Final version at the meeting as of 31.5.1999; Last link corection: 7.1.2002.

1.6.1999 Elmau, Bayern, Max Planck Gesellschaft MPG

Workshop The Transformation of Science

Max Planck Gesellschaft MPG

Integration of Research Institutes into the future world-wide network of scientific information

E.R.Hilf, Institute for Science Networking ISN an der Carl von Ossietzky University Oldenburg, Germany
This talk resides at:

Thanks to the Max Planck Gesellschaft for the invitation and the idea of this workshop.
Thanks to the organizers for their flexibility and supportiveness.

Tnis document is a mixture of notes for the talk and more detailed comments for the MPG chief executives
(The latter part is confidential to the MPG and its referees).
This is due to that it became more and more apparent during the preparation of the talk, that one purpose of the meeting is to discuss the concept of the MPG for a new MPI (Max Planck Institute). The text is preliminary and needs overhaul after the meeting in conjunction with MPG representatives and experts.

------------- a personal view ----------------


An industrial revolution is taking place. Thus it is worthwhile to draw some paralells:

From Handwritten Information to Printed Information lead not only to more effective services but to a new era of worldwide science and their information network.

The printing age then required

Therefore one needed as service and research institutions: central professional structures (large publishing houses, FIZe, MPIs, University Libraries, Computer Centres; (professional large bodies with a small surface)

The digital era allows

Therefor one needs as service and research institutions: distributed service structures (large surface to the user, lean central structures): author's institutions as publishers, distributed networks for computing, information distribution and services.

As an example: teaching (say Quantum mechanics) in theoretical physics at a University, that is what I do, meant reading books of other authors, working out the own lecture and presenting it to the students in a linear (time and complexity ordered way, called 'Vorlesung').

In the digital age we have 65 Departments in Physics in Germany, where a colleague teaches the same material, and may be 1.000 worldwide.
These 1.000 colleagues will (or do) put their material on the Web, with interactive material, exercises, computational examples, demonstrations, etc (see e.g. PhysNet: Education).

This almost infinitely rich material will almost for shure contain material much more suitable to your or my own class than a single person ever can work out in a life time, if only one teaching term.
In addition: all this material is freely available to the students. And the learning style of students changes from linear systematic learning to 'try and error', to 'interactive working'.

Thus in the digital era even the way of learning changes qualitatively, and the quality of teaching will reach a much higher new level.

Some Requirements of professional information systems

The Role of the MPG

Short summary:

An MPG Structure for Scientific information would greatly boost the transition to a new era of digital information, and improve the MPG's Institutes visibility on the web.

Two first steps are envisaged:
1.) Setting up a visible internal structure for the Information Infrastructure of the MPG, for coherent service across all institutes.
It is important that all Institutes use for all their documents the same set of Metadata definitions with MPG-specific additions.
Thus a responsible organizational structure should be set up which defines and organizes the MPG-wide document management system, assures the homogeneity, and the professionality by embedding it into the worldwide document systems.
This step is what all companies, institutions etc. have to do at present.

2.) The second step is to work out in international context the introduction of XML, SGML, RDF, the means for separating structure of content from format, and the hierarchical metadata.
This needs an intensive scheme for establishing a standard with formluae or text DTDs with automated metadata capture to ease the MPI-members to structure their scientific information.

The MPG should gather a group of experts to design and operate the new distributed structures.
Specifically, Dr. D. Rusch-Feja as well established expert should belong to it.
Referees could be found by asking the IuK Initiative Information and Communication of the learned societies in Germany for referees and experts. The IuK has proven to be able to name on short notice really outstanding experts and referees in Information and Communication from all learned fields.

Founding a new research institute.

[more precisely we mean a
Distributed Innovation Structure for Information Management for scholarly publishing and scientific information dissemination.]

(The role of the MPG as seen from outside)
The MPG used to install a new MPI whenever a new research problem came up which called for prime research an d which needed as a means of its realization the full throttle of a new research institute with permanent staff.(e.g. the large supernova explosion codes in the MPI Astrophysik).

The MPG used to build the new institute 'around a worldwide known excellent scientist in the new field'. This as a means to assure quality.

The MPG used to give the resources needed to do the research with the necessary internationally competitive speed.

We assume that plans for such foundations were designed, then promoted, then refereed by independent internationally known experts, then discussed on an international workshop, and finally put energetically into action.

