Open source doesn’t always represent best value

Despite much hype we did not see a breakthrough for open source CMS last year. When I launched the discussion last year, we  received some great comments, e.g. on intellectual property and warranty, suggesting that in some cases open source is not the right decision.

In the past decade, several governments have issued statements with strong support for open source, e.g.  UK government backs open source and Denmark’s endorsement of Plone. Often these statements were driven by an underlying desire to save money and drum up competition for Microsoft and the de-facto Windows & Office monopoly.

Our usual advice is not to start by deciding on open source or not. However, in our community of practice, many technology selection projects often start with a debate around whether open source is good or bad. Many members report that they have experienced quite expensive open source projects, indicating that open source is not always cheaper.

The open source debate is often based around emotions, eg. the very strong urge to avoid Microsoft. Cartoon by Hugh MacLeod

A significant factor in terms of value is the cost and quality of the implementation. If you’ve selected an open source system, say Drupal, WordPress or Umbraco, for your new website, but cannot find any experienced implementation partner, then you may be forced to take a step back and rethink your selection process. You might have enough resources to do the implementation yourself, but I don’t recommend doing it without proper training and expert assistance. Most open source projects have really weak documentation.

Also, if you don’t have any resources to engage in a vibrant open source community, you are missing out on one of the big advantages of open source. Except for the really big vendors, e.g. IBM, Microsoft, Oracle, which have decent communities for developers, most commercial vendors don’t have communities where you can meet other practitioners and share experiences.

If you are concerned about risk, it is worth noting that some relatively well-known open source projects, e.g. Mambo got in trouble back in 2005 when most of the developers associated with it decided to start Joomla. HyperContent, another open source CMS, was announced dead in 2008. Commercial systems don’t live forever either, but typically you can continue to buy support from the vendor.

Your requirements may deflate the value of open source. Those with strong requirements for Microsoft Office integration, e.g. seamless Word integration, might struggle to find an open source solution that support the requirement, while many commercial alternatives have offered this for 5+ years.

The past decade saw the rise of the so-called “commercial open source vendors”, e.g. Alfresco, eZ or Jahia. These vendors have open source solutions, but earn their money on selling enterprise licenses, training, and support agreements. Some even do consulting. In their own words, they provide the best of both worlds, although I’m yet to see any of these firms develop a serious community.

In your view, when does open source software not represent the best value?

I am presenting on Thursday 7th January in London at a free event run by BCS – The Chartered Institute for IT on Public Funds in the UK: Open Source for Document and Content Management. Whether you can make it to London or not, I invite you to participate in the discussion by posting a comment below.

“I think you are getting it wrong.

Companies buy open source because is better, not cheaper.

To think about open source as free as in beer is quite right but no the only criteria for choice. It is true that with open source you don’t have licensing cost upfront or in every upgrade. No cost of ownership.

It is also true that it is reliable, precisely because it is open. What about Apache, Linux, NetBSD, Samba, Drupal, SSH, Zope, etc…
Apache is the number one web server on the Internet with more than 106 million websites, far from Microsoft. Linux is the number one server operating system on the Internet. I ran Windows servers before I changed to CentOS and my headaches are gone!
Open source is contributed by passionate people just for the sake of making it better, not just with financial interests. That, IMO, is the best ownership. For most closed source software, that feeling of responsibility that any one and every one can see the code and hence you might be held responsible simple won’t be there.
Open source enhances and improves security (due to the open peer review, open source software bugs are usually fixed before they are exploited) and encourages open protocols (internet’s and world wide web’s recently rapid growth is based on open standards and open source code, such as BSD’s TCP/IP and DNS code, the NCSA and Apache web servers, and the Sendmail email routing software).

Support is normally where anti-open source advocates come in. I honestly think that open source offers the best support, in cases you have competent people in your team (who is easier to find than for a closed system), the community or in the same way you get it with licensed software, via a vendor or shop.

Every time you can find more and more specialized vendors supporting open source who are reliable, and I am not talking just about Red Hat or IBM, if you choose Drupal for your CMS, companies like acquia, Development Seed and others.
You mentioned that all these “commercial open source vendors” like Alfresco do not develop communities like Microsoft… you are wrong. The communities are around the products they support not themselves. Same for Microsoft. You have communities on Windows, Office, Sharepoint…
You talk about risks? Like in the case of Mambo diverting into Joomla… This is normal!! you can always take an open source project and fork it! I see more risk in betting for a closed product that if it dies, you are basically dead as well.

