What is e-Research?

What is e-Research? This question was asked at the Victoria University workshop on the Framework for e-Research Adoption and gnerated considerable discussion.  A few weeks before I had considered this question for a presentation on e-Research and thought I would share my thoughts.

There are various definitions.  When Tony Hey was Director of the UK e-Science programme he often described e-Research as enabling faster, better or different research.  This describes the expected outcome rather than indicating how it might be achieved.  The iVec website defines e-Research as the use of advanced information technology to enable better research outcomes. This now includes ICT tools but still doesn't explain what is actually happening.

My suggested definition is that e-Research is the (rapid) evolution of research methodologies enabled by information technologies and tools. It is a fundamental step change in the research process.  Jim Gray agreed, arguing that the explosion of data was in fact a new dimension to the research process.

A typical research methodology consists of a set of steps such as observe phenomena, propose hypothesis, experiment, measure, analyse, conclude and publish. The experimental step is aided by new instrumentation such as sensor networks, able to measure and collect data as never before.  Indeed Gray's argument was that we could now carry out a successful experiment by examining existing data.  We don't even need instruments! Simulation is also a powerful tool enabling researchers to use models to investigate processes that defy measurement, such as geologic processes over tens of thousands of years. All of these tools: instruments, networks, data mining and simulation, make use of advanced ICT technology.

All of these processes may generate enourmous amounts of data that present all sorts of challenges.  If the data is to be reused it needs to be discoverable and searchable.  The researcher must also analyse large amounts of data.  Increasingly this involves the use of visualisation tools to aid understanding. And visualising one experiment's results often leads to new hypotheses, so ICT can also contribute to the start of the process.

Finally publication is changing dramatically.  When did you last visit the library?

But this is only the tip of the iceberg seen in the changing research process.  Underpinning this are ICT services that are not dependent on particular disciplines, such as high performance computing, data management, collaboration tools and a network infrastructure.

The key point is not the underpinning technology but how it is changing the research process and how rapidly that change is occurring. The old processes aren't invalidated, they just don't enable the range of research paths that the new processes do.

Submitted by John Hine on

Comments

The use of technology to advance

John,


I agree with your sentiments. eResearch is a practice defined by research outcomes  achieved with ICT. I do not think it is dependent on the technology being used - innovative use of common place tech could constitute eResearch.


eResearch may be the use of advanced technology. eResearch may be the advanced use of technology. eResearch is the use of technology to advance @enabledRambler, Twitter, November 2010


The mission/definition for eResearch at Curtin University is "to enhance research outcomes through the use of information and collaboration technology". We define enhance as increasing the scale, scope or nature of the research, including new lines of enquiry – bigger, better, faster, new. eResearch is the emerging approach to research practice in the anytime, anywhere, anyone digital world. 


I have submitted an abstract for a presentation at eResearch Symposium NZ about two models for eResearch developed during the planning of our eResearch Support operation in 2009. You can get a peek at http://dbs.ilectures.curtin.edu.au/lectopia/lectopia.lasso?ut=2817&id=58404.


I like your highlight of innovation in the use of scholarly information AND the use of information and collaboration technology. They are distinct but interdependent. It highlights the emerging need for multiple skillsets in research practice. 


Regards,


Peter Hicks

On definitions

Thank you for this post John. It agrees with so much of what we've heard in the workshops for the Framework for eResearch Adoption. The discussions on what the term 'eResearch' means were very lively and there were many points of view, but there did seem to be some consensus emerging.

To summarise what I heard at the workshops:

  1. eResearch is not just about using High Performance Computing
  2. eResearch is about using technology to change the kind of research that can be done. That is:
    1. enabling research to be done at different scales (temporal, spatial, social, multi-displinary) than were previously possible
    2. increasing the degree of resolution (temporal, spatial, social, other measures) beyond what was previously possible or affordable
    3. using the above to ask new or different questions, or solve problems that were previously intractable
  3. eResearch is about using technology to change the way that research can be done:
    1. by using simulation, modelling, or data mining and pattern analysis to approach questions in different ways
    2. by involving more collaborators with different expertise, in different places
    3. by enhancing particular steps in the research process, or enhancing an entire research methodology in some way
  4. eResearch is about using technology to change the efficiency with which research can be conducted:
    1. by making it faster or cheaper
    2. by making data, analysis methods, and outputs more accessible for reuse by other researchers
  5. eResearch is also about using technology to change the way that research outputs can be taken up by others, to make a difference in the world, by delivering research outputs as electronic tools, services and information resources
  6. eResearch includes all technologies that support research, including research instrumentation, but has a particular emphasis on information and communication technologies
  7. eResearch is something all researchers can do, it's not limited to just a few disciplines or to those with advanced computing/IT skills

These seem well aligned with Tony Hey's faster, better, different, and your thinking about the evolution of research methodologies.

