Open source has been hailed as the obvious solution for saving money, being flexible, more transparent and avoiding project failures. After attending a recent open source event in London, I came away confused and it made me think about what open source actually means here in 2010. Source code access hardly seems either relevant or valuable to most online professionals. In addition, many commercial proprietary vendors, e.g. IBM, Microsoft, Oracle have bigger and more vibrant communities than most open source projects. That's being the case, what's the real benefit of open source?
The event was called: Public Funds in the UK: Open Source for Document and Content Management? Let me start with an excellent quote from the day:
Bernard Woolley: "Well, yes, Sir...I mean, it [open government] is the Minister's policy after all."
Sir Arnold: “My dear boy, it is a contradiction in terms: you can be open or you can have government.”
It was former Ovum analyst Mike Davis who cited the very first episode of the BBC comedy series Yes Minister first broadcast 30 years ago in order to very eloquently express his view on the state of open source adoption in the UK government.
My talk was based on last week's commentary on Open source doesn’t always represent best value and I covered the confusion and several issues with open source, including support, risks, requirements, implementation and community. Illustrated with a few case studies from our European members, I tried to look beyond the hype and shared my recommendation regarding open source. You can download a copy of my slides below.
As mentioned initially, I came away confused. The event was interesting and organized by UK project management guru Graham Oakes and run by the British Computer Society. The event was free to attend and there was a good mix of speakers, including vendor, agency and user perspectives. Moreover the BCS served a great lunch. I learned that when digital agencies respond to a technology selection tender, they tend to avoid open source unless the tender explicitly states anything about open source or typical open source requirements, such as active number of contributors or source code access. If in doubt, they'll suggest a proprietary commercial alternative thinking that the customer might have reservations about open source.
Dear reader, perhaps you can help me out? Am I am the only one thinking that:
- open source has lost its meaning and moved far beyond source code access to either a way of life for some very transparent vendors, or a new way to capitalize on software and attract snazzy investors
- open source does not have any real benefits for the buyers
- open source does not equal open standards. And, by the way, which open standards (except those from the W3c, e.g. HTML and XML) are used and referred to by all the analysts and vendors?
Here's another relevant and great quote on the law of inverse relevance from the same Yes Minister episode:
The less you intend to do about something, the more you have to keep talking about it.
This certainly seems to be a very valid statement about open source adoption in 2010. As long as the real open source benefits remain unclear, at least from the buyer perspective, my message to both government and private buyers is: Don't rush into open source.
Thank you to the many members and other peers who contributed with helpful feedback to my talk. Download slides: Open source doesn't always represent best value (PDF, 1 Mb)
Other perspectives from the day:
- Jon Marks, LBi: BCS Open Source Presentation and An Incomplete Directory of Open Standards
- Will Abson, Alfreco has shared his talk on Slideshare
It was a fine event. Pity the beers afterwards were cut short by your mad airport dash. Anyway, I’ve tried to answer your question “which open standards”. I give you:
An Incomplete Directory of Open Standards
http://jonontech.com/2010/01/10/an-incomplete-directory-of-open-standards/
I wrote a response to your question from the open source side, in fairly general terms.
http://blog.technologyofcontent.com/2010/01/open-source-and-content-management-for-janus-boye/
In terms of the specific questions here, I think the key points are that the community is enabled by the sharing of source code, as that is a minimum requirement for truly empowered community; the sense of community fostered by what is enabled with open source makes it much more important than say the communities around Oracle or Microsoft. Code is so much more important than many people realize.
I do also think it is key that the role of open source in content management is just beginning, and the bigger effects are just starting to be seen, and the rise of standards and agreement about core requirements is building a broader consensus and allowing many companies to start to collaborate on open source cross vendor software, as seen with CMIS. Expect much more to come.
