Just the Pure truth...

On Friday we had another demonstration from Elsevier of their research management system, Pure.  We had a demonstration/sales pitch a few years ago, but at the time things were not as desperate as they are now.  The research FTE has grown by over 100% in the last eight years, with journal articles alone growing by a whopping 230% in the last five years (as submitted for ERA evaluation).  So, in spite of the Library having chosen a preferred new institutional repository, the Research Office decided that they knew better and organised a new demonstration of Pure.

It say that it was interesting would be an understatement.  It was an informative demonstration that showed that there are so many holes in Pure that I’m surprised that it is keeping afloat!

But, let us look at the reasons why our institution is interested.  The Research Office uses Research Master currently for Grants, Ethics and HDR management.  They used to use it for Publications but have recently turned this module off in favour of ‘linking’ their reporting to our very flaky and unstable Access database which shadows our institutional repository (a whole other story).  The HDR management is being migrated as we speak to PeopleSoft, which just leaves the Grants and Ethics.  In an ideal world the institution would source a new system that would be able to cope with both the Research Office’s needs as well as the Libraries (which include research outputs, data management and digital collections), but it seems that just such a system has yet to be made.  The Library is keen on signing up for ExLibris Esploro, which initially would only cater for publications however research data management and possibly grants management are on their roadmap.  The Research Office is “luke warm” about Esploro (I know not why),  so when the invitation came for another demonstration of Pure, I was ready and waiting to see what had changed in the last few years.
I have to say, not much.  Below are some random thoughts from the demonstration (which was delivered as a webinar).

• Many institutions, including University of St Andrews (UK) which was featured in the Elsevier PPT, use Pure but also still maintain a separate institutional repository.  When I asked the Elsevier presenter why this would be, he responded that there may be a number of reasons including the need for a system that could handle collections (Pure doesn’t handle collections at all).  As a Library we have digital archival and research collections (for example, the K'gari (Fraser Island) Research Archive and the USC Art Gallery Ephemera Collection) that are currently using the ExLibris Alma/Primo VE platforms.  Although there have been and still are some teething problems to using this, on the whole it is a big improvement on our previous system, Canto Cumulus.  So it would be likely that we would continue using Alma/Primo for these types of collections.  However there are other research collections housed in our institutional repository that would not be able to be catered for in Pure, such as publications relating to a particular project a particular research group (especially if it were an informal group), theses, conferences that USC hosts, etc.  

• Reporting and analytics are always a problem with any system, and often the deal-breaker if it doesn't deliver.  During the recent CAUL Research Repository Days in Melbourne it was mentioned that Pure’s reporting was “diabolical” and that some coding experience was required.  Pure doesn't share direct access to the database, but provides all reporting through APIs.  The API structure is the same as the Scopus API.  Reporting of the backend is via APIs whereas the reporting from the front end is via the dashboard.  Elsevier is currently enhancing their reporting module including building a ‘write’ API.  Interestingly my advance question on output analytics (page views, downloads) was ignored and I didn’t realise until afterwards that it wasn’t answered.

• In our previous demonstration some years ago, Pure had no Ethics module.  Now they have a basic Ethics record, with some institutions such as Monash University using Infonetica as their main Ethics system.  I do not know enough about Ethics workflows to know what we would need, however it seems that Pure will not cater for the Ethics requirements of the institution.  

• When I asked at the end about compliance with the ARC/NHMRC OA mandate, I expected a system such as Pure to be up to code.  However I was surprised to hear that they are not compliant.  Elsevier is meeting with the ARC in the near future to discuss the requirements, so who knows when it will be released into production.  That being said, our current institutional repository is only about half compliant.  But shhhhh!, don’t tell anyone.  

• Preservation is becoming increasingly important for digital data, and research information is no difference.  It is something that I am very passionate about, although acknowledging that I am a novice in the field.  When asked about Pure’s preservation strategy the Elsevier presenter mentioned that a history is kept of metadata records in the background (no clue as to if this is accessible to the administrator or only Elsevier staff).  No versioning of documents is kept as the system is not designed for this.  However I believe that Pure can plug into third party proprietary preservation systems, although this wasn’t confirmed by the presenter.

• During the demonstration several years ago it was mentioned that any metadata harvested into Pure from Scopus was unable to be edited.  Scopus is not the most perfect of metadata aggregators and there are often mistakes with the metadata.  So to hear that the system would have a subset of records that are unable to be edited was alarming.  Happy to say that this has now changed.  Pure now lets you edit Scopus records, and is even considering allowing users to edit records for errors which would then feed back to the Scopus database!

• Editing records is a bug-bear in our current institutional repository with many of our older records unable to be bulk edited.  The Elsevier presenter said that all the records could be edited in bulk.  The documentation however states that not all fields could be edited, and this was one of my advanced questions, but like the analytics question I forgot to ask about this during the demonstration.

• One of the advantages of Esploro is that research data management, in particular dynamic data management plans, are on the roadmap to be developed.  Pure is not going to have research data management capabilities but is going to rely on Mendeley Data for the data management tool.  I haven’t heard anything about Mendeley Data, so will need to look into it to see what it’s capabilities are. 

• The Grants management in Pure seems to be a fairly superficial module, at least compared to the richness of the Research Master data.  However, like Ethics, I do not know enough about Grants to know if this would be a suitable system.  I do know that Esploro will interoperate with Pure, so if the Research Office chooses to use Pure for Grants and we end up using Esploro for our institutional repository, they will at least talk to each other.

On the whole, Pure is a system that could be used as long as you make allowances for it’s limitations - in reporting, institutional repository, collections, Ethics, Grants and research data.  No system is perfect, however some systems are more perfect than others.  And I fear that Pure isn’t one of them.