Not sure license can be MIT

[diegoceccarelli]

The License in pytrec_eval is MIT but the repo includes https://github.com/usnistgov/trec_eval/tree/de6a29f8ba9312c73f978aa9739695aa8ebf48eb that is under the copyright of Chris Buckley
see also https://github.com/usnistgov/trec_eval/blob/master/convert_zscores.c#L2-L5
and usnistgov/trec_eval#10

[dmlls]

This could also have repercussions for BEIR, which includes pytrec_eval as a core dependency, and is licensed under Apache 2.0.

[seanmacavaney]

It looks like the license was added in b8358d4. cc: @cvangysel

[cvangysel]

Thanks all. My apologies for the delayed response. We're in the process of transferring ownership of this repository to the Amsterdam IRLab and so hopefully the future maintainers will have a better response turnaround time.

I'm not a lawyer, but my understanding is that trec_eval is a transitive dependency of pytrec_eval. The MIT license applies only to the source code present in this repository, and not the trec_eval source code itself (since I have no control over that). The fact that trec_eval uses a different license from pytrec_eval is mentioned in the README: https://github.com/cvangysel/pytrec_eval#license.

We are not redistributing the trec_eval code (either source or compiled) as part of the published Python package, since that triggers this code path and pulls in the trec_eval source code at installation time. However, that may not be the case for the binary distributions published in the Terrier fork which is maintained by a 3rd party.

The license linked by @diegoceccarelli allows for non-commercial, research usage/modification and applies to trec_eval as well. Hence, if you are allowed to use trec_eval as a tool, you are also allowed to use pytrec_eval.