If we talk about Drupal it powers websites like United Nations (end of poverty campaign), World Bank, Warner Bros, Discovery Channel, AOL, Sony, NATO, MTV UK, BBC, the Onion, NASA, Greenpeace UK, New York Observer and thousands of others.

You mention documentation worse in open source, and I couldn’t disagree more. It is often better in open source as well. Most popular open source solutions are extremely well documented and a variety of free and commercial technical support options are available. Due to the nature of community development, documentation and instructions are often written from a variety of viewpoints — creating well-rounded information, instruction and tutorials. In addition, open source projects can’t hide usage techniques, due to the free availability of the code. Free technical support is often available in the form of mailing list or newsgroup discussions.

In many cases administrators prefer closed source because it offers some kind of protection in that you can always say “we are talking to the vendor, they are working on it…”. So, the responsibility is conveniently shifted to the vendor. And this ripples up all the way to the top management. Net effect? Problem remains unsolved. Most surprising fact is that nobody seems to bother about it, but as I said every time more open source vendor are coming into the market so if there is no expertise in house you can turn to your vendor, the community or hire a couple of interns to help you developing at any level (not just through APIs, in the best case, or via the vendor…).

I would recommend to do not fall for the sales pitch. The samples they’ll show you will quickly differ from your reality and you will have to start customizing the software and then the more you know and have access to all the source, the more options to go somewhere else if not happy.

If people still don’t understand the open source is the way to go, they are missing something fundamental.

Did you know that Google is the biggest contributor to open source? All it releases is open source, from chrome, to wave or android. They have understood that in this era, to be the number one the customers should be the ones developing what they want, so the platform should allow that.

To a some extend that is the secret success behind the iPhone and its 100.000 apps, a good API and a lot of people developing, but the police role Apple plays is not something developers like, and that is why the top developers of iPhone have already switched to android. I believe android will be bigger than iPhone in a very near future.

What I am talking about here are the fundamentals of how companies are working know, far from the model we were taught in the MBAs or in the business schools. Before they decided what the client should have and got the most out of it. This way of thinking and acting is over.

Now the key to survive and grow is to let the clients decide what they want, enable them to do it, and with the minimum benefit to survive. Again Google is avery good example of this. Also from the pharmaceutical industry who through collaboration can get the most out of a spread community of scientist that otherwise they could not afford.

If you don’t follow this pattern you will disappear.”

— dani, January 4th, 2010 14:25

“By the way, I see you are using Apache, PHP 5.2.12 and WordPress, all 3 open source… Good choice Janus.”

— dani, January 4th, 2010 14:25

“I think that 2010 will see the rise of more pragmatic commercial open source vendors such as ourselves (Umbraco). Taking the best from the traditional closed source world (someone to call, official support, training, certification programs, SLAs, warranty, etc) without cannibalizing the ecosystem with a consultancy/professional services department. The latter is often kills the motivation for a professional ecosystem and then you’re likely to be better off with a closed source vendor with a bigger ecosystem (partner network).

When evaluating open source vs. closed source, I believe that the open source needs to be on pair with what you expect from a closed source vendor, especially when it comes to finding an integrator (Solution Provider) and getting the level of support/warranty that meets the demands of your organization. In addition to this, a good open source project will often have a great dedicated community which is the best (= most transparent) way to monitor the health of the project/organization.

When it comes to CMS I think that one of the areas where open source sometimes falls short is that they’re quite generic in their nature, where you might find a more specialized (vertical) closed source CMS which might fit better to your needs when it comes to OOB functionality.

Niels Hartvig / Umbraco”

— Niels Hartvig, January 4th, 2010 14:25

“I agree that it’s more complicated than “open source” versus “closed source”. I’ve argued that “open source is the wrong question”. Not the wrong solution, but the wrong way of driving the evaluation process. For me, there are three key aspects to consider:

* Who will be the authors, and how usable does the CMS need to be?
* Who will be the site owners, and how technical are they?
* Will there be any customisation/development, and if so, done by who?

I’ve covered this in a slidecast, done last year:

http://www.slideshare.net/jamesr/opensource-web-cms-the-right-question-audio-presentation”

— James Robertson, January 4th, 2010 14:25

“> when does open source software not represent the best value?

When the out of the box features of a proprietary product provide your organisation’s short term requirements, assuming you cannot find an open source solution that meets those requirements as closely.

