We talk a lot about new technologies but less about some of the implications. Take the software provisioning model, yes it’s all about ‘as a service’, but go a little further into this. Yes, we are changing the provisioning of software, but that is because the purposes people want to use software for are also changing. It then follows that the development of software itself will also change.
Little more than three years ago open-source software (OSS) was being positioned as something not suitable for the mainstream IT market with a series of implied risk statements about its fitness for use at an enterprise level. The use of Linux as a low cost operating system was acceptable, as was the increasing use of the so called LAMP stack, standing for Linux OS, Apache web server, MySQL database and PHP, Perl or Python scripting languages, to support web-based activities. Today many of the major proprietary software vendors who were vehemently attacking OSS have moved to embrace it as part of their product portfolio and can be found making contributions. In the USA the White House website has been shifted to OSS, in the UK the London Stock Exchange has adopted it for its high profile ultra reliable environment, and in Norway using proprietary software has become the exception with OSS the de facto approach.
The original debate on the use of OSS focused almost entirely on cost, the standard rating value of conventional IT, but the reasons for adoption have changed as the business and technology environment has also changed over the last two years. Today, time to change and frequency of change are big issues as enterprises struggle to cope with recession, global competition, new markets, as well as the overall shift to an ‘online’ business world. The major issue has become the terms of the license for the use of software. Proprietary software is licensed in a manner that is deliberately restrictive, and deterministic with respect to the number of users or processors, as well as the fact that access to the core elements is closed. All of this made sense in the context of the original enterprise use, the more limited skills in software development and the need to guarantee a revenue stream to pay for bug fixes, upgrades and other ongoing requirements.
Today the freedom of OSS licensing models, in combination with a development model, suits the faster moving, more distributed and fragmented use of software across the entire enterprise. Indeed the people-driven activities taking place at the edge of the enterprise and the marketplace can arguably only be served by the OSS model. Why? Because it is comprised of an unknown number of people, on an unknown set of processors whose interaction is neither predetermined, nor likely to be long lasting!
The acceptance of OSS is as much a feature of new business demands introducing new usage models as any other factor. These are the same factors that are driving the cloud as a new way of provisioning technology and software towards a distinctively different set of business activities. This shift in business priorities shows clearly in this analysis by Gartner made available on their website for download, as does the misalignment with traditional IT priorities! Every one of the top ten business priorities, except one, is focused towards the edge of the enterprise, people-centric activities and markets with the likelihood that the OSS license model will be required.
Quite frankly in this new world the question might well have become ‘is there any reason not to use open source?’ And then the rest of the question might be ‘is it time to review our overall approach to development for this new environment and set of business requirements’?
Kind of puts a different spin on the whole open source debate when you think like this!