Research Portal News

Eve Gray kindly shared this via Twitter: VERY interesting presentation by Jason Priem

 

Peter Dunsby of Maths & Applied Maths here at UCT has shared his experiences of coping with only an iPad while on the international visit road - and it makes very interesting reading, simply from an "I didn't know you could do that!" perspective.

"Have a look at a few posts I wrote while on the road with the iPad last year. Plan to update them soon. 

• Post 1
• Post 2
• Post 3
• Post 4"

Now - if only ICTS could stop the constant demands for authentication, that make it painful to use an iPad on the UCT wireless network...!

iTunes U

The announcement by Apple Inc recently about the launch of the iTunes U application for the iPad and iPhone, and related enhancements in online courseware development for iTunes U, has created rather a buzz in academic circles around the world.  You see, heavyweight institutions like Harvard and Oxford Universities, MIT, and the mighty British Open University, seem to be making a lot of courseware available there - for free.

There has accordingly been some interest here at UCT in getting official approval for courseware submissions.  It's difficult to know what's not to like about this: visibility from all over the world, promotion of the UCT brand, uptake of UCT-developed materials by other institutions, and new things to do for the Centre for Educational Technology, among others.

Oh, there is the "Apple factor" to possibly worry about, meaning the locking-in of participants to one technology and one interface - but iTunes works on both PCs and Macs, which should take care of that objection, and I remind you, most offerings are FREE.

While this may not immediately appear to have relevance to the eResearch Portal (as we are coming to know it), it certainly does for the instruction of research students - which is after all, a core activity of researchers at this institution.

iBooks Author

Another announcement from Steve's Company was of less general interest, but loomed large in the interest bubble of committed iPad enthusiasts.  Like me, it seems almost redundant to say. 

This was that Apple was making available - for free - the iBooks Author app, for constructing iBooks for use on the iPad.  There has been a lot of grumbling about the EULA for this as well, given that it is necessary to use ONLY Apple's online store to sell anything developed for sale - which completely ignores the fact that it is possible to circulate iBooks for free via email or other electronic means, to any iPad user you like.

I have been having a great deal of fun with this: it is also an EXCELLENT means of painlessly making PDFs for instructional or informational purposes, which can of course be opened on any platform.

We WILL assimilate you....

 

 

 

A new tool in SciVal's Spotlight (ask the Research Office - Dr Dianne Bond - if you want a login) is under the "Explore Collaboration" tab.   

This brings up a world map with an overlay showing how many institutions in any given country UCT has collaborations with.  The following slideshow will illustrate what happens on a drilldown - taking Australia as a starting point.

In simple terms, it is possible to see - very quickly - just who UCT researchers collaborate with, where they are - and possible collaborative partners we may have missed.

I can actually see my car, too - good job, shows I was at work that day.

 

Seriously, and although UCT does supply a set of maps of campus, these are hard to find.

It is a whole lot simpler to simply use Google Maps directly - and the depth of coverage, and extent of naming, is incredible.  For Upper Campus.  Pity about the rest...B-)

 And yes, it did help me find the Menzies Building

The coverage goes right down to street views - where you can almost see people smoking weed in the bushes you know, walking about.  Great tool for nostalgic alumni to retrace their footsteps.

UCT has recently acquired the Scival Spotlight institutional evaluation tool from Elsevier, via a grant from the Carnegie Foundation: this allows graphical analysis of an institution's output in terms of "co-citation analysis", which allows the creation of "Competencies".  The tool allows one to drill down via graphics or tables, from a "whole of UCT" level, down to individual researchers - and to individual publications, h-index calculations and much, much more, given that we now also have Elsevier's Scopus - thank you , UCT Libraries!!!

IF the researchers are represented in a competency...which is not a given!  

However, what it does do at the institutional level is allow very quick identification of emerging vs established vs declining areas of scholarship, and what is most attractive is that it does it without regard to the kinds of straightjacket that makes such a nonsense of cross-, inter- and transdisciplinary research.

We will have fun in the years ahead.  It will also be available via the Research portal.  Meanwhile, here's a slideshow for your amusement!

Given that these days a fair bit of the business of academics here at UCT seems to consist of trying to find out how good we are, at least in terms of rating of our publications, it is useful to know that there are an increasing number of good ways of doing just this.

