For Approval: Academic Citing License

Johannes Kaiser jkaiser at
Mon Sep 27 21:52:12 UTC 2004

> On Mon, 2004-09-27 at 06:08 +0200, Johannes Kaiser wrote:
>> On 26 Sep 2004, at 23:17, Evan Prodromou wrote:
>>> This is where you're mistaken: this is _nothing_like_ the requirement
>>> to have license notification. That's a restriction on the 
>>> publication of
>>> the work itself. You're trying to restrict the publication of the
>>> output of the program -- a very clear restriction on use.
>> OK, I see that the condition is a restriction on use *in general*.
> Yes. The problem is that, as the copyright holder, you have limited
> rights to tell people how they can use your software. You can keep them
> from distributing it or making modified versions and distributing 
> those,
> but you can't really tell anyone how to use it day to day.

You tell me and I've got to believe you. I understand that this holds 
for open source licenses. But how about the creative commons licenses? 
They are telling people how to use it, e.g. not commercially. So your 
statement cannot be true in general.

> At least, this is how I understand the "no restrictions on use" idea.
> Why isn't it part of the OSD, you may ask? I'm not sure. My guess is
> that it was assumed that since copyright holders don't have the right 
> to
> tell people how to use their software, that shouldn't show up in
> licenses anyways.

Yes, to get a better understanding how this thing works, 
I would be very interested in what the criterion for OSI approval is. 
The web page says it is the OSD, but you are telling me it is (also) 
the somewhat vague "no restrictions on use" idea. Is this consensus?

>>> I would suggest that you license your work under one of the already-
>>> approved Open Source licenses, and add a _reminder_ in the
>>> documentation for your work that scientists who use your code to
>>> make new discoveries
>>> are obliged through both professional courtesy and thoroughness of
>>> reference to cite your original article.
>> That is exactly what I think science should work like! Unfortunately,
>> it does not far too often. At least, I know a couple of pretty bad
>> stories.
> This is unfortunate. However, I don't think your license will solve the
> problem. If someone publishes a paper that uses your software but
> doesn't cite you, what will you do? Sue them? My guess is that your
> restriction is unenforceable, and you would lose in court.

I agree that the license would not solve the problem entirely in 
practice. Instead, I would think that it increases the awareness of the 
user, thereby reducing the problem. And I am sure that my more 
experience coworkers would never ever give any code out of their hands 
before the recipient has signed a license agreement containing such a 
citation condition (and much stronger one requiring prior notification 

> If you're not going to sue, or if you can't, then it's probably not a
> good idea to include the restriction in your license.

Thanks for your explanations!

More information about the License-discuss mailing list