RDF
hacking ILRT Home

Report on Key Action III Open House March 12-13th, 2001 - Luxembourg

Author: Libby Miller <libby.miller@bristol.ac.uk>

Date: 2001-03-19

This open house seems to have been primarily aimed at organisations who wished to bid for options in Call 6 of Key Action III (in the 5th Framework Programme). This meant that many of the sessions were aimed at consortia who had already made plans to bid under this call. There was also a lot of time allocated to people who wanted to discuss proposals with EU staff or network to find partners under the call.

There was also a specific session for the semantic web, an action line of Call 7, to be held in June 2001.

I attended

Below is a short discusson of each of these, some conclusions from each one, and my overall conclusions about bidding for call 6, call 7 (semweb) and in general.

2001-03-12 General discusssion of Framework Programme 5 and IST

There were several presentations intended to situate calls 6 and 7 into the context of FP5 (Framework Programme 5) and IST (Information Society Technologies). FP5 is a time-based grouping of projects funded by the EU. Each framework programme budget has to be ratified by the European Parliament. They last 3-4 years each, and FP5 continues through 2002. Within FP5 there is the IST (Information Society Technologies) theme of EU research and technical development which will also be a theme in FP6. Key action III is a vertical grouping of activities within IST about multimedia content an tools. Each Key Action needs to have a balance of activities from basic research to demos to take-up products.

Key actions are divided into 'action lines'. Projects can be across action lines, and can be multi-disciplinary.

Norbert Brinkhoff described the aims of IST:

IST aims to create tools for knowledge acquisition, representation and management. Functionality includes creating, accessing and filtering information (e.g. search engines). It also includes the creation of intuitive interfaces for multimedia access (also in FP6)

55-60 MEuros are available for each of calls 6 and 7. As a measure of the likelihood of success in bidding for these calls, in previous calls, about 1/4 of applicants were selected from the total of applicants (in call 3), 1/3 (in calls 4 and 5).

In Action Line III the focus is on long-run R&D (5-10 years) where appropriate. In addition the new concept of the 'semantic web' is introduced.

Spanning actions apply to topics which fall between Action Lines or beyond Action Line III. They are:

General tips for bidding

You are expected to show:

They prefer paper submissons

In the future they will be looking to fund a few large initiatives (see FP6 discsussion)

Safer Internet Action Plan

There is a link between Action Line III and the safer internet action plan. Currently there is a call under this plan open until 31 May 2001 for filtering software and services.

E-Content Programme

Also connected to the e-content programme (Yvo Volman), a new, separate programme to Key Action III, with the following priorities:

The e-content programme started on 18 Jan 2001 for 4 years. 100MEuro is available under this programme.

The e-content programme has three action lines:

There is an information day for this programme on 26 April 2001 in Brussels.

Conclusion to the general presentations

The structure of EU funding calls is extremely complex and confusing. More detail about the Action Lines mentioned here and others, and some useful diagrams, are contained in The Information Society Technologies 2001 work programme. I've got a paper copy, and there's an online copy at [1].

2001-03-12 Cultural Content and Digital Heritage - Bernard Smith

'Heritage for all' and 'Next generation digitial collections' are Action lines III.1.2 and III.1.3 of IST repectively. This was an excellent presentation which explained exactly how to go about applying for funds under these two action lines in call 6.

Smith first explained that it was not enough to ground any bid solely in the relevant action line. Instead you also need to ground it in both IST and also the 5th Framework Programme. He talked about how project bids were evaluated. To succeed in general in the evaluation process, the proposer needs to:

In these two specific action lines, projects to be funded require a very high standard of research. It is essential to have partners in the right areas and maybe in non-EU countries like the US (NSF) and Russia.

Projects under these action lines need content and users. They will be evaluated on 'access to scientific and cultural content through networking of libraries, archives and museums'.

'Scientific content' is interpreted very broadly - it could even refer to soon to be created scientific content. They are very keen on scientific content.

Economic sustainability is vital. Smith could not emphasise this enough. He said that the project needs an economic motivation and an exploitation plan. He told us to find new combinations of partners who have not worked together before. These two calls apply to varied multimedia content (?film). They concern improving functonalities of archives, and are open now - from Jan 2001.

