European Research Exchange – EUREXH

Need and relevance

Whereas individual centres and small groups of research excellence add great value to the quality of life within their reach, connecting these entities together is likely to have dramatic effect. As Percolation Theory says: “At a certain point (Percolation Threshold) even small increase in the density of the network may result in sudden, dramatic increase in it’s power”. Many networks, with their feedbacks and cross-catalyst effects would exhibit powers that are beyond current modelling capabilities and so human’s comprehension.

Scale of ambition and critical mass

As we are already in possession of a horizontal exchange we are confident that with relatively small amount of extra work we will achieve the following:

  1. Deployment of an exchange where market places can be created on-the-fly by participants and so allow the exchange to grow in sync with user needs. Within these market places users “trade” information services and products.
  2. Allow all market places to be highly parameterised, listing all their properties and property scopes and constraints.
  3. Allow bids (requests) and offers to be matched in real-time and information disseminated in variety of ways: browser, web service, SMS, voice, fax, pager, etc.
  4. Allow incorporation of multimedia sub-documents
  5. Have exchange that personalises interests and means and timeliness of distribution
  6. Have exchange with configurable quality of service (quantifiable resilience in terms of availability and guarantee of distribution)
  7. Have exchange that is fully XML-ised.

In addition to 1-7, that are in existence already, we are proposing to augment the capabilities of the network by following:

  1. Allow distribution (over Internet) of servers (instances of exchanges) with single directory master and fail-over capabilities.
  2. We will create clusters of markets within the following broad categories:
  3. Artificial Intelligence
  4. Communications
  5. Molecular Biology
  6. Investment Banking
  7. Management
  8. Higher level research services

The following institutions have already agreed to participate:

  1. JRC Online, a financial services SME which spends 50% of its turnover on R&D projects.
  2. FinSoft, a UK-based financial and e-commerce software company which develops platforms for major European financial traders.
  3. Ibermatica, a Spanish computing and IT organisation that specialises in presentation software and computer vision applications.
  4. University of Surrey works in the area of information extraction in financial applications and in cultural heritage.
  5. University of Sheffield works in the area of information extraction in close collaboration with major news agencies and pharmaceutical companies.
  6. University of Birmingham pioneered key linguistic methods and techniques to examine large corpora (20GB).
  7. University of Trieste works in the area of translation and interpretation, focusing on economics and financial subject areas.
  8. Q-Sphere, a horizontal Internet exchange of information, services and products.
  9. University of Southampton works on a GRID computing project under the UK national e-Science initiative.
  10. Scientific Generics is an integrated technology consulting, development and investment company

In addition to these institution this consortium will allocate large proportion of funding to no less than 10 additional participants. Extensive consultation and visits will be conducted with no less than 20 European centres of excellence.

Integration

EUREXH the following key integrational properties:

  1. Creation of market hierarchies and market-places within them is dynamic. Users with appropriate privileges can do it at any time. This way, EUREXH is consumer centric and allows for fine tuning towards what users need.
  2. Market places are highly paremetrised. This way (more parameters, stricter constraints) users are in the charge of the strength of typing. “Noise” (misplaced information) can be controlled to a large degree.
  3. Exchange is asynchronous. Users don’t only get what they need when they are online. They also get the information as and when it becomes available.
  4. Distribution is resilient. If importance of information, on a certain topic, is very important, EUREXH can be instructed to distribute it using alternative means should the primary one fail.

Three-dimensional nature of EUREXH (time (bids/offers can be limited in time/frequency), market-places, communication channels) forms an basis for covering most of what Internet can offer at the moment in addition to adding sharp, qualitative edge to the effectivness of information exchange. Here are a few examples to illustrate the point:

  1. Lightly parameterised market place “AIResearch” with mandatory and searchable parameter “topic” and mandatory, non-searchable parameter “document” where distribution is allowed only by e-mail would equate to what most companies use to distribute information over the Interned today. Key feature on EURECH is the ability to accept documents from multiple publishers.
  2. Lightly parameterised, password protected, market place “GenomicsWithinEC” with two, non-searchable parameters “headline” and “URL” where distribution is by online news ticker (primary discipline, if user is online) or e-mail (secondary discipline) will equate to discussion group where whenever some says something on the topic headline ticks on everyone’s screen.

EUREXH’s nature is open-ended. Client facing part of the application is implemented on plug-and-play basis. Subject to approval (grants of various rights) clients can extend client-facing behavioural semantics by attaching their own modules.