Thomson-Reuters' ISI Web of Science:
This is available via the Libraries site (and right here), is a tried and trusted means of seeing just who has published what, how often, and what their Hirsch, or h-index is.

Elsevier's Scopus database:
This is also newly available, via UCT Libraries' generosity, and can be accessed here.  It won't give you as high an h-index if you were publishing before 1996, because it doesn't do citations from before that - however, it has a wider coverage than WoS, so may find more papers.  It also has a MUCH better way of identifying papers by the same authors.

Google Scholar:
This is becoming increasingly useful these days - and, which will please all those software libertarians, is FREE! There is a useful Mozilla add-on for Scholar, which calculates h-indices for you.  Better still, there is

Google Scholar Citations:
With a little training, this will find ALL your publications, and immediately give you lifelong and 5-year h-indices - as well as an "i10 index", which is NOT another Hyundai, but "... is the number of publications with at least 10 citations". 

Incidentally, you will almost certainly get a significantly higher h-index via GS Citations than with the other two proprietary systems.

Another, far more Humanities- and even Computer Science-friendly option is 

Harzing's Publish or Perish:
This is "...a software program that retrieves and analyzes academic citations. It uses Google Scholar to obtain the raw citations, then analyzes these and presents the following statistics..."

And then, for those those who don't like the "pay for" options AND are in disciplines not well covered by the above offerings (Scopus is NOT good for Humanities, it turns out), AND who want alternative sorts of evaluations, there's...

AltMetrics! 
I played with this during Open Web 2.0 guru Cameron Neylon's very interesting talk at UCT recently under the auspices of SCAP, when I was able to determine during the talk (via iPad) that a certain story which shall not be mentioned in polite company, had garnered some 1560 mentions in Twitter and Facebook in just two months. A very good tool for assessing blog or other Web posting impact, it would appear - and here's a great blog on the subject, courtesy of my guru, AJ Cann, via Google+.

Cameron also mentioned Total-impact.org, dedicated to uncovering the "invisible impact of research", but I haven't played with that.

Celia's Blog here at UCT has a very useful list of AltMetrics as well. 

Article describing the AltMetrics movement:

The altmetrics movement is "...a sprawling constellation of projects and like-minded people working at research institutions, libraries, and publishers" - an article in The Chronicle of Higher Education.

SO there you are: a catalogue of means of tracking yourself (although auto-AltMetricking doesn't have same vaguely sinister ring that Auto-Googling does) and improving NRF rating assessments - and possibly your promotional prospects.  If you care.  And they'll all be available via the Research Portal.

 

 

 

After a good deal of crystallisation of ideas with the Enterprise Content Management (ECM) team in ICTS, I came up with the following concept slideshow:

Of course, there are several things wrong with it:first, that's not what it will actually LOOK like; second, various of the modules will probably end up being called something else; third, aspects of it willprobably change without notice, and new things will be added as great ideas occur to people - like "My Contracts", this very afternoon.

This is an excerpt from a longer report which details facts and impressions gleaned by Ed Rybicki on a tour lasting nearly three weeks - 24 March - 5 April 2011 - taking in four Australian cities and six of their major universities, undertaken largely in the company of Sakkie Janse van Rensburg, UCTʼs Executive Director of ICT Services.

Visiting Australian eResearch infrastructure, the first thing that impresses is the sheer scale of their national support for the endeavour: they have the Research Data Storage Infrastructure (RDSI) Project, the National eResearch Collaboration Tools and Resources (NeCTAR), just being set up; the Australian National Data Service (ANDS), linking data collections of national interest for discovery and re-use; the Australian Research Data Commons (ARDC), being built by ANDS; and the Australian Research Collaboration Service (ARCS) which is now being superseded by NeCTAR and RDSI. These initiatives are very well supported in terms of funding; there is also the prospect of the National Broadband Network (NBN) coming shortly.

 Individual Australian universities have a varied degree of support for “eResearch”, ranging from the slick, comprehensive and sophisticated (Monash, University of Adelaide) through the developing (Universities of South Australia and Tasmania, Australian National University), to the rudimentary (University of Melbourne). The top end offerings were a bundle of collaboration and visualisation services, data storage and management, and software development (eg: Monash); the lower end was a more fragmented set of offerings, which did, however, include some form of centralised data storage and backup facility (eg: Melbourne). The offerings seemed to reflect the state of integration of IT services: the more centralised and standardised the service, the better and more seamless the service; the more fragmented the service (eg: due to “legacy” Faculty- or college- specific systems), the lower the service level. It was interesting, however, that the service always included data storage and partly because of “good practice” and IP requirements.