In terms of partners, 1/4-1/3 should be from cultural institutions, and 8 participants is a good number. The projects should last 24-30 months. The award on average is 1.8 MEuro.

For 'Next generation digital collections', prospective applicants should look at previous applicants to see what themes are already covered and think about synergies between those already funded themes and new themes.

Smith suggested using the networks (e.g Cultivate) to find partners. Applicants need a partner from a spcecific area of cultural heritage - the project must not be about generic cutural heitage. Projects must include a partner from a specific cultural sector. Projects must refer to the technical state of the art, and must have a problem and solve it.

The first evaluation criterium is R&D - technological innovation, especially in WP 2001.

The aims of a project under these calls should be:

The project should be such that:

Projects must not be just digitizing a collection. They should be long term (5-10 years, maybe longer). For this action line, it is probably too late to start from scratch now: intead, you should find an existing consortium, which should have spent 4-5 months planning already. Only 4 or 5 projects will be funded this year.

Heritage for all

Heritage for all is different: you could start a proposal to bid for this now. Buzz-words are 'online communities', 'social and cultural inclusion', 'tomorrow's digital content, and how to preseve it.

Examples of topics under this action line include:

Note: this action line has a 3-7 year horizon, with project results in 2004-5: this requires matching vision and technology.

In general

Always check out project proposals with EU staff before submitting them. They always try to give useful advice. [From questions: most cash from call 6 will be recieved in Feb/Mar 2002].

Don't forget x-content futures: projects like I cubed, lasting 2-4 years with a vision of 10+ years.

In all projects you need to propose a proper structure and evidence of continual motiviation over time, and stamina. Don't forget that it might be easier to get a small project funded with 5 partners, but if you want to make a big change and you need consensus for this, you will need lots of partners.

Conclusions to cultural content and digital heritage presentatoion

This presentation gave very specific advice about bidding for these aspects of call 6. I hink the ILRT would be well-placed to bit under Next generation digital libraries, since we have the library experience plus the advanced (RDF!) technology to do something interesting. Plus we have TASI, BIOMED to represent multimedia scientific content. The worry is that we are too late to start creating a consortium for this call. Heritage for all is also interesting - Dan B's 'seeing ghosts' project might be fundable within this call.

2001-03-12 Semantic web technologies

Note: There will be semantic web calls in 2002 as well, and also 2003.

Several presentations were made on current and past semantic web technologies funded by the EU, others on the EU's criteria for funding projects under call 7 (June 2001) of IST.

Previously there have been semantic web projects funded by the EU: in the first batch these were about intelligent web search, especially images; metadata, mpeg; and innovative interfaces. Recently funded projects have concentrated on information visualization (of databases); mobile and domestic multi-media; streaming video and interactivity. Now there is going to be a new action line for 'semantic web technologies'. The presenter noted the current semantic web activity at the W3C, but said that unlike the W3C, the EU were looking for tools, applications and technologies.

Hans-Georg Stork discussed the semantic web action line. He talked a lot about mobile technologies and the semantic web (they think that Europe is ahead in this area). He put a lot of emphasis on the projects already funded as being the most important semantic web applications currently available. He talked about Onto* projects and OIL (not 'DAML-OIL'). He had a picture of the semantic web as envisaged by the EU containing the terms 'OIL' and 'XML' and a big mobile gateway. He placed heavy emphasis on automatic content analysis; on ontologies and how they must be made more sharable and interoperable; and on agents. He made brief mention of information retrieval and standards.

There are four strands to the semantic web technologies action line:

Stork also mentionned sustainability (especially with respect to reference ontologies); scaleability; platform-specific constraints - TV, mobile, UMTS, CC/PP (especially ease of access (see FP6)), content presentation, and geographical information integration [would be nice but would need some heavyweight partners]). Stork suggested we should find partners through Dieter Fensel's Ontoweb. He said: 'the future is the web of knowledge!'.

Questions:

There was then a discussion of currently funded activities by Pierre-Paul Sontag (I have a handout for this). All seemed small scale, non-scaleable and not that interesting, with emphasis on agents and ontologies. Lots of projects have already been funded; 10 R&D projects are in negotiation in the field of mobile and domestic systems and information visualization. He said that there was scope for following through sucessful previous projects. It is possible for EU to directly fund participation in the standards process.