Implementation, as discussed elsewhere is a key factor, and of course as implementation may be the majority of the budget this matters too. but a lot of the same questions may apply here – parts of implementation may be software development, and owning the code source here may matter more than owning the product code, in terms of reducing vendor lock in. And open standards in implementation are key too to lowering cost of future modifications and being able to take control internally. So looking at implementation work with the same typee of analysis as the product is a useful thing to do.
A very interesting article Janus,
It is nice to see some of the analysts looking into the other side of Open Source. I, being a frequent WCMS developer works with both open and closed source systems (JAVA and .NET). To an extent, you are 100% right; the source very often does not need to be open source for a developer, – and code, while being important, is not as important as a good extensible architecture.
Just as long as the WCMS vendor provides a real, pluggble interface, with opportunites to override the standard functionality, this is actually quite ok: And sometimes I even prefer closed source: Because such systems, being closed, has been carefully designed to allow 3rd party developers to customize it. This means that you are bound to modify the system making extensions. And very often, by not modifying the core product, it is much easier to upgrade the software.
Off course, some (most) closed source systems does not provide sufficient extensibility layers, – and this is where open source excels.
Also, from a monetary perspective, I see two kind of Open Source; those who are run and formed by vibrant communities, and those developed by companies. Each has benefits, – of course; 100% community driven WCMS is often cheaper in the very end, because no company has to make money out of it. But they are also way harder to get a “phonecall” help by, – and therefore often not a good choice for somebody with a business critical web site. Company driven OS WCMS has often lesser communities, – and you must in many cases purchase services in regards to support, training and training material. In the other hand, such vendors supplies a services interface which may move these systems into the selection list of companies in need of a business critical web site.
One thing, though (now I might be hammered by the Open Source Community). None of the Open Source WCMS compares from a functionality perspective with the large (and often expensive) systems such as Interwoven, Day, (RedDot, once it lived), Sitecore or Tridion:
I often hear, “for that money you can develop that functionality yourselves”. And often, this is true. But it is really just a matter of functionality and requirements matching. Does the WCMS provides the functionality you need? I think, software costs is less important; because our customers are typically spending 5-6 times the cost of the WCMS for the implementation. Too much functionality tweaking, to get the system to do what you want it to do, and the free or cheap solution becomes expensive.
Thanks for a great post, – with an alternative and perhaps critical view of Open Source. This discussion is much needed by independent analysts!
> One thing, though (now I might be hammered by the Open Source Community).
> None of the Open Source WCMS compares from a functionality perspective with the
> large (and often expensive) systems such as Interwoven, Day, (RedDot, once it lived), Sitecore or Tridion:
I don’t hope you get hammered*, despite open source and beers often goes hand-in-hand ;-)
However, I like to ask the question – do you miss any of the features from the large systems? We (Umbraco) win markets because we have less features, resulting in a more elegant, easier to learn and better performing system.
Our main wins comes from more experienced buyers who’ve already experienced the non-joy of cluttered software with mandatory maintenance contracts giving them more features they don’t need with every update ;-)
Best,
Niels…
* The Danish term hammered doesn’t translate very well (http://www.urbandictionary.com/define.php?term=hammered)
Hi Janus
I too have taken up the theme of your conversation with a posting here:
http://ecm-stuff.blogspot.com/2010/01/open-source-in-content-management.html
As every, thanks very much for thought provoking commentary :-)
Janus,
I’ve tried to add some of my thoughts to this… ended up being a bit of a long ramble, but hopefully answers some of your questions. In short, Open Source always has been more than just the license. And as it starts to become more widely adopted and ‘normal’, and hence less a decision factor, it doesn’t mean it is any less valuable.
http://www.netsight.co.uk/blog/2010/1/11/open-source:-does-it-matter
-Matt
Here is my followup thoughts on the matter:
“WCM Field Notes: Give Open Source A Chance”
http://www.cmswire.com/cms/web-cms/wcm-field-notes-give-open-source-a-chance-006369.php
It’s probably too long for you ;-)