Session 3: Survival of R&D and Repositories

Audio from the session
To downlod the MP3 click here

This session focused on the idea that the uptake of technology and innovation is not only a technical issue.

The human aspects that came up in session 1 were discussed in this session – from the software and repositories perspective.

Neil Chue Hong from the Engage project explained why good software sometimes dies and what we can do to save it! He defined the human factors needed to build successful sutainable comminities and feed the ‘Free Puppy’ of open source software…

Bill Hubbard reviewed the history of the ‘mountaineering’ initiative to get repositories generally supported by the UK academic community. 80% of the community completely supports it now, but that is not the end of the discussion…
Obstacles and policies were discussed at a dazzling speed and from great hights What is the future for repositories?

The session kicked off with a lively and often amusing talk from Neil Chue Hong on the reasons why good software sometimes dies and what we can do to save it. His full Powerpoint is here: Why software sometimes dies and it’s very useful and easy to follow. What follows is an overview of the talk and the powerpoint will add more detail. Neil’s talk received a good reception and people really got involved during the group work.

He talked about his involvement with the Engage project, which has done some research into the issue of why people use (or don’t use) software.

Neil then went on to talk about how we can engage software users in a community around the software. He talked about how software communities have parallels with other types of communities, for example a cinema community he is involved with or a gardening community – all these communities need to be inclusive and grow together.

Neil moved on to long tails – do people really know exactly what they are? Audience members gave examples of Amazon, eBay and small software companies. Niche companies can gain access to a large marketplace through the long tail. Libraries are another example of a successful type of long tail organisation.

However, does this approach work for software asked Neil? Can we compare an online retailer such as Amazon with a software repository? Neil likened this to comparing…apples and books. It is about shelf life.
Books have a very long shelf life. Software, like apples, does not. Its users’ needs change and operating systems change also.

Another of the problems with looking at the long tail is that software doesn’t have negligible stocking and distrbution costs. Neil talked about collaborating to reduce these costs, asking how can we, as developers look at ways of reducing these costs? And how can users help developers to do this? We all have a part to play, he said.

Neil asked the audience if they thought open source software was “free speech” or “free beer”. He said (quoting something he had heard) that it is actually more like…a free puppy, because it needs attention, it may lose its charm and it means some long-term costs (see powerpoint for very cute dog picture!).

He said that researchers and developers need to be good at estimating software costs and finding ways to reduce them.

Neil pointed out that we need to be aware of the ways that open source software is different to commercial software offerings and this means being aware of the benefits that it has as well as the limitations.

The audience were asked to do a short exercise in groups. They were asked to think about the types of software that they used and specifically why they chose the software. After three or four minutes of lively group discussion some answers were:

“It already exists”
“It works”
“It’s usable”
“It complies with standards”
“It doesn’t interfere with other programmes”
“It’s cool” (As opposed to just using Miscrosft products!)
“It’s easy to implement”

Neil then asked what would make people stop using a certain piece of software.

Answers included:

“If it became no longer supported”
“If something better came out”
“If it no longer did what I wanted it to”

Neil then showed a slide (the full presentation is available on this post as a link above) which showed some feedback that the Engage project had received on the issue of software choices.
Neil pointed out at this point that he is keen for any more interviewees for the Engage project!

Neil continued the question of how can we develop a sustainable community around a piece of software? He used the quote ““Give a man a fish, and you feed him for a day.Teach a man to fish, and you feed him for life.”

To create sustainable communities, we need four things:

– cohesion and identity
– tolerance of diversity
– efficient use of resources
– adaptability to change

What does this mean for the long tail and software? To get a large enough availability of choice requires a comparable marketplace, eg other projects going on within JISC, and a large population of customers. The main area to address is reducing the costs.For developers of a software component it is the efficient use of resources and technology that can make the biggest impact, and allow the long tail to work for software.

Neil talked about basic software economics discussing finding out what the value of a piece of software is and who it is valuable to. Small and niche pieces of software can be extremely valuable because they often do exactly what the user requires of them, even if it’s only a tiny amount of users.

Neil asked the audience if they had ever contributed to a piece of software that hadn’t been written by their group. If they hadn’t, he asked why not – was it because you didn’t need to or because it was too difficult? Nobody said it was because it was too difficult.

Neil talked about creating a community around a piece of software. It is about developers knowing what users want to do NOT how they can do what they want with what’s currently available. It can be difficult to decipher what it is that people actually want to do though.

Smart Growth through collaboration was discussed with questions raised such as “How do you turn users into collaborators and then into contributors?” And “How do you move from a single team at a single organisation to a more diverse, sustainable development?”

The points that followed were on How to leverage infrastructure (for example making it easy for new developers to use), issues surrounding governing sustainability and software as a shared facility – and some of the issues this raises such as the lack of cohesion if communities become too large.

Neil talked about thinking about several small communities rather than one large ones.

He summarised with his views on how to keep good software alive:

  • Understand the valueIdentify the community
  • Leverage technology
  • Improve process
  • Keep people engaged
  • Encourage contribution
  • Above all, if you use it, make sure people know!

Bill Hubbard then took over. He is manager of the Repositories Support Project and manager of SHERPA.

His Powerpoint presentation is available here:The Repository Story . It provides more details on his presentation.

He talked about the development of repositories and the possibilities for their future. Bill likened the Open Access Repository journey to a mountain to be climbed – daunting but full of promise.

The landscape

Bill first gave an overview of the Open Access landscape as it stands now. He went on to discuss the route to institutional repositories and the “trail” into this new area.


Institutional repositories have gained support from areas such as Institutional mandates, funding mandates and an NIH mandate (further voices of support are detailed in the Powerpoint).

Reflections on the repositories journey so far

Bill reflected on the repositories “path” so far, looking back at what has been achieved and what has come of early worries. There were many positives, such as the fact that Open Access has an 80% approval rate (even if people aren’t actually using it. 80% of academics asked reported that they would be amenable to using it). Bill also acknowledged that there are many obstacles still to overcome, such as copyright fears, authors not wanting to upload their own work, metadata-only tagging and embargoes to name but a few (the full set of obstacles identified is available on the powerpoint).

What next?

There is plenty to work on in the future to reach the top of the repository mountain.

The next steps are:

Mandates – support them! support compliance, integration, stakeholder involvement
Focus – choose a clear path
Metadata repositories – remember the advantage has to be for users, not statisticians
Engagement – academics support the idea: allow them to support the repository
Services – offer services that make life better!

The important issues that need to be addressed are about people not technology.

Hopefully the view from the top of this mountain will be a good one.

Q and A

Pro VCs often feel that repositories are showcases and only want to include peer reviewed articles or separate pre and post-peer reviewed articles. Has thinking moved on?

Splitting repositories doesn’t matter as the search will pick up everything and can filter anyway. The repository is a service for academics to use however they want within their own research. It is a discussion for the Pro-Vs and the researchers and not really concerned with the repository itself. It is political, not structural.

Alaistar Darling – JISC
You say 80% academics use Open Access repositories….80% people want world peace too but how many do anything about it? The next step is surely to talk about this issue outside JISC and in the wider academic community where concerns still exist?

Yes, there is a continued need for advocacy. Culture changes take a long time

1 thought on “Session 3: Survival of R&D and Repositories

  1. Pingback: National Centre for e-Social Science BLOG » Blog Archive » JISC Innovation Forum

Comments are closed.