Presentation about Cross-Project actions CPA1 and CPA9

CPA1: Home of the future. The aim is the integration of novel and exciting technologies for the home: 'adaptive intelligent environments'.

CPA9 - the GRID. The Grid is a project that has been going for the last 2 years for world-wide distributied computing and to make data resources seamlessly available for computer intensive disciplines such as physics, meteorology, health. The Grid needs semantics [lots of interest]. CPA9 is for community-wide Grid development in the EU.

General info

16-20 MEuro will be made available for the semantic web technologies action line in 2001. They would prefer a research/industry mix of partners. Possible projects would include demo/pilot applications, good/real content.

[Questions: A very disappointed guy said that the money was peanuts; A: the EU is specifically focussing on international issues, not reinventing the wheel.

2001-03-13 Georg Stork - general talk on the Semantic Web

This was a presentation to all attendees about a vision of the semantic web. Stork stood in at the last minute for someone else.

Stork started with Tim BL's diagram with 'ontologies' in the middle. He said he regarded ontology as the most important part of Tim's diagram. He defined an ontology as a 'common understanding of a domain', playing an analogous function to theseauri, classification systems etc in traditional information systems. We will eventually get a 'web of knowledge'. Stork talked about the expense of cataloguing, and emphasised automatc content extraction.

Conclusions

The EU are interested in the semweb, but in the very long term, and their interests do not apprear to currently coincide with ours at the ILRT too much. In particular they don't appear to have a web-sized vision. The projects the fund are not part of a new web; they are separate semantic applications.

The EU are interested in 'ambient intelligence' which means barely perceptible user interfaces - getting ordinary people using technology all the time, especially in the home and particularly with repect to mobile devices. In contrast, the hertitage and next generation digital libaries strands of call 6 seem to be very much in ILRT's area: they concern the application of very new technologies to real collections. Also we now have very good advice on applying for funds in these action lines, although it may be too late.

We should also bear in mind that money from the call 6 (Jan) mostly won't come through until Feb 2002; we can therefore expect cash from the 7th (semweb) call in June 2001 also to take a year to come through (part of the problem is the limited funds available in each financial year).

The EU are moving towards very large scale projects in this area. ILRT will have to decide whether to go all all out or remove themselves from the running and seek funding elsewhere.

Appendix - interesting talk on Digital libraries

2001-03-13 Dieter Felner

This was a presentation about 'nagging/challenging research issues for digital libraries for the next 10 years'. The initial was list based on V3D2 strategic research initiative and discussion with many distinguished colleagues in the field, but is open to more suggestions. First open discussion: EU-DL ALL Projects Consultation meeting Feb 8 2001

Felner is interested in digital libraries for many different applications, for example help desks, company libraries; any form of 'data holding environment'. The problems he identified were too much information and lack of relevant information: he talked about the expense and skill required to organise and describe the information, and the need to deskill and cheapen this work. He described the functions of digital libraries as

Key issues from consultation.

user interfaces. Users get lost in hyperspace. Navigation in multimedia documents. Level of detail, progressive transmission for video; links in non-text documents.

Huge amounts of data. (Intellectual copyright issues - give away info down to a certain level of detail)

open problems:

http://graphics.tu-bs.de/DLResearch/issues - a 'living document' d.fellner@tu-bs.de

questions and comments:
library as 'data-holding environment' i.e. include museums and archives, libraries.
linux and open source was going to be important to EU, but not heard a lot about it. A: in education side, encouraging open source software. Ariadne.

References

[1] Information Society Technologies 2001 Work Programme
http://www.cordis.lu/ist/workprogramme.htm

Information Society Technologies (IST) website
http://www.cordis.lu/ist/

Proceedings of the open house:
ftp://ftp.cordis.lu/pub/ist/docs/ka3/ka3_oh_12_13_03_2001.zip

November Luxembourg workshop presentations
http://www.cordis.lu/ist/ka3/iaf/swt_presentations/swt_presentations.htm

Ontoweb
http://www.ontoweb.org/