The Multiple Meanings of the Term “Open”

Over the last couple of months I’ve found myself involved, both actively and passively, in several conversations that contained terms like “open” or “openness”. The adjective “open” was associated to nouns like “format”, “standard”, “source”, “government”, “data”, and so forth.

Quite often the use and misuse of the term “open” leads to almost hatred discussions, exuding religious fervor on both camps (those who are for “open” and those who are against). In my honest opinion, most of these battles are fought on the basis of a misunderstanding on either or both parties about what “open” actually means. Usually people confuse the end (such as making something more accessible) with the means (adopting one particular approach).

I would argue that many of those who fight the battle for “open” often lose sight of the reason why they do so. Nobody would deny that making something (a specification, a file format, a data set, a process) more open and transparent and accessible to a larger number of people is a good thing. This clearly implies that there should not be any obstacle for people to get that.

“Open” as in open source and open standard

Now, if I start with one the epical battles around “open”, which was Microsoft vs. the rest of the world, one could argue that what many were really  looking for were cheaper, more affordable office tools. Had Microsoft given Office for free or as a much cheaper value proposition, for instance making it pay by functionality used, I suspect that most proponents of open (source or standard) alternatives would have stepped back. Sure, the issue of not being locked in into a vendor is very important for governments, but would have they been equally vocal if they had got a (much) better deal from the incumbent? It is not by chance that many of the conversations I have with clients today are about cloud-based office suites rather than OpenOffice and the likes.

Isn’t it interesting that those who have benefited most from open source have been vendors? I see very few clients who would even consider to use an unsupported version of whichever open source software. They lock themselves into a vendor and external skills anyhow: sure, in theory they can ditch that vendor and get to a different one, assuming there is one and – if not – assuming they can get enough skills (internally or externally) to make that work. How likely or easy is that?

And what about all those Linux boxes that user organizations can deploy without paying extra licenses for the operating system? All very true, but those boxes are being hosted more and more by – guess who? – vendors, who sell access to those as a service (here is the cloud, again).

As I have been saying for quite some time, from a user perspective open source does make a real difference at the application level, where the open source licensing and development model together can change the way government organizations (but also other industries) can procure mission-critical applications, by relying on reuse and community development.

“Open” as in open government and open data

One might say that this is a completely different topic, but it isn’t. Today more and more governments put their faith to become more transparent and participatory in the provision of increasing quantities of open data sets. What “open” means here is somewhat subject to interpretation, as it ranges from XML and text files to XLS formats. All this openness is meant to benefit citizens, but it will not do so alone. There will always be the need for data aggregators, intermediaries, activists, businesses that turn that raw open data into something meaningful and useful to people. Be  it just a better way of visualizing data, or countless mashups and geolocations, the relationship between citizens and open data is the same that exists between IT users and open source: they also need intermediaries (i.e. product or service vendors) to make that work.

Of course there are exceptions. For instance, research organizations or statistical offices do use open source a lot, often manage their own IT infrastructure, and possess IT skills that are above average. Similarly, researchers or students or statisticians can benefit directly from open government data for their research papers, thesis work and reports.

But the average citizen is pretty much like the average IT user organization. He or she needs somebody in between who can package that data (or support that open source software).

Open for whom, open to what?

It would be great if future uses of the term “open” in whichever context gave a little contribution to clarifying the meaning of the term in that context. As I said, nobody denies that open is a good thing, but could we possibly know who is supposed to benefit most and how those benefits are going to be realized?

In my view, focusing on what is the value, cost and risk of choosing “high value data sets” in the context of the Open Government Directive is pretty much the same as articulating the value. cost and risk of choosing a given open standard or selecting an open source alternative.

Understanding both the value and the value chain is what will unlock the real power of “open”.