 The more sophisticated offerings tended to be by a specialised group of personnel with research backgrounds (generally PhD qualified), either as a separate organisation to regular IT services (Monash), or integrated within it (UniSA, Tasmania, ANU). The latter two also had library services within the IT departments, making for better integration of services.

 There seems generally to be the recognition that eResearch support should be tiered - that is, high-impact research groups get the most attention. A common attitude was that the support should be in the form of academic-type collaboration rather than service, and IT support of research should be planned for and budgeted for in research grant applications.

 The integration of IT support of teaching and research was variable: Monash eResearch did not; however, UniSA was very well integrated, with very impressive integration of teacher/researcher information and course offerings allowing prospective research and undergraduate students to audit the University from outside. Learning management systems were impressive, especially at UniSA and Univ Tasmania: both these had to service widely-separated campuses, and did so using video and webcasts. The integration of SAKAI (Vula at UCT) as a research collaboration tool was also widespread.

 Supercomputer (HPC) access was generally good, and this could be configured quickly and easily to support a variety of user needs - from computationally-intensive to memory- intensive - using virtual machines.

Recommendations for a UCT Research Portal

 Lessons from the Australian institutions visited were that eResearch support is potentially a wide spectrum of services, ranging from the basic - email and storage - to the sophisticated, which would include video conferencing, data storage and management, and HPC virtual machine computing. While a number of institutions did not think a “Research Portal” was such a good idea, most did in fact have researcher-specific Web pages or service access - and several had facilities for outsiders to search the institution for specific researcher skills, including engines which could throw up a biography, research profile and publication record for any specific researcher.

 My thinking, informed by the Australian eResearch profiles we saw, is that UCT needs a two-way layer for a Portal. This would consist of a Web-based environment, and have one layer for the researcher to get out, which has the following:

•  "My UCT” and “My eResearch" (eg) tabs on a UCT home page visible after login on chosen PC - with only the standard UCT page available to non-logged in or outside users
• My EResearch goes to pages which can be personalised - eg: "My Publications, My Tools, My Library" eg.
• My Publications should have a dynamic link to some search engine (or several), like SCOPUS if we get it, ISI Web of Science - and should be seamlessly available when off-campus via login or a MUCH improved version of EZProxy. It should also have tools available like Endnote Web.
• My Library should go straight into an improved library front page, again available from off-campus.
• My Tools is important because this could be very personalisable - it should have links to a variety of chosen cloud tools and tools made available specifically to that user by tiered support from ICTS / eResearch support. One of these should be “My Storage”, which would allow easy drag-and-drop file storage, for example, and even complete machine backup services. 


 The other layer would be for outside in: this should include things like "UCT Researcher Search" - which pulls in SCIVAL Strata / Spotlight / InCites data, etc, for profiling UCT researchers. This would require dynamic databases of publications / research specialities / grants / student projects, for example, which could be created with single data entries for an integrated system like the UniSA model (PeopleSoft integrated with HR / LMS systems, eg).

 An important aspect of both the in-out and out-in layers is that they should support as wide a range of devices as possible: it should be as easy to access the Portal(s) via desktop computer as via tablet as via smartphone. There should also be integration of Twitter and Facebook and other social media feeds and chat facilities. 


 Thus, it is not a trivial exercise to create, and requires an integrated approach - including web page renewal and systems integration and purchase of new capabilities. It could, however, significantly empower UCT researchers as well as aiding in more effective collaboration, and markedly improve UCTʼs “shop window”.

 

This blog will be the vehicle for news coming out of the Research Portal Project - which, for those among you who read the pronouncements in things like the UCT Research Report, is supposed to be launched in November 2011.

This is partially true: it will be partially launched by the end of November, given that integration of all of the content we would like to put in it will take a lot longer than that.

My next post will be a description of what it is we think the Portal should look like, and what it should do for UCT researchers.

Watch this space...!

Ed Rybicki

Academic Liaison to the Resarch Portal Project