be helpful in a given situation, or for a given user or company. From a developer point of view, a good example of the perils of launching an open source project is the Mozilla project (launched by Netscape in 1998), considered by some people a complete failure because it took more than an year to get a first beta out. On the contrary, other people perceive, with respect to the same project, that the open source community was capable of producing a totally new code base (larger than the Linux kernel) starting with only some new infrastructure. In fact, now half of the developers of Mozilla work outside of Netscape, and the Mozilla browser has begun to be embedded in many innovative devices, thanks to its modularity and flexibility (for a detailed discussion on the evolution and results of the Mozilla project, read [6]).

Footnotes

... exist¹⁵

The most widely known is Freshmeat, at http://freshmeat.net.