An explicit license for the infodump?

March 17, 2017 1:54 PM

My impression is that all of the infodump data is in the public domain. Is that license explicitly stated anywhere? And, if it is stated somewhere, could we get that statement reprinted on the infodump’s home page and possibly bundled with the zipped downloads. It’d be nice to have some legalese posted somewhere so that no one feels at risk of getting into trouble.
posted by Hooray For Socks! to Feature Requests at 1:54 PM (8 comments total)

That's a decent point; we've consistently treated the content as free to use but having a little bit of language making that intention clear on the Infodump page is a good idea.
posted by cortex (staff) at 1:57 PM on March 17, 2017 [3 favorites]

Also - didn’t there used to be data in the Infodump on projects? Or am I just making things up?
posted by Hooray For Socks! at 6:07 PM on March 17, 2017

Makin' things up, in that case; Infodump in its current form has as much data as it ever has.
posted by cortex (staff) at 8:53 PM on March 17, 2017 [1 favorite]

Wait...so comments and posts are copyright user, but this as metadata isn't? Is that the idea?
posted by Miko at 10:45 PM on March 17, 2017 [2 favorites]

I'd actually be moderately surprised if the Infodump would be protected under copyright at all.
posted by jacquilynne at 12:02 AM on March 18, 2017 [2 favorites]

Note that the outcome of Oracle v. Google thus far hasn't been that software APIs, which are basically metadata in one manner of re-use in question—the "structure, sequence, and organization" of code as a lower court found, couldn't be copyrighted or weren't protected by copyright, but that a fair use defense was valid in Google's case:

It's still the case that APIs can be protected by copyright under the law of at least one appeals court. However, the first high-profile attempt to control APIs with copyright law has now been stymied by a "fair use" defense.

I would hope that the decisions will survive further appeals even in the post-truth era, and that there won't be any legislative "clarifications" of copyright law in Oracle's favor, but with the case still being appealed I'd think that you might not have any substantial degree of certainty that a legal answer about the current status of infodump downloads would go unaffected by future events. (I'm not any sort of legal professional, though.)
posted by XMLicious at 1:38 AM on March 18, 2017 [2 favorites]

Wait...so comments and posts are copyright user, but this as metadata isn't? Is that the idea?

That there's a difference in the basic disposition of the two categories is the general idea, yeah. I'm not gonna tender a legal argument about it; practically speaking the metadata skeleton of people having-been-on-the-site, all of which is publicly scrapable/calculable, falls into a pretty different feeling bucket than the actual written content of comments and posts and in the nine or ten years the Infodump's been operating we've never gotten any substantial expression of concern about that metadata beyond a couple users availing themselves of the munging function.
posted by cortex (staff) at 8:35 AM on March 18, 2017 [1 favorite]

My sense is that the infodump is less like an API and more like a telephone book, thus recalling Feist v Rural Telephone which is why I suggested it would not be protected. It was late, and I didn't think about jurisdiction outside of the US, because of course it would be protected in the EU under the various laws that implement Directive 96/9/EC. In Canada, it's considered a bit of an open question how far database protection goes, but I'd tend to argue that there's not a lot of skill and judgement in the selection and arrangement of the Infodump itself (as opposed to the original site software designs) since they are in most cases quite literally just a dump of some of the fields in an existing table.

I don't think any of this is all that pressingly important to the question of whether it would be a good idea for Cortex to add an explicit release for use of the Infodump. People "licence" all sorts of things they don't actually have any legal backstop on their "ownership" of. Adding some words of reassurance that people can use the info is the least bad version of that. And because of the EU protections (and other places which might have EU-like protections), it actually makes sense.

Mostly I just think it's interesting to think about, because I am a Copyright nerd, though not yet a legal professional.
posted by jacquilynne at 9:40 AM on March 18, 2017 [3 favorites]