Hitchhiker's Guide to Software Architecture and Everything Else - by Michael Stal

Monday, March 13, 2006

The Open Hype Scale

To earn money in the [IT] industry one of the potential preconditions is the availability of innovative solutions in your products. Innovative can mean a lot of things, for example faster, better, easier, or nicer. Only seldomly this kind of innovation improves or extends the functionality. Application domains have turned out to be very stable over time. So, what could a text processor do more than editing, spelling checking, or printing? On the other side, it is impossible for a vendor to publish a product and then live with it for years, without any regular updates. Same for other industries. What if your car would be the nicest, fastest, energy-saving car in the world with all features anyone could ever wish and what if this car could never be damaged by accident or age? Consider the truth: No one would ever buy a new car anymore. End of car industry. Point. Same holds for software development. That's why your new brand of software tools support MDA, UML, SOA, AJAX, VIIV, you name it. That doesn't necessarily mean that software vendors themselves invent all of these technologies, each of them promising a cure for all world problems. But if someone comes up with such a potential technology innovation, all vendors will jump on the wagon, no matter whether this technology is really useful. At the end of the day all of the software engineers, business accountants, and other personel must receive their cheques. So do I and so do you. It is essential for the IT industry to always improve and innovate, even if some of these improvements might be questionable. But don't blame our industry. Do you really thing that all of these features in your mobile phone, LCD-TV or car are necessary? Nonetheless, there is one implication for us architects and engineers. Make sure that your management never gets addicted to a new hype. I remember management guys to be totally convinced that the invention of components and their wiring with graphical IDEs would solve everything. That were the stone ages, you might think. But just consider the current SOA Hype. Most hypes come in waves and even if they don't succeed they reappear again wearing other clothes. To be able to prevent the worst, look at all of these new technologies very soon and try to keep yourself informed about all their strengts and weaknesses. And we as engineers should also be careful not to enter the trap. It is easy to come addicted to all this stuff as it promises to improve our lives as architects and engineers. Be realistic!


Post a Comment

<< Home