Open Source in the mainstream
Isn’t it all just software?
Comment A few things have appeared from various sources lately resurrecting the old discussion of whether Open Source software is “safe” or “right” for mainstream adoption. Whilst many of us consider this issue to have been dealt with long ago, there still seem to be some out there who want the debate to continue.
The problem is, however, that such debates often get muddled, confusing discussion of the Open Source approach with the pros and cons of specific Open Source solutions. It is relatively common, for example, for people to use the terms Open Source and Linux interchangeably.
As part of this confusion, there is then a tendency to regard any challenges with a particular solution to be somehow associated with its Open Source status. Desktop Linux is a prime example of this. Whilst research indicates that early adopters have often experienced problems with software compatibility and availability, for example, this is simply a function of Linux not yet having achieved critical mass in the desktop environment - it has nothing to do with the origin of the software. Any new commercially developed desktop operating system released into the market today would have to overcome the same hurdles on the way to mainstream acceptance.
It is easy to see how confusion occurs, however. Solutions coming out of the Open Source movement are sometimes perceived as being relatively immature at a generic level, because at any point in time we can identify many offerings that are young and/or have not yet achieved the critical mass we were referring to before. But again, this is not a function of the Open Source approach per se, but of the rate the community is growing, and if we look elsewhere, we can see exactly the same dynamics in some parts of the commercial software industry – e.g. in the mobile and wireless solutions space. In the meantime, there are clearly many mature and well proven Open Source solutions out there – Linux (on the server), Apache, JBoss and MySQL - to name a few of the obvious ones.
There is then the more fashionable debate about software innovation and Open Source developers not having access to the same R&D budgets as the big commercial software developers. There is a whole philosophical discussion that we won’t get into here, but looking at it pragmatically, it is clear that there is creativity coming out of both camps, suggesting that the approaches they use are just different ways of achieving the same thing. And let’s face it, with organisations like IBM, CA and others donating significant amounts of resource, funding and intellectual property into the Open Source arena, can we really unravel who is driving what kind of research and development and on which agendas?
But having gone through all this, the critics then come back with the supposed killer blow of Open Source being a complete lottery from a maintenance and support perspective. This might possibly be the case if you are trying to run your business with software installed from a disk stuck to the front of a consumer magazine, but there is enough money to be made from delivering Open Source solutions now for mainstream suppliers to back them up with full blown support and maintenance services. As an example, when we spoke with some of the management team at JBoss recently, the conversation was no different to the equivalent discussion with a traditional commercial software vendor – it was all about customer needs, component bundling, product lifecycle and release management, software maintenance services, routes to market, integration partnerships and so on.
These kinds of developments, along with the commercialisation of Open Source by the likes of IBM, Dell and HP might feel a bit uncomfortable for those at the other extreme who advocate free software and free community support. The reality, though, is that IT Managers and Executives feel more confident with traditional solution delivery mechanisms with clear supplier responsibilities. This can only be achieved if money is available to invest in technical staff, call centre operations, etc.
Commercialisation of Open Source also helps to fund the competitive drive into the mainstream, which, quite simply, requires cash and a mechanism for generating it on an ongoing basis. Without this, battling with the marketing machines of commercial software companies and creating interest and activity in the IT delivery channel is extremely difficult.
So, what does all this mean from a customer perspective?
Well, the obvious conclusion is that most organisations should probably not get too hung up on whether a solution is Open Source or not as this is unlikely to be a significant factor in determining the capability of the product, quality of support, coherency of release cycle, and so on – provided, of course, you are using an appropriate source. Maturity, skills availability and cost of acquisition and ownership are all considerations too, but again, none of them are specific to Open Source. Even the value of savings on licence costs needs to be considered in the overall context as other factors, such as cross training costs, different operational and support requirements, etc may neutralise the benefit in some cases. It is impossible to generalise in this area.
In terms of practical advice, we would always recommend people include at least a couple of new or different alternatives on the evaluation list when going through a significant selection process, then offer them up against the business objectives alongside the more obvious solutions. If you do this without prejudice, you will find quite naturally that Open Source solutions will increasingly end up in the mix and it is important to include these rather than dismissing them purely because of where they come from. It is then a case of evaluating options on their merit in the context of what you are tying to achieve. Whether individual solutions are Open Source or otherwise is largely irrelevant to the process.
After all, it’s all just software and services at the end of the day.
Dale Vile is Research Director at Freeform Dynamics, an independent industry analyst firm.
Sponsored: DevOps and continuous delivery