Open source is years behind commercial products
Well that got my attention as a viewpoint! Especially as it was a Computerweekly.com article quoting the Socitm president Richard Steel.
I had been perusing the net to see what the reaction was to the recently announced government's policy on open source software , and came across this fascinating debate. I guess it struck a chord with me as it tends to be the common theme in my daily grind, working with organizations and companies that are considering whether or not to pursue an open source content management route.
So, I followed the thread and read the blog from Richard Steel. Herewith the main points of Mr Steel’s blog:
- Open Standards are definitely required.
- I don’t like the term “Open Alfres”. It’s misleading; what many people mean is “anything but Microsoft”; few businesses actually use open source directly – they buy software derived from open source that has been commercially packaged and sold with support, which, in practice, is little different to licensed software.
- Nevertheless, competition is great for keeping suppliers focussed on delivering customer value, and “Open Source” has certainly played its part.
- All the same, software is only one part of the Total Cost of Ownership equation; don’t consider it in isolation, but as part of the full TCO and lifecycle costs.
- “Open Source” software development, in my experience, lags proprietary development by several years. I don’t think we could achieve the anytime, anywhere fixed and mobile infrastructure with tele-presence we require, now, for flexible and new ways of working using only Open Source.
- I agree with reuse, and it’s a very significant factor in the Microsoft Public Sector software licensing project I’m involved in (and not allowed to talk about).
- If it works for you – fine. I wouldn’t rule-out so-called “Open Source”; Newham has used it for some applications since the time it did its deal with Microsoft (probably the first UK public sector procurement of Microsoft as a supplier) and continues to do so.
So, I have to say, I think there are some very valid points made by Richard Steel, and in my experience they often get overlooked.
I agree that the term ‘open source’ is misleading, and as is the way with all coined phrases, it conjures up different perceptions for different people. The majority of my interaction with open source is in the presentation of Alfresco open source enterprise content management ( www.alfresco.com ) to interested parties. This basically means that all of my discussions are with people who are considering whether to invest their precious budget in procuring a subscription to an enterprise open source subscription, or not.
Therefore I get to hear the real thoughts, fears, objections, misconceptions, and I have to say. It is utterly fascinating.
On more than one occasion I have been drinking coffee with a CTO of a Local Authority, and explaining away the fear that an open source product is not actually a bunch of spotty teenagers and train spotters coding away in their bedrooms, and that if they all suddenly get girlfriends then the product will fade into obscurity. Who would have thought that such a basic misconception still exists in 2009? Interesting huh?
Anyway, having designed, built and delivered over 20 open source enterprise content management projects to date, the common theme in every procurement has been a focus on features and functionality every time. The fact that Alfresco is open source has been a secondary factor, and frankly – that is the way it should be. If the product was inferior to other systems I am extremely confident that none of our clients would have gone down the Alfresco route.
Mr Steel’s point about ensuring that the Total Cost of Ownership has been evaluated is exactly right. You can not, and should not, avoid a well thought out evaluation of technology in the light of your specific requirements, and to assess the implementation costs, the long term support costs and the on-going product roadmap of your selected technology.
If your selected open source product is a flash in the pan – meeting a need today - kind of application, with no infrastructure to constantly develop the product to meet future needs and demands, then you could be faced with an expensive replacement project in the near term future.
Without considering these things, you will not be best placed to stand over a decision to invest in open source, or not.
However, if you do find that the product meets your functional, technical, and business needs, then I would be doing open source an injustice if I didn’t point out that having no capital expenditure on a license fee is a major advantage – especially if you have a specific budget that you have to work within.
For example, if your total project budget for a new website, or a new document management system is £250k. You go through your complete due diligence and selection process, and have selected a proprietary content management system. You then pay £100k up front for the software for example, and have to budget for an annual £20k support and maintenance contract. This leaves you with £130k, which is probably going to buy you 150 – 175 man days of resource to cover design, architecture, build, configuration, system test, user test, deployment, and transition into support, with a potential additional cost for application support thereafter. This is not a lot of resource to deliver such a project, and therefore you typically work with a lot of ‘out of the box’ functionality, and deliver a highly prioritised, minimum list of functionality back to your business.
If however the CMS you selected was an open source product, you would only pay approx £20k for annual support and maintenance, providing you with exactly the same kind of SLA’s, quality assurance, and product roadmap, leaving you with £230k to spend on design, architecture, build, configuration, system test, user test … etc …etc … you get the point.
If I was to cite a specific real world example, without naming any names, we undertook a public sector project, supplying and implementing an enterprise open source subscription for Alfresco to replace an existing proprietary technology that was currently in use. Although the existing technology had been procured a few years earlier, and the license fee had already been written off, the organisation was still paying over £50k per annum for the support and maintenance contract.
Their subscription for Alfresco web content management, document management and collaboration, on a 4hr SLA contract was just over £22k per annum.
So before they engaged any Ixxus Alfresco consultants, they were saving over £28k per annum. Cool huh?