Is "open" data free as in speech, free as in beer, both, or something else?

Question

In open data, should I understand open to mean:

• Free as in speech
• Free as in beer, gratis
• Both?
• Something else?

For example, the Swedish Meteorological and Hydrological Institute (SMHI) has dedicated itself to open data, and more and more of its products will be available with free licenses; first for non-commercial uses, but later without usage instructions.

For some data, such as data archived on tape, they will still ask for a one-time distribution fee, I think to cover costs they need to make to retrieve the data. The data are still freely licensed, so the user is free to put it on the web etc.

Is this still "open data"? The data in question is licensed under CC BY 2.5 (Swedish link).

Answer 1

The Wikipedia page for Open Data states it better than I can. That is, open data should be:

... freely available to everyone to use and republish as they wish, without restrictions from copyright, patents or other mechanisms of control.

To answer your question, truly "open" data should certainly be free as in speech. I would say, however, that open data need not necessarily be free as in beer. In your example, even though a fee might need to be paid to access the data, as long as the end-user is able to use the data as they wish (including for commercial purposes) then I would agree this still could be considered "open data". In practise, data that is made available for public use (by government, say) is generally also free as in beer, but I don't see this as a requisite.

The idea of open data is, in my view, synonymous with the notion of open source software. Just as open source software can be sold if desired by any party, I have no problem with open data being used in commercial ways. If a business is able to add value to open data (or software) then they should be able to charge for their product. The original source will always be available to anyone wishing to use it.

Of course, the above view is utopian in some ways. I suspect a lot of "open" data these days is in reality controlled in some way via partially restrictive licensing, copyright, and so on.

Answer 2

I would say that open data is free as in speech - not as in beer. @Snubian's quote from the wikipedia page is a good starting point. But free actually means subsidised - because it is never free to collate, check, record & store meaningful data over any length of time. So there has to be a motive to maintain the quality of open data if it isn't to degrade into worthlessness.

And the quality of open data is really the key point. Like going and listening to someone rant on Hyde Park Corner you can get lots of free speech but verifying its quality is essential to building something worthwhile based on it.

I think that the situation in the late 19th century for the western world where lots of libraries were built for the common good is a good analogy for open data. Back then lots of well meaning (and rich) people built libraries to help improve society and implicitly improve the culture & world that they lived in. Now rich organizations can release structured data often to encourage an ecosystem to be build around their products.

In both instances something other than a profit motive can energize both sides of the coin, but replacing Victorian philanthropy with corporate brand building as a motivator.

So open data really has to be free and is better without a profit motive - but for the quality to be maintained you need a different motivator, or the data will not be maintained. If there is no way to justify subsidising the maintenance of open data as in the SMHI example then you have to question its medium term viability in any case.