I am currently developing a fitness utility application and have reached the point where I need to populate it with a real data-set.

After quite a lot of research I found that there were literally no companies that allow their data-sets to be used openly. Now, although this is understandable I don't exactly understand how they can copyright their data when it comes to things like a fitness exercise.

For example, if we take a record from bodybuilding.com like:

    name: "Barbell bench press",
    type: "strength",
    main_muscle: "chest",
    other_muscles: ["shoulders", "triceps"],
    equipment: "Barbell",
    mechanics_type: "Compound",
    level: "Beginner"
    sport: "No"
    force: "Push",
    reputation: 9.2,
    description: "Lie back on a flat bench. Using a medium width grip (a grip that creates
                  a 90-degree angle in the middle of the movement between the forearms
                  and the upper arms), lift the bar from the rack and hold it straight over
                  you with your arms locked.
                  When you are done, place the bar back in the rack."


How can this data be copyrighted if just recreating it would yield the same result, given that the values are not ambiguous/subjective?

The only thing I feel could be copyrighted is the sequence of words (phrasing) in the description – please correct me if I am wrong.

If this is the case: Would it be legal for someone to use the exact same data-set, rephrasing the words in the description? and, for example, replacing beginner with amateur?

Edit 1:

Except from the description and the reputation, all the other fields, and the data structure in itself are completely standardized. The exercise names are not thought of as so by the author of the data-set, they are standardized; the same is true for equipment names, force, level, etc. (again, excluding description and reputation field). So if one were to remove the description and reputation field, could it still be thought of a copyright infringement. Being that everything else is not subjective or does not require creativity to add?

  • How can this data be copyrighted if just recreating it Sorry, but recreating if from where? There is a publicly available source that already categorizes the information about that contraption with the same categories? – SJuan76 Oct 22 '16 at 21:06
  • Well, yes, what i meant by the phrase you quoted was that if i were to sit down one day and start typing out my own exercise dataset, in the end i would get a result that would be identical to that found on any fitness website (aside from the description field). – WiserTheBassist Oct 22 '16 at 22:46
  • After seeing your other comments, one friendly reminder: This is not legal counsel. If you are planning about doing something, talk to a lawyer. If you do some illegal or get sued, saying "but the guys at law.stachexchange.com told me I was right!" will not help you at all. And for all I know (again, check with a lawyer), everything you propose is against copyright laws (but you should check with a lawyer about that). Check with a lawyer. – SJuan76 Oct 23 '16 at 1:22
  • Thanks again @SJuan76. Thank you for pointing that out. Just for the record what i am actually planning on doing is compiling my own database. I asked this question because i realized that even by creating it myself i would still get something that has the same data structure and some of the field values equal to any other data-set around. So this was kind of a hypothetical 'worst case scenario'. Thanks again – WiserTheBassist Oct 23 '16 at 9:30

In the United States, you can't copyright something like a phone book that is just a collection of records sorted in an obvious manner.

According to the US copyright office, you can't copyright a recipe either:

Copyright law does not protect recipes that are mere listings of ingredients. Nor does it protect other mere listings of ingredients such as those found in formulas, compounds, or prescriptions. Copyright protection may, however, extend to substantial literary expression—a description, explanation, or illustration, for example—that accompanies a recipe or formula or to a combination of recipes, as in a cookbook.

This seems like it's something along those lines. On the other hand, it might not be so clear-cut, as much of this "data" is really opinions (like the level and reputation.)

As you say, the "description" field is problematic. But just changing a few words will probably not cut it; that would be a derivative work. You'd probably want to rewrite it from scratch.

| improve this answer | |

"Data" is a slippery concept, especially if you're trying to distinguish it from a "program". A simple fact such as the length of a stick or a phone number is not protected by copyright. What you've provided is a structured record, which is closer to being a program, and these days, having the right data structure is most of the work behind making a program. Also, the values are not automatically assigned by an automaton: some element of creativity goes into creating the data structure and the specific records. This is the "modicum of creativity" that is required for copyright protection.

| improve this answer | |
  • Hi @user6726 thanks for your answer. From what i take away from it the key question i should ask myself is if i am copying anything that the original author would not have been able to create had he not added some of his personal creativity. If this is the case i would suspect that i was originally correct in my question because all the other fields and the data structure itself are pretty much standardized. I have edited the question to add this detail. – WiserTheBassist Oct 22 '16 at 23:06

The data is probably subject to copyright – it is a literary work stored in tangible form – it is not just a collection of facts.

On the other hand, your proposed use is probably fair use: If you are using it only in house to test your app then it cannot affect their market and this is one of the criteria that you need to show for a fair use defence. Also, how would they ever know?

However, there is an easier solution that avoids the issue altogether: Write a program to create a random but legitimate data set and test on that.

| improve this answer | |
  • Yes as a matter of fact that is exactly what i have done, i have been testing the application with random data. However, the closer i get to my release date the more concrete this problem becomes. So i will eventually end up hiring some freelancers to create a legitimate data-set. – WiserTheBassist Oct 23 '16 at 21:25
  • If you hire contractors to write copyrightable materials, make sure you understand the limitations of the "work for hire" doctrine in your country, assuming you want the freedom to duplicate, modify and distribute their works as your own, i.e., without having to seek further licenses from them. – Upnorth Aug 20 '17 at 4:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.