There was significant debate about why we would put the letter (e) in front of Research, and whether we should use the term at all. One reason was that it was so we could have a useful conversation about how to enable improved research outputs and outcomes through technology, while excluding other topics such as available funding, grant processes, the PBRF, national research priorities etc. Another was that when we use (e) in front of a word, like eCommerce, we are denoting the use of technology in quite specific ways involving a transition from one stable state to a new stable state. The jury was definitely still out as to whether we were in a transition phase and we'd see the (e) disappear again, or whether the pace of technology change would mean we need to continue focusing on the (e) in research for a long time into the future.

Following the definition was an hypothesis, that putting more (e) into research (i.e. using technology to change the kind of research that can be done, the way it can be done, or the efficiency with which it can be done), will lead to improved real world outcomes. While most people believed this hypothesis was valid, and would be proven as more evidence was collected, the degree of the difference that more (e) would make was seen to differ widely across different disciplines. There was therefore no "one size fits all" approach, but rather a range of infrastructure, tools and services that different researchers needed. There were also interesting discussions on why we, as a country, would focus on eResearch, more on that in another post.

And finally, to concur with your last point, there was definite agreement that we shouldn't be focused on the underlying technologies and a technology centric view of the world, but rather on the research process and on the researchers, and on how technology could support them.

More on this and other workshop findings soon.

eResearch definition and openness

thank you for a very useful and wide-ranging definition of eResearch, Julian, I would like to suggest your forum also be more explicit on the HOW, e.g., does not keep questions of openness out of this definition. This is in regard to the following:

Openness ranges among “science's most important ethical norms”, writes Resnik (David B. Resnik. Financial Interests and the Norms of Academic Science. In: The Commodification of Academic Research. Science and the Modern University. Edited by Hans Radder. Pittsburgh: University of Pittsburgh Press, 2010, 65-89, 71. Collection is non-OA except Chapter 1, see http://www.upress.pitt.edu/BookDetails.aspx?bookId=36159, free TOC). Academic research in the age of the Web has at its disposal a variety of technologies, of which some have been put to use in publishing. These, I would argue, do shed a new light on Resnik's description: “Openness: Share data, ideas, methods, tools, and results; be open to criticism and suggestions” (Resnik 2010, 71).

Maybe we could discuss further in an open wiki environment? (see my suggestion here http://friendfeed.com/claudiakoltzenburg/5e0067f5/eresearch-definition-j...)

Apart from Kleinman quoted above, I recently checked out and find recommendable the following definitions of openness (or defs of openness included in): http://www.opendefinition.org/okd/, http://sciencecommons.org/resources/readingroom/principles-for-open-scie..., http://scientopia.org/blogs/bookoftrogool/2010/03/15/battle-of-the-opens/, http://pantonprinciples.org/

Looking forward to more of your findings and reports

Claudia

> > 1 eResearch is not just

>
> 1 eResearch is not just about using High Performance Computing
>

I find it interesting that we are seemingly looking to define the, admittedly rather nebulous, term "eResearch", in terms of a negation of what often comes across as an equally nebulous term, vis, "High Performance Computing", when most of the "technologies" that the following points go on to describe, would appear to require computational resources that few, able or not to define HPC, would question as being of a "Much Higher Performance" than those computational resources which they usually have access to.

A slight rephrasing of Point 7 would, for me, seem to be the most important point, in terms of outlining a "framework":

a lack of HPC (whatever that gets defined as, and presumably it will get defined) skills should not be a barrier to using the HPC resources that you WILL end up using to do eResearch.

I would thus, also, rather see 1 phrased along the lines of:

eResearch is about not being aware of what computing resources/technologies you are inceasingky using to achieve your Research ends.

When did you last visit the library

I don't find this sentence helpful in the debate or that it relates to publishing. 

In terms of libraries - a more appropriate comment might be that we way we engage with libraries is rapidly changing and we may visit libraries electronically or virtually rather than always visiting a specific geographical place in person.  Libraries are an important part of the eResearch concept.