I think you have to ask some questions first.

The most important one is whether the software product/tool is meant to be used in conjunction with a hardware product that you sell. Usually in this senario the software tool or utility is "included" and itself does not represent a separate revenue stream. This means that the decisions that you make as to what environments to support will be driven by quite different factors than a for sale software product.

Another consideration to be made is to contemplate how future looking that you want to be. Is this software product one that would be better implemented as a an "App" that works on the current trend of 'carry around' type devices instead of conventional PC architecture devices? Windows and Linux devices that are prevant today will be around for a long time but do they address the needs of the user audience say 3, 5 or 7 years from now?

It also seems important to ask the key questions regarding the rate of return that you get for a particular part of the development effort relative to the energy, time and cost involved with doing that development. Having Linux support may be great but if it represents 35% of the development effort but only supports 2% of your user base it becomes really necessary to make some strong economic type decisions.

Michael Karas