Issues with collecting a data set via crowd-sourcing

Question

Consider a tight-knit community that might be willing to contribute to a communal data set via a website, Facebook group, etc. As a concrete example, let's say that the data set is a list of landmarks.

The goals of the project would be clearly stated, including that the data set will be explicitly licensed as open data.

What are the best-practices (or risks) with this approach? Should contributors be asked to acknowledge a "terms of service" before submitting?

Suppose further that another data set is a list of businesses with operating hours and a contact number. Does that change the complexion of the risks?

Generally speaking, how does one determine what is acceptable in these cases? I'm curious in general but especially vis-a-vis Canada and US law.

Answer 1

Crowd-sourcing businesses name/hours/number is a big part of what Wikivoyage does, and they have been doing that successfully for quite some time, so their case is worth studying.

• They state their license at the bottom of every page
• Any place where users can contribute data has a very visible notice saying that all text descriptions must be the user's original work or from a source with a compatible license. No "I acknowledge blabla" checkbox though.
• A data dump containing everything is easily downloadable. A previous version of Wikivoyage did not have that, and that lead to a rebellion and the community splitting from the administrators. Most people crowdsource for the public good, and can get very upset when the administrators try to keep the data just for themselves.

So, just decide what license to use (I would recommend public domain as it is hardly copyrightable data anyway), state it in any place where people will be contributing data, and make the data easily downloadable.