Free open-source software kernel helps unlock web resources

Eighteen months ago, the EU-funded OpenKnowledge program set out to provide internet users with a simple set of tools to find and interact smoothly with trustworthy partners across the web.

OpenKnowledge created a new computer language, crafted smart solutions to the problem of semantic mapping and found ways to ensure that partners are reputable.

The researchers packaged these capabilities into an open-source software kernel that is now freely available on their website.

“You can go to www.openk.org and download quite a small piece of Java code that will allow you to do this coordination and checking and sharing of processes,” says Dave Robertson, OpenKnowledge project coordinator.

To test — and showcase — the capabilities of the OpenKnowledge system, Robertson’s team has applied it in three very different areas: coordinating healthcare services, emergency management and proteomics research.

Robertson hopes that these case studies will inspire people worldwide to utilise OpenKnowledge to forge potent new partnerships across the web.

Coordinating proteomics research

Just as OpenKnowledge can make useful healthcare choices more widely available, it can help researchers access and analyse the rapidly growing mountain of data on the structure and function of proteins.

Working with proteomics researchers, Robertson and his colleagues identified several areas where OpenKnowledge could make a difference.

They found that researchers worldwide relied on a small number of databases. “That puts a lot of stress on those databases and on the people and institutions that curate them,” says Robertson. “They act like a bottleneck.”

When the OpenKnowledge team assessed the quality of the information in those heavily used databases, they found problems. “Actually, they’re of very mixed quality,” Robertson says.

In addition, researchers in different groups found it hard to share data and results with other groups, even when those could be of great value. “One group’s garbage might be exactly what another group was looking for,” says Robertson.

Taking control of your own healthcare

The OpenKnowledge team responded to the UK’s Grand Challenge Project for information-driven healthcare with a proposal to “completely change the way patients view the management of their healthcare pathways”.

“If you manage your healthcare in a conventional way,” says Robertson, “you go to your national health service or a private provider, and they’ll tell you the pathway you have to follow to get a diagnosis, acquire medicines or receive treatment. ‘Here’s the plan; you follow it!’”

The OpenKnowledge approach turns that process on its head.

“There are many different pathways out there on the internet,” Robertson says. “You pull one out that has a high reputation, you inspect it, and if you’re happy with it, you take it to a healthcare provider who may help you enact it. So you, the patient, are taking much more responsibility.”

Of course, many people already seek healthcare guidance on the internet. However, that information tends to be disorganised and comes with no guarantees as to its accuracy.

In contrast, OpenKnowledge can provide patients with structured information that has been checked for accuracy.

OpenKnowledge is currently working with Cancer Research UK on a project to better manage cancer treatment protocols.

“You can say, ‘I’ve found this set of keys that I think would work for me’ and see if they will work in the medical care system,” says Robertson. “It’s a new way of thinking about the problem.”

Making emergency response more resilient

The OpenKnowledge team is working with emergency response authorities in Trentino, Italy to explore decentralising the command and control process.

“Typically, information from people and sensors comes into a control centre, and that centre tries to get information out to people who might need to evacuate, for example,” says Robertson. “You need that kind of control, but what happens if that centralised system breaks down?”

The researchers used OpenKnowledge to model Trentino’s centralised control system, and also to model and simulate more distributed systems.

“It turns out that we get comparable performance,” Robertson says. “If you were a bus driver trying to evacuate flood victims, you could find out if the road ahead was open even if the control centre were taken out. We can now imagine the decentralised system being used as a backup when the centralised system fails.”

OpenKnowledge has helped solve all three of these problems. As more researchers join and use the system, it becomes easier for them to share both raw data and results directly with others. That relieves the pressure on the main databases, while feedback continually improves the quality of the data being shared.

“We’ve linked into an existing proteomics network in Spain, ProteoRed, which is evaluating the system,” says Robertson. “It’s no longer just a simulation.”

Robertson is encouraged by the demonstrated ability of OpenKnowledge to coordinate and add value to endeavours as different as proteomics, emergency management and healthcare.

Still, what he’s really hoping is that inventive people everywhere will ask what OpenKnowledge can do for them.

“I’d like to see people use OpenKnowledge to write and use and share keys that would let them unlock and coordinate the dynamic information that’s on the internet for them, but that until now has been out of reach,” he says.

OpenKnowledge received funding from the ICT strand of the EU’s Sixth Framework Programme for research.

ICT Results