Introduction

A logical consideration for applying OSS might be the costs involved. Because the software is often free to get, it seems much cheaper to switch to OSS, but not only the purchase of the product matters. Other costs involving hardware requirements, maintenance, merits for the company and the instruction of the users are all influencing the total costs. Check out this comment on costs (http://www.nosi.net/node/27|) as well. It gives a nice oversight.

All these factors have to be considered. First of all we should make a comment on the idea of free software. OSS is not always totally free of costs. It depends on the license that comes with the software. Some software comes with special modules with extra functionality or services that should be paid for. Others (called shareware) ask for a voluntary contribution. The idea behind OSS is mainly based on a software development model that offers openness, quick development, flexibility and conformation to standards. It needn't be gratis although this is often the case.

Integrating it into the organization

Integrating the OSS in the organization requires two things: investments in knowledge to install the software and make the necessary changes to the profiles of the users and/or the system on one hand, and educating the users that work with the software.

It depends on the kind of software, how much the integration will cost. If the software is used by end-users that are not software experts, some training is always necessary. Although the same holds for proprietary software, there is quite some software on the market from which the user interface is similar or equal to that used by the end-users at home. Take for example the programs of Office: it is hard to find people that never worked with programs like Word and Excel before. Of course such a user needs less or no training to use this software.

Another argument in favor of proprietary software, which is used by Microsoft for example, is that it is easy to start with. A lot of effort is put in the usability of their products. Although the situation is improving, the usability of OSS was not always that good. The community was driven by software experts that put more effort in making the software fault tolerant, robust, bug-free and elegant. An easy to use shell around it was not so necessary. But since the OSS is moving towards the consumer market and is now also considered by companies and institutions, developers are more aware that �real persons’�have to work with it too.

A counter-argument stated against this, would be that the software from proprietary software vendors doesn�t do what it promises at first glance by his fancy User Interface. In the end though it should be said that the user interface of OSS is poorer and a lot could be improved to gain on this headway of the proprietary software. A good article on |usability (http://www.cs.waikato.ac.nz/~daven/docs/oss-wp.html) is written by David M. Nichols and Michael B. Twid from the University of Waikato in New Zealand. Of course, if the software concerns a server, the usability issues are restricted to the way this server influences the daily work of the end-users. In most cases the impact of migration to other server-software is very low or almost to neglect. It may be for this reason that the very stable Linux (see for example: this survey (http://www.linuxworld.com/story/33838.htm|)) and Apache software has replaced lots of Microsoft Servers. Today already 63% of the servers are based on OSS like Apache (See: [1] (http://www.nosi.net/node/32)). This kind of server-side software has showed to be of good quality and cost-effective. (They calculated in ([2] (http://www.nosi.net/node/27)) that to maintain a Linux server will cost half of what a Microsoft server would cost within a period of 4 years). Of course the technical personal has to be instructed how to install and maintain the software. Depending on the experience of the in house software engineers it might be wiser to have an external consultant be responsible for the whole integration of the software. In general the OSS might be a little more complicated to install if your existing consultant is not familiar with this software.

Maintenance

It is hardly never the case one has to pay for new releases. It is right away available to download via the web. Sometimes payment is necessary for special support. But often this is free too �although not garanteed- or these services are not really needed (for small companies e.g.) reducing the costs to nil. When bugs are found one can influence the development team by contacting them directly, which is not the case for proprietary software. This way, possible evolving problems might be sooner solved than would have been the case otherwise.

In general the OSS tends to be of good quality and in lots of cases the software turns out to be quite reliable. Once everything is installed well, the main maintenance costs will be upgrading and repairing crashes and/or bugs. If the software is quite stable, the maintenance is merely reduced to do administration and upgrading once in a while. Although this seems good news, the latter might be problematic for a system administrator with less technical (OSS)skills, making a more expensive consultant necessary.

Hardware costs

Most OSS does not require specific hardware and so you could run it on the same hardware you had or buy relative cheap standard hardware for it to run on. As opposite to proprietary software that requires sometimes special certified hardware to run on. In general hardware shouldn’t be an obstacle with respect to costs in applying OSS.