(I use the phrase short term requirements to mean the next 1-2 years; looking at criteria beyond that is generally too speculative to be useful. And by requirements, I take a holistic view: software features, support requirements, training, et al.)

In relation to your comment about commercial open source vendors not creating sizable communities: this is hard to do because the company culture must embrace that growing the community is a long term investment that takes planning, commitment, time, cost, energy, and may from time to time compete with short term corporate interests (like exceeding sales targets, etc.)”

— Sigurd Magnusson, January 4th, 2010 14:25

“Just wanted to provide an example to illustrate that open source can indeed be costly.

According to an article on iTWire, The Spanish Presidency of the European Union has just awarded a 12 million Euro contract to their web site built on OpenCMS…

That is a lot of money to maintain a website (regardless if it is based on open source or commercial software).”

— Peter Sejersen, January 5th, 2010 14:25

“Thank you for all the great comments. I very much appreciate it.

There’s an additional aspect, that I’ve forgotten so far: Many customers do like to be treated like a key account and get the attention of a good key account manager. When it works, this can make a big difference in terms of the project outcome and avoid many problems. You can’t get this with traditional open source, e.g. Mediawiki or OpenCMS, which don’t have a company behind them.

Dani: In my experience, many consider and buy open source specifically to save money.

Cheers, Janus”

— Janus Boye, January 5th, 2010 14:25

“@PeterSejersen: I think that a very “generous” buyer is more to blame than openCMS and open source in general.

Finding one extreme example seems a little unserious and populistic IMHO. Let’s raise the level on this blog, no”

— Niels Hartvig, January 5th, 2010 14:25

“It do agree to some extend that open source doesn’t always represent the best value. One example is around OpenCMS version which I used for one of the projects. It lags the basic content publish functionality, its inability to have staging and production environments. We ended up writing a custom code. It is very important to understand requirements and product’s capabilites before blindly selecting open source product just to save money.

Cheers,
Shishank”

— shishank, January 5th, 2010 14:25

“@James Robertson: Great presentation and an interesting angle to look at “open source” vs “closed source”!

One thing I disagree on though is the general statement that “low cost” is a benefit of an open source solutions.

The cost for implementing and maintaining an open source solution can be the same or even higher than that of a commercial solution, depending on the goals of the organization implementing it, available resources, time lines and a million other factors. (But of course the same factors can make it the better solution)

Of course cost ist not the only factor determining the value of a solution but certainly one of the most important ones.

Oliver Jaeger / e-Spirit”

— Oliver Jaeger, January 12th, 2010 14:25

“Over the years we have integrated a number of different closed source CMS’s into various companies such as Obtree, Immediacy and Microsoft CMS.

We made the decision 12 months ago to concentrate solely on Umbraco and for us it has been a great decision. We recently beat 3 of the top closed source .NET CMS’s in a 4 way pitch where we were brought in as a last choice and a kind of ‘lets just see what Umbraco is like’, not expecting it to be any good.

In our opinion, CMS’s all do the same thing – manage content, be that in slightly different ways. The main differences being cost of licence and quality of implementation by the partner or implementation specialist. You can pay upwards of £20k for a closed source licence, but if the implementer isn’t up to scratch, then it doesn’t matter how much you pay, it still wont be any good and the project may be a failure.

The thing with Umbraco is it can do anything its closed source competitors can do, but the money you’d spend on the licence fee you can spend on the implementation, meaning a better quality implementation all round as you don’t have to scrimp on development.

Anyway, thats my 10 pence worth…long live Umbraco and open source is what i say…and we’re very excited about 2010, we’ll be going along for the ride all way”

— Adam Shallcross, January 12th, 2010 14:25

“Not sure if the open vs closed source debate is that relevant for CMS these days.

Any platform will incur considerable cost of ownership in terms of development and maintenance over the long term, it’s just that some do not carry up-front licensing fees. Let’s face it – on a £250k development, the licensing fees for most middle-market commercial CMS platforms can be a relatively small proportion of the overall project cost.

Platform governance can be as much an issue for closed source as open source. Mambo users may have been left with egg on their faces a few years back, but RedDot users are looking pretty twitchy at the moment too.

As has been pointed out already, the role of the implementor is pretty key. if you have a good long-term relationship with a competent implementor then that’s half the battle won.”

— Ben Morris, January 23rd, 2010 14:25