One of the significant new exceptions introduced this year enables Text and Data Mining (TDM) to be carried out on bodies of copyright material as long as it is for a “non commercial purpose” (and the sources should be acknowledged where possible.)

TDM includes a range of advanced techniques for analysing vast quantities of data in order to draw out new facts or statistical trends, or gather evidence of previously unexplored relationships (for example between chemical substances and medical conditions). The potential uses of TDM are very wide ranging and may occur in all disciplines.

The new exception makes the application of TDM to copyright works possible by removing the copying of material (which is an essential part of the TDM process) from the realms of infringing activity.

A Jisc report on the value and benefits of text mining from 2012 mentions that at that time TDM activities in higher education were mainly focussed on Open Access materials because the latter were more readily available. Since the new TDM exception was introduced in June 2014, the content of a vast number of e-journals to which universities such as UCL subscribe should also be available for TDM. This results from the fact that under the legislation, the terms of our contracts with the publishers of those journals cannot over-ride the TDM Exception:

“To the extent that a term of a contract purports to prevent or restrict the making of a copy which, by virtue of this section, would not infringe copyright, that term is unenforceable”, Copyright, Designs and Patents Act 1988, Section 29A(5).

This means that any terms in suppliers’ contracts which sought to restrict advanced computer analysis of their repertoire will no longer have any weight, at least where non commercial research is concerned.

Libraries are given greater flexibility by one of the less obvious changes made to the Copyright legislation earlier this year. This concerns requests for a copy of a periodical article from one library to another and is governed by Section 41 of the Act. Many of the updated exceptions include a clause stating that they cannot be over-ruled by the terms of a contract, which tends to reverse the situation prior to the 2014 changes. Section 41 of the Act  includes such a clause:

“41(5) To the extent that a term of a contract purports to prevent or restrict the doing of any act which, by virtue of this section, would not infringe copyright, that term is unenforceable.”

It follows from this that a library may now supply a single copy of a periodical article from a given issue of an “e-periodical”, just as they may supply a copy from the print publication. The terms of the contract under which the e-periodical is supplied may formerly have prevented this, but it seems they can no longer prevail over the exception.

Naturally the libraries concerned need to be sure that they are complying with the legislation, for example the library fulfilling the request must have legitimate access to the e-periodical and must not supply more than one article from a single issue. It is important to note that this “library to library” exception does not permit the supply of copies from books unless it is practically impossible to contact the copyright owner in order to seek permission.

Now that both the IPO’s licensing scheme and the EU Orphan Works Directive are available, it will be interesting to see how user-friendly they really are. If you have any UCL projects in mind where we could test the licensing procedure with a limited number of orphan works, please do let me know.

We will be eager to see examples of successful use of the new licences but currently there are no applications to view on the IPO Register of Orphan Works. There is a lot of background information on the IPO website. The Orphan Works Licensing Scheme Overview for Applicants includes details of charges and full Terms and Conditions of the Licence. A big unknown is what the IPO will accept as a “diligent search” for the copyright owner in specific cases, although there is plenty of information on sources for a diligent search on the IPO website.

The admin charges start at £20 for one item but rises only to £80 for 30 items, that being the maximum which can be included in a single application. For non-commercial uses the licence fee itself is 10p per item. For this purpose the IPO uses a definition of non commercial which excludes charging to recover costs (see paragraphs 35 to 37 of the “Overview” linked above).

Licences last for seven years with the potential for renewal. Licences cannot be longer term because of the possibility of a copyright owner emerging after the grant of licence. In that eventuality the licence would run its term but would not be renewable. The IPO is in effect indemnifying us, the licensee, against action by the copyright owner, as long as we have given them accurate information.

Did you know that a broad swathe of historically important unpublished works is in copyright until 2039? A campaign has been launched by CILIP, the Libraries and Archives Copyright Alliance (LACA) and others to persuade the UK Government to press ahead with a small but significant change to the copyright term for unpublished works.

Once the copyright term for these work is reduced to the standard (and more reasonable) author’s lifetime plus 70 years, then museums, libraries and archives will have greater freedom to display and reproduce the unpublished works they hold. The more historical items will come out of copyright immediately. (See the previous blog post of 21 August 2014 for more background).

UCL Library supports the Free our History Campaign and would encourage colleagues in the cultural and educational sector to add their voices. There is a petition to sign on the CILIP web pages Libraries and museums can also participate by  displaying a “blank page” in place of a significant unpublished work they would like to exhibit but are prevented from doing so by the current copyright term for these works.