Library Out Loud

I was recently asked an interesting question – “I’ve written a report that hasn’t been published as either an article or a conference paper but I’m keen to be able to circulate it with a DOI (Digital Object Identifier) so that it has a permanent place on the web and is easily discovered and cited. It’s not really suitable for a journal but I think it could make an impact on people working in my field. What are my options?”

 
This is classic “grey literature” – work that could be useful in getting your research message out there but that might not appeal to good quality journals with their emphasis on new work and broad appeal. There are various types of research-related work that fall within this category – working papers, reports, conference papers that did not appear in published proceedings and even PowerPoints – that could enhance your reputation and help make your scholarship available to readers who might not be reached by conventional academic publishing. With the growing emphasis on the social, economic and environmental impacts of research this makes sense, but venturing outside the conventional outlets gets you into some tricky territory known as self-publishing. That’s not necessarily the worst thing in the world but it needs to be handled with caution if you want your work to be taken seriously. And you need to make sure that your work is easy for others to discover and read. And that it appears to best advantage.

 

There are a number of options to consider, but first a word about identifiers.

 
An identifier is a unique permanent alphanumeric code that identifies a unique object. It differs from a name or address in that both of those can change and they may not be unique. Think of it like your Massey ID number; no matter if you change your name or its preferred form, shift departments or change status, your ID number is still going to be the same. Even when you leave it won’t be allocated to someone else – the university, in fact the universe, isn’t going to run out of numbers. And it always has to be your number, it can’t be transferred and you can’t change into someone else. And you can’t have another one – there is only one you. Your tax number is another example of an identifier, and there are a number of important identifiers in academia – your Scopus ID or Web of Science Researcher-id for example, your ORCID ID or the DOIs attached to your articles or the Handles attached to the online versions of your thesis or other works in Massey Research Online.

 
The DOI has become a standard element in many referencing systems and at first glance it might appear to be just a tag attached to an article or a chapter. However, it’s important to take a closer look at the name Digital Object Identifier because what it identifies is not the article/chapter and its particular set of words and/or images but a specific and unique digital expression of them – an object rather like a chair that exists in one particular place at any given time, although it could move to another place and still be the same chair. It differs from a web address in two ways –

 
1. Websites frequently change their structure and the web addresses (URIs) of specific pages change with them. It was partly in response to problems caused by this that the DOI was developed. It’s just a character string but behind it sits a system (Crossref) that always “knows” where that number is to be found on the web. To locate a specific digital object on the web you put the string https://doi.org/ in front of the DOI in the format https://doi.org/10.xxxx/xxxxx

 
2. The content of a web page can be changed at any time but an object with a DOI is supposed to be perpetual and unchanging – its most common use is by publishers to identify the version-of-record, the official published article or chapter as it appeared in the journal or book. To some extent this harks back to the days of print when whatever was printed was it, for all time, and could not change. Crossref have loosened the rules recently to allow preprints (i.e. author manuscripts before peer review) to have their own DOIs as expressions separate from the version-of-record, but author postprints (i.e. fully peer-reviewed but separate from the version-of-record) are not supposed to have a DOI as this rightly belongs to the version of record.

 
So, to get back to our question about grey literature, what role does an identifier like a DOI have in making your grey literature less grey, more permanent and even something that could be cited and taken seriously? Well, again there are two reasons. Firstly your reader has some sort of guarantee that the content is not going to disappear and, secondly, that it’s not going to change. It will always be findable and always the same thing – identical in fact. That’s a reason to be really careful when giving your work a DOI by the way, you can’t take it back or make it go away.

 
And how do you get hold of one of these magic little character strings? There are lots of options, but here are a few good ones –

 
Massey Research Online – our very own MRO is primarily a home for Massey theses but it also functions as a “green” repository for published articles and is host to a wide range of reports,  working papers and other grey items. Here’s an example.  Unfortunately MRO items don’t get a DOI – the persistent identifier is a “handle” which takes the form of http://hdl.handle.net/10179/3954 which is a less well known identifier and won’t automatically become part of any citation of the document. A big advantage of MRO is that it seems to be well covered by Google Scholar which means that your work will be visible to other scholars.

 
Figshare – this is another really good option. Figshare is primarily a site for data sharing and it is the used by Nature Publishing and Public Library of Science to provide access to datasets accompanying articles. In other words it has plenty of prestige and a high level of stability. Any file type can be uploaded to Figshare and will automatically receive a DOI so it’s an excellent place to self-publish. Here’s a good example of a researcher who had a good experience with them. You should also look closely at Figshare if you are thinking of making your research data public. One downside is that Figshare content is not routinely covered by Google Scholar so your content probably won’t be found through the most widely used scholarly search engine. (Figshare tell me they are working on this but Google Scholar are notoriously hard to get through to so don’t watch this space too closely.) And you need to be very careful about getting it right, because publicly shared Figshare content can’t be altered, deleted or hidden.

 
ResearchGate – there are plenty of good reasons to have a ResearchGate account  although you do have to take care not to breach copyright in uploading published content. For your unpublished work, however, it looks like a good self-publishing option as they offer a do-it-yourself DOI service. You can’t do this for anything designated as an article, book, chapter, patent, poster or conference paper as these are considered to have been published already which would break the DOI protocol, but for reports and working papers it could be a reasonable option. Findability might be a problem as coverage by Google Scholar is sporadic so make sure you have a strong and unique title.

 

 

The other important thing about all of these options is that they are all FREE. That’s right, no money.

 

 

On a personal note, I’ve moved on from the Library to the position of Copyright and Open Access Advisor but the Library have very kindly allowed me to continue posting to Library Out Loud so I plan to appear with words of advice from time to time.

 

Bruce White

26 April 2017

Postscript 27 April

Some sage advice for Massey researchers from Doug Franz – “A suggestion could be to include a DOI from either Figshare or ResearchGate in the Elements information which then gets the DOI onto our profile page which is then spidered by Google Scholar. Alternatively to deposit the newly minted DOI into MRO through Elements (via the Enter an OA location link).”