The Three Meanings of Open
The OKF is the Open Knowledge Foundation. But what does it mean for knowledge to be open? We take open to have three distinct senses: legal, social and technological.
Knowledge is legally open if it is free of most of the standard legal restrictions and requirements. In particular it should be accessible without restriction, reproducible freely (at least for non-commercial purposes), and reusable – that is, freely incorporatable in derivative works. In short, it should fall within the bounds of one of the Creative Commons licenses.
Social openness consists of ensuring that a work is made available and not kept secret or mouldering on a CD at the back of the drawer. It means supporting sharing and reuse as well as collaborative working processes.
But most importantly it means an ‘open source’ approach to knowledge. That is, knowledge should be made available so that access is given to the raw, underlying data and not simply through a particular, usually limiting, interface (such as a human-only-usable web form).
This parallels the distinction with software programs, emphasized by the term open source, between access to the underlying source code and access simply to the compile version. Thus Open Knowledge in this sense can stand for access to the underlying ‘source’ rather than purely access to the ‘compiled’ end product. To illustrate consider the following examples.
For data in a database the ‘source’ form means the raw data and the ‘compiled’ form is any of the multitude of interfaces such as web query pages that can wrap that data. Providing access to the source data would be a major change – even open databases that are freely searchable rarely provide their data in source form – the only form in which it is any use to a computer.
Another example is provided by the common practice of providing a PDF version of a document rather than the original text file. This, perhaps intentionally, hinders access to the underlying text and inhibits activities such as annotation or indexing.
Technological openness requires that knowledge is provided in a form and format that does not unnecessarily hinder access to humans or machines. This can be achieved by utilizing data formats and tools that are open – meaning that a full specification is publicly available and unencumbered by legal restraints, and that access and use of the formats will not require proprietary tools or products (for more information on ‘openness’ of formats see the Information Accessibility Initiative).
It also means providing the necessary documentation, structuring and presentation of data so as to ensure comprehensibility and usability. One should aim to achieve these ends not just for humans but also for computers – something that is increasingly essential in an information age.