[euopendata] Study says charge for public data...

[James McKinney]

> Marco Ricolfi explained this better than I can do:
> http://www.epsiplatform.eu/guest_blogs/re_use_licenses_commercial_or_non_commercial_this_is_the_question

Marco Ricolfi's second point ignores the possibility that the person
licensing their work under a non-commercial license could charge for
commercial use:

"Second, we should consider that, once digitization costs are
incurred, it does not make any difference whether the re-user makes a
profit from re-use. No marginal cost is incurred because there is an
additional re-user. If he or she is smart enough to create a business
model which enables her or him to combine this input with other inputs
and make money out this, nice for him. Nothing is taken away from the
public."

What Marco says is true, but he forgets that, in his case, the
Sicilian local administration could charge for commercial use but
allow free non-commercial use - which is a pretty serious omission.

> I agree with you to a certain degree. But I do have two counter arguments:
>
> 1. The fact of the matter is that most of the untapped value in
> opening data lies in unforseeable innovation, i.e. innovation that
> will result from a long line of trial and (mostly) errors, until
> somebody comes up with something that turns out to be valuable and is
> able to extract value from that. This innovation can come from the
> academia, from small startups or large corporations, but the fact of
> the matter is that none of these will be playing around with the data
> if it is not freely available for such tinkering. Even big companies
> only spend money on things they believe are highly likely to be
> successful. A minimal fee can therefore prevent enormous creation of
> value.

Assuming that, in our hypothetical jurisdiction, non-commercial use is
free but commercial use is not, you can play and tinker with the data
up to the point where you start making commercial use of it.
Entrepreneurs, academics, and companies can do the trial-and-error you
describe without paying for the data. If it turns out to be valuable,
it can release it commercially and cover the costs of the data,
assuming the price of the data is reasonable. (I find it interesting
that you first say that big companies are willing to invest in
trial-and-error, and then later say that big companies only invest in
things that they believe are highly likely to be successful.) So, I
don't find this argument convincing.

> 2. Successful commercial enterprises are beneficial to the economy as
> a whole (and thereby the government) as they pay salaries, taxes, and
> buy goods and services from other companies that do the same.

I don't disagree with this. But successful commercial enterprises have
the funds to pay for the data (again, assuming the price of the data
is reasonable, which it should be.) I don't think paying for the data
would put these companies out of business.


I think there are good arguments for avoiding non-commercial licenses,
but I have yet to read them on this list or in the linked articles.

The only good arguments I've read have less to do with commerce and
more to do with re-use; for example, the argument that releasing your
work under a non-commercial license means works like Wikipedia cannot
include it, as its license conflicts. I don't think this is a serious
threat, as Wikipedia would still be able to do all the things it can
do with works under "all rights reserved" copyright, like commentary,
sampling, documentation, and other uses. So, if you choose a
non-commercial license, I don't think you are often prohibiting
activities which you would like to allow.

Another decent argument is over the ambiguity of "non-commercial." But
it is, AFAIK, a legally untested ambiguity, so I would not assume, for
example, that non-profits putting ads on their web pages that reuse
NC-licensed works would be violating the license.