The limits to openness in research – As open as possible, as confidential as necessary

To open or not to open? The openness of information and data is increasing as tools and attitudes evolve. Publishing results and conclusions is a clear issue in the case of parliamentary committees and scientific articles, but communicating and being open about background data and work in progress is more complex. In this blog, Tuula Hämäläinen, the leader of VTT’s team responsible for knowledge discovery, discusses the limits of the openness of research output and data.

Openness of research is now being required by funders and even legislation. Open access, or open publishing, has become a familiar idea in the research world. However, open science involves many conceptions and assumptions that are either slightly distorted or even “fake news”.

Assumption 1: All information is free and freely available online – false

Although the number of open scientific articles is growing sharply, they are still of marginal importance and relatively small in number. For example, only just over 10% of scientific articles uploaded by VTT researchers are open. The others are based on subscription fees.

Other relevant information is also subject to a fee. Payment is required for either information content or its usability. The financial statements of private companies or standards are not available for free. Classified and easy-to-use patent or news information is not free of charge.

Copyright applies to almost all information, including free information on the Internet. The use of ordered information is also restricted by subscription and licence agreements.

Assumption 2: Open publishing is cumbersome and costly – both true and false

Research funders often compensate for publication fees, so that the research project or organisation does not have to pay them in full. Publication and the related costs should be planned in good time and be included in the project budget.

Publication in an openly available publication (“Gold Open Access”) is easy in administrative terms, but costs are incurred due to article publishing fees. “Green Open Access”, i.e. the self-archiving of articles in an organisation’s own systems, involves embargo periods and version control.

In most organisations, the library, information service or a similar function supports and assists in open publishing processes. This leaves the researcher to focus on the substance, while requesting administrative help.

Assumption 3: Corporate clients prevent all openness – true and false

Data associated with confidential projects cannot be published. Similarly, nothing can be disclosed about an innovation that is to be patented.

However, the fact that an organisation is engaged in confidential projects does not preclude disclosing of results and data associated with publicly funded, collaborative projects. You need to understand the type of data being processed on a case-by-case basis. You can then consider how and with whom it will be handled, and how to share it with others.

Assumption 4: The opening up of research data leads to abuses – true and false

It is increasingly required that data in the background of a publication or research project should also be open. But just uploading a set of files into the cloud and opening them up to the world benefits no one. Such data could also be abused.

Data and files must be uploaded in accordance with a research data management plan and systematic metadata must be created for them. When publishing your data, you can impose an embargo, only publishing it when it is no longer of use to you. The most rational approach may consist of opening up metadata or publishing certain datasets.

The concept of no longer disclosing anything to anyone no longer works. You need to decide what course of action is best for the competitiveness of your own organisation: what to disclose, to whom and when.

Data management becomes more important

Good data management and the control of personal data are even more important in a world where modern tools enable the easy sharing of data. Not everything should be tweeted or posted on the social media, and if it is uploaded into the cloud, then not just into ‘the cloud’ as such. Secure cloud solutions exist – we have long trusted in online banking services.

However, you cannot afford to retreat into a corner and never open up data to anyone because you fear the risks, but should think more carefully about what to open up, and to whom.

VTT is committed to complying with good scientific practices and ​promotes open science – see our open science principles

 

Tuula Hämäläinen
Tuula Hämäläinen
Team Leader, Knowledge discovery
tuula.hamalainen(a)vtt.fi

 

 

Find out more on VTT’s AI pages here:

https://www.vttresearch.com/services/artificial-intelligence-ai/

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: