From peter.robinson@english.oxford.ac.uk Tue Oct 26 10:11:43 1999
Date: Sat, 16 Jan 1999 10:58:17 +0000
From: Peter Robinson <peter.robinson@english.oxford.ac.uk>
To: mjd@coco.ihi.ku.dk, adolf.knoll@nkp.cz, Zdenek.Uhlir@nkp.cz,
     lou.burnard@computing-services.oxford.ac.uk, reb@de.ibm.com,
     bove@fotomr.uni-marburg.de, anne.korteweg@konbib.nl, Hans.Jansen@KONBIB.NL,
     Ad.Leerintveld@KONBIB.NL, lalou@msh-paris.fr, gougerot@cnrs-orleans.fr,
     cwd3@columbia.edu, jsimons@dmu.ac.uk, Patricia.MANSON@LUX.DG13.cec.be
Subject: meeting minutes

Dear folks
Here are the minutes of Tuesday's meeting done in a multiplicity of
formats.  First, they follow here as text. If you want to read them in
nicer shape, you can read them on the web at:
http://www.dlib.dmu.ac.uk/~peterr/master/private/minutes199.html

User id: PRIVATE
password: anastasia

I will wait till Monday for comments and corrections.  I will then revise
them (if necessary) and make them available to associate partners also
Alas, my struggle with DMU networking to get our web site active on the
address we want is still not won
best wishes all

Peter


MASTER
Manuscript Access Through Standards for Electronic Records
Draft Minutes of kick-off meeting, 12 January 1999
9 am - 5.45 pm, Room 0.35, Trinity House, Leicester

Key: DMU De Montfort; IRHT Paris;  NLP  Prague; OU Oxford; KB  Royal
Library, the Hague; EAMSS US partners; VL Vatican Library; AP other
associated partners; EG Expert group; BFM Marburg
Present: Anne Korteweg (KB), Lou Burnard (OU),  Zdenek Uhlir (NLP),
Elisabeth Lalou (IHRT), Muriel Gougerot (IRHT), Pat Manson (CEC), Consuelo
Dutschke (EAMSS), Belinda Egan (EAMSS), Peter Robinson (DMU-CTA), Adrian
Welsh (DMU-CTA), Matthew Driscoll (AMI)
Apologies: Professor Judy Simons (DMU-CTA) due to heavy snow in Baslow

These draft minutes also include some material arising from discussion
since the meeting.  Parts of this document will be absorbed into the
Project Handbook.

9.00-10.30
PR opened the meeting with a heartfelt greeting: at last, we are here and
ready to start!  He thanked all for their commitment to the project through
the long two years between the Studley meeting which conceived MASTER and
the start of the project itself, on 1 January 1999.  He outlined the agenda
for the day, and then asked Pat Manson to give the admininistrative
introduction to the project, as designated EU Project Officer.
PM took the meeting through all the various contract conditions and
administrative procedures laid down by the EU, and expressed in the various
documents surrounding and in the contract.  She stressed the following:
1. It is the project's choice on how the 15% advance on the project
expenditure should be distributed (see below).
2. The need for efficient reporting procedures within the project (see below).
3. The project must conclude a Consortium Agreement to define the project's
internal rules and procedures, and especially ownership and rights to
intellectual property developed by the partners
4. Currency rates for the purposes of computing EU payments will be those
applicable on the first day of the first month following the end of the
period being claimed for (for us this will be 1 November 1999), as given on
the official EU web site.  Euroland partners will claim in either local
currency or Euros dependent on the accounting practice of their institution
5. On the all-important part D cost statement form, to be completed each
ten months by every partner:
1. Original forms must be submitted to the EU WITH SIGNATURES IN BLUE (so
that the EU can be quite sure that this is the 'original' and not a print
of some kind).  We were entertained to discover that the EU appears not
have heard of colour printing.
2. The financial manager and technical manager who sign the Part D Cost
Statement should NOT be the same person

11-12.45
Reporting PM outlined the following reporting requirements:
1. Two monthly management reports from each partner, with statements of
resource use but NOT cost statements.  These to be consolidated by DMU/CTA
and passed to the EU, within one week of the end of each two month period
(one paper copy, one electronic copy).  The first of these will be due on 1
March 1999.
2. Ten monthly progress reports from each partner, with statements of
resource use AND with cost statements (the Part D form).  These should be
the consolidation of the previous five management reports, and must be
passed to the EU within one month of the due date (three paper copies, one
electronic copy).  The first of these will be due on 1 November 1999.
3. An annual report, capable of publication.   This should be largely
extracted from the Management and Progress reports.
4. A peer review, to be held in Luxembourg as and when the EU requests it.
This will most likely be in the last three months of the year.  The review
will be based on a status report IF the annual report is NOT available.
Deliverables PM discussed the deliverables format, and handed out copies of
the EU deliverables guide.  She pointed out that we, the partners, are to
provide the 'keywords' in the deliverable submission sheet.
Exploitation PM explained that the consortium agreement must specify ground
rules for exploitation of property rights, and that these ground rules must
then shape the project exploitation plan.  The project and each partner
must specify a 'technology implementation plan' in which each partner will
declare what each will do with the materials developed in MASTER.  The EU
is in the process of creating guidelines for exploitation appropriate to
libraries or similar.
Concertation  The partners must attend EU concertation meetings and must
develop relationships with complementary projects
Staff costings  The EU require completion of the forms on page 3 of the
overheads questionaire for ALL staff funded by the MASTER grant.  PR will
ask the partners who have NOT yet completed these to forward them to him.
Staff contracts The EU requires a COPY of the contract for EACH billable
person working on the contract.  For the 100% additional cost partners,
this is crucial to ensure that only 'additional' staff are funded by the
project.  This lead to some discussion as to the exact definition of
'permanent' and 'additional' staff, for these purposes, and the
possibilities for uncertainty over the exact status of individual staff.
Timesheets  Each partner is to keep a timesheet for ALL staff.  For costed
staff, this must be signed by an independent person (ie not by the person
himself or herself)
Invoices Each partner is to keep a copy or other record of all invoices,
etc, relating to reimbursement of travel costs, equipment.   SO long as the
copy or record is sufficient to satisfy the partner institutions own audit
requirements, this will be enough.  That is: if the partner does NOT need
to keep the physical originals of tickets, receipts, etc, then the EU does
not require this either.
VAT No partner anticipates having to claim VAT from the EU for purchases,
etc.  That is a relief!

At 11.45, PM handed the meeting back to PR.  The meeting then discussed the
reporting requirements for the project, and how they might be achieved most
efficiently.  At LB's suggestion, the meeting agreed that the most direct
way to handle reporting would be to have ALL the partners complete all
forms (so far as possible) over the web.  AW agreed that he could implement
such a system, for the two monthly reports at least.  This will greatly
simplify project reporting, and was greeted with enthusiasm and relief.
(Following the meeting, PR and AW have discussed this further, and think we
can do this in such a way as to make it clear to partners what they SHOULD
be doing each two months, on each task, as well as allowing partners to put
in what they actually DO each two months.  See below for a fuller
description of what we propose)
PR then began to go over the whole project in outline, in order to remind
us all of what was to be done, by whom, in what order, in each part of the
project.  This provided many opportunities for lively interjection and
clarification.  The closeness of MASTER and the key related initiatives of
EAMMS and the TEI work group were stressed.  We intend that the MASTER
standard be completely conformant to that prepared by the TEI group, and
the presence of so many MASTER members in the TEI group should guarantee
this.  See further the discussion below
By lunchtime, the outline had reached WP 5.  There was particular
discussion of the relationship between the SGML and database branches of
the project.  There is general agreement that the project must enable full
translation between records made in each format for first level records.
Indeed, first level records are defined precisely as records which can be
exchanged without loss of information between the SGML and database
implementations.  However, we have not yet decided exactly what the
database will run on, what features it will contain, etc: this is a task
for the technical committee, and must be addressed in the next meeting.

Lunch: for some, a walk in the rain to Leicester Castle.

2.00-4.30
Consortium agreement
Over lunch, PR had copies made of the consortium agreement established for
the ELISE project.  This was handed to all the partners and made the base
of discussion for the consortium agreement for MASTER.  Only two areas of
that agreement appear to need altering:
1. The section describing the consortium governing committee.  The Project
Plan sets out slightly different procedures to those specified in the ELISE
agreement.  This will then have to be adapted to conform to those in the
ELISE agreement.
2. The section relating to IPR.  It is crucial that we establish the
ownership of materials, software, and all goods created by, within, and for
the project.  Further, we must specify what happens to them after the
project, in terms of the base for consideration of these by the
exploitation plan.
With advice from Pat Manson, and after discussion by the group, we
determined three categories of IPR goods to be created by the project, and
agreed an outline of how we thought these should be treated.  These are:
1. Standards materials related to the TEI standard, in the form of the TEI
Document Type Definition to be developed in partnership with MASTER, and
all derivative materials associated with the standard (as: rules files
generated from the DTD, subset files of the DTD, documentation of the DTD).
The partners agreed that we shall not assert exclusive rights to any of
these
2. Software and derived products developed for the project by the partners.
The partners agreed that all software and derived products are jointly
owned by all partners.  The partners will determine an exploitation
agreement which will set out transfer and exploitation rights for all such
products after the project's conclusion.
3. The Intellectual Property Rights in the manuscript records made as part
of the MASTER project.  We agreed that these will be the property of the
partner contributing these to the project.  Each partner contributing such
records will grant all other partners non-exclusive right to the use of
these records for demonstration and development purposes.  This right to
non-exclusive use for demonstration and development purposes will continue
beyond the end of the project.  Other use of the records by the partners
after the end of the project will be determined in the exploitation
agreement.
Since the meeting, PR has spoken to DMU's European Funding office (Sue
Boorman).  This office will take legal advice on these issues,will advise
us on their practicality, and commence drafting an exploitation agreement
embodying these points.  Partners should check with their institutions that
this outline is acceptable before we proceed further.

Discussion of the project outline continued, led by PR, up to 4.15.  During
this outline (and the later description by the partners of their work in
the next six months) there were four particular areas of discussion: the
composition of the expert group; the relationship between the TEI and
MSTER; the relationship with the associated partners; technical issues
involving the relationships between the database and SGML strands of the
project and between the server implementations in DMU and OU.
The expert group
PR reported that he had asked Ian Doyle, formerly keeper of manuscripts for
Durham University Library and the doyen of English medieval manuscript
scholars, to act as chair of the expert group.  Dr Doyle has a
long-standing interest in standards issues relating to manuscript studies
and is very alert to the possible use of computers in this context.  The
group should contain two further members from Europe.  Dr Doyle suggested
Gilbert Ouy.  The group discussed briefly other possible names: in
particular, it was felt that a person from the German/Austrian or Italian
communities would be desirable.  Suggestions please!  (Since the meeting,
IRHT has suggested Mme Lucie Fossier in place of Ouy.  PR will discuss this
with Dr Doyle: as chair of the group, he will need to agree its
composition).  Although it had earlier been planned that Ambrogio Piazzone
of the Vatican Library would be a member of this group, his role as
co-chair of the TEI work group will make this inappropriate as the expert
group should be independent of BOTH Master and the TEI.  This matter to be
discussed by email^
The relationship between the TEI and MASTER
Pat Manson expressed concern that the relationship between the TEI and
MASTER was not fully articulated in the Project Plan.  There was discussion
of what is meant by the terms 'partnership' and 'collaboration' between
MASTER and the TEI.  We agreed the following:
^ MASTER will NOT develop a standard distinct from that developed by the
TEI workgroup.  There will be one standard, and one standard only, and the
TEI is finally responsible for issuing that standard
^ The standard which the TEI produces must be usable by and acceptable to
the MASTER partners.  Therefore, it is in our interest that we define
exactly what we need, and communicate this to the TEI.  It is in the TEI's
interest that its standard is accepted by the manuscript community: if it
is not accepted by MASTER then it is unlikely to be accepted by anyone
else.
^ Achievement of this agreement between MASTER and the TEI will need
communication and negotiation.  We need to articulate our needs as quickly
and clearly as possible, convey them to the TEI workgroup, and be prepared
to respond to the TEI's suggestions
^ Accordingly: we should start work immediately on the standards part of
our work, and meet as soon as possible to agree a preliminary draft which
we could pass to the TEI (who plan to have their next meeting in March).
^ We will concentrate in this first phase of MASTER on definition of first
level records.  We will use the fields defined by EAMMS as the base for
discussion of these.
The development of the standard will be the responsibility of a standards
committee in MASTER.  For the time being, this committee will be made up of
ALL the MASTER partners.
A further consideration is that the task of the TEI is NOT just to prepare
a standard to be used by MASTER.  It is to prepare a standard to be used by
all manuscript cataloguers.  There is actually a danger of the relationship
between MASTER and the TEI becoming too close (for example, if the
membership of the two groups were identical), so that TEI loses sight of
the wider constituency it must serve.
The relationship with associate partners
Several groups are already linked with MASTER as associate partners.  These
are: in Europe, The British Library (BL), the Bodleian Library (BOD), the
Stofnun rna Magnssonar  Islandi (SAM), The Wellcome Institute (WI).
Outside the EU - The Vatican Library (VL); Electronic Access to Medieval
Manuscripts (EAMMS).  In addition, a group based at the Biblioteca
Ambrosiana has also expressed a wish to become associates of MASTER.
In the Project Plan, it is specified (in the preface to Part II) that
'Non-funded associated institutions will take part on a voluntary basis in
standards development and testing, but are not committed to making a
minimum number of manuscript descriptions using the MASTER scheme or
contributing any records to the on-line union catalogue'.  In Section 9 of
the Project Plan it is further stated:
In addition to the funded partners named above, other manuscript-holding
bodies inside and outside the EU have expressed an interest in association
with MASTER. As indicated in the preamble to Part II, these bodies will
take part on a voluntary basis in standards development and testing, but
are not committed to making a minimum number of manuscript descriptions
using the MASTER scheme or contributing any records to the on-line union
catalogue (the Demonstrator, WP 7). However, as key manuscript holding
