The visible bit of the software, the user interface is, as its name suggests, high profile and important to the system user. It's the bit they see, and it's therefore the part of the system that frequently determines whether the software is considered to be good or poor value. It is so important for the installer to be able to sell a software product that will do the job and delight his customer. Success with this is very likely to bring additional business in a way that tidy cables in a power supply cabinet will not.

Features and usability
One of the worst types of software product is one that has just been allowed to evolve over the years. Applications suffering from this syndrome may have started off being a logically written package, but, over time, features requested by customers are bolted on. If this is not handled properly by the manufacturer, and with courage (sometimes the courage to say no!), the software becomes a real mess. It looks unattractive and has too many features making it impossible to find the basic feature you want. Another common failing among software writers is to assume that the more features you have, and the more flexible and powerful the features, the more you can charge for your work. Unfortunately, many features are put in for the wrong reasons. (I have even known designers include a technology because it's trendy and looked good on their C.V.) Unless the feature is a core part of a system user's requirement, it's a distraction, and has negative value. Less is more! Software Engineers are sometimes reluctant to discuss real customers' needs with them. This communication breakdown is another reason for software lacking common sense features.

There is a culture in some IT circles to scorn system users or installers who haven't memorised every wretchedly written word in a thick manual before daring to ring a helpline. Just as a waiter should not ask his customers to wash up, support departments should not turn away their customers in this way. If the question is about a basic feature, the manual may not be required - the software should make the task obvious. Obscure features are more likely to need explanation, and a friendly and efficient support department is an essential resource for the busy installer. Companies who protect their support technicians behind email, automated telephone systems, group meetings or any other such barrier will struggle and lose against competitors with good support departments.

Licensing - yes or no?
Licensing is a thorny issue with many software vendors. They rightly want to charge for their products in order for their business to be viable. However, software can sometimes be sold at too high a price, and can have attached unforeseen conditions. A per seat license fee for a client server application can come as a last minute shock unless it has been planned. In general, the more an application is locked down with copy protection, the more likely it is that that protection will become a serious nuisance to the legitimate user at some time in the life of the product. From coded software key files which become corrupted, to a hardware protection device that fails, copy protection can cause irritation at best and disaster at worst. If a protection device fails and the manufacturer has ceased trading, there is no way for the system user or their installer to recover the situation. As the supplier, the installer could become liable for the costs of such a failure.

Free software
A recent trend has arisen for software to be given away free. At one time, such offerings were seen as risky and potentially of poor quality. This has changed as mainstream providers have observed the problems of licensing mentioned above, and the value of mass adoption of their software. Without the unreliability introduced by copy protection, free software is easy to install and use, and operates as a powerful vehicle for selling the chargeable features. There are still many vendors stuck with the idea that not much should be given away, and these companies will fail should they try to join the free software world half-heartedly. Only the minority of the application's function should be chargeable as this encourages a huge active user base and brings in more revenue as a result. In addition, the large number of people using the software ensures that bugs are reported, and quality is improved. The logic of this is that the more you have to pay for your software, the poorer the quality is likely to be.

The challenge for installers
Familiarity with software and the operating systems under which it runs is the best preparation an installer can make. Learning how to use a mouse on site whilst commissioning a system is a startling tactic, but it has been known! It seems an obvious thing to say, but when an installer has been round the block with a piece of software, it really shows and his client is more likely to place more business with him if he seems rehearsed. Again, the manufacturer can be a great source of help here. Is there demonstration software available, or is it a flashy imitation put together in a presentation package? The latter is of no use. Also, the demonstration software should be the real thing, so that scenarios particular to the client's requirements can be tested and demonstrated; they can even be left running for free on the client's PC. Is demonstration hardware available, and does it look the part? It makes a much more convincing sales argument than just a standard letter and a slither of leaflets.

The challenge for manufacturers
Having cautioned against over complicated software, I predict that the future is an inexorable path to more and more intricate systems. Fantastic applications facilitated by TCP/IP connectivity and systems integration bring such huge benefits that they are worth the extra fiddling. As in modern cars, much of this intricacy will be under the bonnet in well-designed software - but the installer may occasionally have to peer at the engine. This means that the installer, like the modern mechanic will have to be more sophisticated. However I do think that well designed electronic security systems will have more and more good diagnostics built in. Bugs will appear with more complexity. The way in which a manufacturer deals with these in terms of access to support staff, and fixes available quickly for download will be an index of their competence as well as their ability to minimise the bugs!