This procedure, a continuous path of symposia, refereeing, plans designing, discussion with experts of the steering committee seems to be indispensable to find solutions accepted outside the MPG.

Octopus structure of a future 'Institute':
The organizational structure of a future MPI Institute has to reflect that the topic is scientific scholarship and 'information and communication' for and between human scientists (and not a research topic in exact sciences).

As a realization the number of permanent (internal) positions should be much less than the number of guest positions and externally operating collaborators.

The new institute in its very structure should reflect the task of boosting 'information' .
It should draw upon an international collaboratory, and set up effective means of reaching, informing and training (and thus possibly convincing) the scientists 'out there'.

While in the 1950ties the solution was to set up large central databases (FIZ Karlsruhe, Spec-Info,..) and while for research requirements (such as finding new geometrical solutions for the early Universe) large and competitive Research Institutes (such as the MPI for Physics and Astrophysics) were the answer, -
for promoting the transition to the Information Age we need to form and boost and study distributed structures.
Certainly this also needs some permanent structure to organize it.. But the amount of funding should be less than 25 % as compared to the guest positions and to the external 'representatives'..

Organizationially a detailed plan could be developed by an expert committee, then to be analyzed by an independent referee group, with a subsequent workshop and discsussion.

What is different for founding an institute for Information and Communication from a research institute.

The transition from printed to digital information took place in 1994 with the LANL-Conference.

Since then it has been clear, what the next steps should be, how the new information scenario should be structured, and which technical means and tools are needed.

In the mean time the means and tools have been developed, both of and for the scientific institutions as of and for the industry. This may be read of e.g. from the presentations at the recent fairs and conferences.

Thus it no longer a technological challenge.

The real challenge in an industrial revolution is the transition of work habits: here of institutes and of individual scientists, - and of the public in general.
A rapid change and adaption of the work habits and its support by understandable, suggestive and easy to use tools and support is essential for future competitiveness in science.

We give two examples: Max Planck Institutes have been dragging behind in setting up WWW-Servers, whereas the Physics Departments in Germany virtually all had by 1994 WWW-servers of their own. We account this to the larger 'contact surface' to the outside world, the many more contacts at Universities, the larger flexibility of University groups and the existing network of operators (information management officers) at most Physics Departments.
Also we account this to the higher level of professionality the public expects from prime Research Institutes, especially from Max Planck Society Institutions.
By this one gets an understanding that institutions with higher level of professionalism have to drag behind in setting up new service schemes.

Personally we experienced ourselves: Simultaneously we had to install and adapt an inhouse document management system for our own University Department and for the AWI (Alfred Wegner Institute for Polar and Marine Research at Bremerhaven).
Whereas quickly we could install the system WUFI for the department and make it public while doing the adaptions, corrections, improvements, tests etc. on the fly, the requirements by the AWI were much more stringent: several levels of operators, editors, responsibles as well as the large set of document databases and research groups had to be taken care of by a document flow which contains a set of registration, checking, approval anbd decision levels to assure quality, correctness of scientific content, and of fitting the respective data base.
This had to be implemented before making the service public.
The advantage was that the installed adaption of WUFI is much more sophisticated.
But another outcome is that the AWI-document system might not go public before the year 2000.

More specifically to your plans:
we propose

Some remarks in addition:
The MPG and some of its institutions are at present behind the international main stream in electronic scientific publication and communication services and its usage.
There is from my view point no need (and chance) to do research on your own on designing and developing anew services which are standard for companies and research institutions by now .
The failure of the BMBF-programme GLOBAL-INFO, having proposed 5 years ago to set up a full coverage cross-disciplinary information system with all software designed and written from scratch, and with about the same amount of money, as the MPG should now willing to spend per year should be used to learn for your programme.

As you see I argue for deciding now to spend in principal the money designated.
But to design the concept and the organisational structure for the future more far reaching steps with external refereeing and with experts.

A 'Centre of Innovation of the Max Planck Society' is a unique chance to suport and promote the transition from printed to digital in Germany. This is not a research question nor one of designing tools but of action and organisation, of training, of contact, of finding effective ways or reaching scientists, of radical reorganisation of the internal information and communication of the MPG and its network of institututes.
N.B.:In some cases it may mean designing tools or collaborating with other places currently designing new tools (e.g. Global-Info, Dissertations Online, eprint, just to name a few).