You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 5 Next »

This page summarizes some of the content from the internal TU training course Sharing my research output with others: The correct use of licences for data, software and publications *english* which I took in June 2022. By that I mean, this is all I remember from that course after looking through my notes. I can highly recommend taking it if you are interested in that topic.

The terms "license" and "copyright" come from the copyright law and are always related to some kind of "work" of one or more authors. The problem is that the law is usually behind real life, especially when it comes to technical advancements. Fortunately, software is already included in the copyright law, but general data not protected by it. Only if the data is part of some work will the copyright law also apply to it. That's why licensing data is a bit tricky and has to be studied on a case-by-case basis. Nevertheless here are few rules, tips and gotchas I took away from the course.

Table of Contents

General

  • Always add a license if you publish anything. Without a license the authors hold the strictest copyright and nobody can do anything with their work.

  • Once you choose a license you can later only change it to a more open one, but not a more closed one.

  • If your project contains a mix of software and other material, you can include multiple licenses, as long as you are explicit about which license applies to each part of the project.

  • Always check all your contracts (employer, funder, etc.) for terms or conditions on copyright and licensing. The TU Wien, e.g., holds the copyright on all works of its employees. However, since the TU Wien is not a natural person it cannot license anything. This has to be done by its employees.

  • TU Wien employees are not free to choose any license they want for their work. The TU has a say in it (Policy for Research Data Management). Strictly speaking this means that you have to ask the head of your organizational unit or the head of your institute when choosing a license for you work.
    However, in general, the TU Wien supports and encourages the use of open licenses: "In compliance with intellectual property rights, and unless third-party rights, legal requirements, Rectorate decisions, other reasonable interests or property laws prohibit it, research data should be assigned an open use license." Policy for Research Data Management, TU Wien, 2018, Page 3.

  • If students create data as part of a thesis or something, it is recommended to set up a contract that specifies what happens with the data when the student leaves. Otherwise it might be compilcated or impossible to use the data in future research or publications since the student is the data creator and holds all possible rights.
    TODO: Apparently such contracts are already availabe but I can't remember how to get them or whom to ask.

  • For general information about open access (like gold, green, and hybrid open access) and Plan S see https://www.tuwien.at/en/library/research-and-publishing/open-access-and-copyright.

Software

When it comes to open software licenses there are basically two types: permissive and copyleft licenses. Permissive licenses like MIT, Apache 2.0, ... allow you to do basically anything as long as you attribute the original author. Copyleft licenses, like (L)GPL, do not permit commercial use and require derivative works to also use the same license. That's why they are called copyleft or viral.

  • Take a look at https://choosealicense.com/ to choose a license.
  • Check out https://tldrlegal.com/ to lookup popular software licenses summarized at-a-glance.

  • If you use a library under a permissive license than the fact that the library is in your dependency tree is enough to fulfill the attribution requirement of such licenses. For example if you develop something in Python and have a requirements.txt where pip looks up your dependencies and installs them if necessary, then that's sufficient.

Not Software






Journal Specific Stuff

Here I try to find out which journals publish under which licenses, what you as the author of the manuscript can still do after publishing and other things about licensing and copyright that might be interesting. In particular I want to know if it is possible to publish the Git repository that contains the manuscript with the necessary data and scripts to produce all the plots as well as (thanks to CI/CD) an automatically compiled PDF that is always ready for download.



arXiv

  • Not really a journal but it took me way too long to find the license under which an article is published there, so I wrote it down here: It is on the top right under the links "PDF" and "Other formats". There is either a link called "license" or a picture of a creative commons license.

SciPost

  • All SciPost content is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License. That means everything is awesome and Git repositories containing the manuscript as well as compiled PDF articles can be made public as long as the CC BY 4.0 license is added. I would still recommend adding a link to the SciPost article and the DOI.

APS

  • After reading the APS Transfer of Copyright Agreement and the copyright FAQ I am as sure as a non-lawyer can get that making the Git repository with the accepted manuscript publicly visible is allowed. If the manuscript is also compiled into a PDF article and made available on the project's site one should add a link to the APS article/abstract/landing page or a full bibliographic citation and copyright notice. Unfortunately I am not sure which of the two options is really necessary in that case but the latter one should be the safer option.
  • For info about open access publishing in APS journals see https://journals.aps.org/authors/apcs. This also includes a price list for article publication charges (APS).
  • https://journals.aps.org/licenses might also be of interest

Elsevier

  • Supports gold and green open access. Basically: gold = CC BY or CC BY-NC-ND, green = CC BY-NC-ND; For gold you have to pay an article publishing charge (APC) which your institution or funder might pay on your behalf. A download link for the APC price list can be found on the website https://www.elsevier.com/about/policies/pricing. For example the APC for Computer Physics Communications is 2930 € (as of 03.06.2022)
  • Authors can share their accepted manuscript immediately via their research institute or institutional repository for internal, institutional uses or as part of an invitation-only research collaboration work-group. That means, in general, repositories on TUgitLab containing the manuscript and compiled PDF articles cannot be made publicly visible but only at most internally visible.
    After the embargo period, sharing the accepted manuscript via non-commercial hosting platforms such as institutional repositories is allowed. For most physics journals of Elsevier the embargo period is 24 months. See https://www.elsevier.com/open-access/journal-embargo-finder/journal-embargo-finder-results for details.
    If you pay and publish right under CC BY(-NC-ND) you can obviously make your Git repository public immediately.
  • Elsevier requires authors publishing their accepted manuscript to attach a non-commercial Creative Commons user license (CC BY-NC-ND). See https://www.elsevier.com/about/policies/sharing for details.



  • No labels