institutions and as likely future users of the MASTER standard and systems,
they form an excellent user group, and will provide especially useful
feedback at the demonstrator stage. They will be invited to attend
appropriate MASTER meetings, may use the MASTER systems as they wish, and
contribute records to the demonstrator. They will also be asked to
contribute to the survey of needs to be prepared as part of the MASTER
exploitation WP.
This commits us to collaboration with associate partners, but leaves open
the nature of this collaboration.  So far as EAMMS (and the TEI) is
concerned, we agree that this collaboration takes the form of admitting an
EAMMS representative to all meetings, and full participation by EAMMS in
all project workpackages and discussions.  This is in recognition of the
unique relationship between MASTER and EAMMS: without EAMMS, indeed, there
would be no MASTER.  Further, the director of EAMMS, Consuelo Dutschke, is
also the leader of the TEI workgroup, and so her presence at MASTER
meetings is doubly valuable.  In effect, we agree that for all practical
purposes EAMMS is a full partner in MASTER.  If it were not for the EU
rules preventing projects based outside the EU being funded, EAMMS would be
a full partner legally as well as practically.
The question of the relationship with other associate partners was
discussed.  At least one associate has expressed a wish to be present at
future meetings.  The question immediately arose: should this and other
associates be present at the next MASTER meeting, which will focus on
standards development.  There was a strong agreement in the group that this
next meeting should be for MASTER partners only (including EAMMS).  This
follows from what was said above: that the best service we can render the
TEI is to define what WE need, and communicate this as clearly and quickly
as we can to the TEI.  We should invite associates to contribute to our
discussions before this next meeting.  But the definition of what we need,
for the records we are now legally bound by contract to make with this
standard, must be the task of the MASTER partners.
This does not preclude attendance by associates at meetings after the next
(scheduled for Paris, Feburary 26-7).  Once a smaller technical committee
is formed, this might meet with representatives of the associates.  In
particular, we should plan to invite associates to the workshop that will
precede the first testing phase (scheduled for September).
Technical issues: database/SGML; servers at DMU/OU
MASTER is committed to pursuing these two twin tracks of development.  The
commitment arises from our belief that a standard, to be credible, must
demonstrate real interoperability by running on different systems.  Thus,
we must be able to express the standard both through database and SGML
implementation, and through mounting the records on different servers.
This will involve close collaboration between the technical partners DMU OU
IRHT, and these three will form a technical committee to ensure this
collaboration.
There was discussion of the extent of the database development.  We are
committed to developing a database system for the making of first level
records.  We are committed to complete interchangeability between the
database and SGML implementations for the first level records: indeed, we
define first level records as being just those records for which we will
support such translation (there is a specific WP addressing this
translation).  There was discussion as to whether we would develop a direct
web interface for manuscript records held in local databases.  We are not
committed to doing this: the server implementations planned in WPs 6 and 7
are designed to be built on SGML format records held on the OU and DMU
servers, not on local databases.  However, we are committed to
investigating a Z39.50 database interface (WP 5.3), which could allow
searching of local databases in addition to the central servers planned in
WPs 6 and 7.  We are aware that several libraries have already stated that
they wish to retain manuscript records locally and have them searched
through the web on the local system: this is the clear experience of EAMMS.
This could be achieved through various architectures, and the
interoperability WP 5 commits us to investigate these, and leaves open how
far we implement any such system.  This will be a matter for further
discussion as the project develops.
Similarly, the relation between the two central servers to be implemented
by the project is not fully defined in the Project Plan.  Will ALL records
be mounted on BOTH servers? To what degree will they offer a common
interface? How deep will the interlinking between the two servers run? What
software will be used by each?  (The exact use to be made by DMU of the IBM
Digital Library software is still to be determined.)
These are matters for the technical committee.  This committee should meet
during the Paris meeting.  The first tasks for this committee will be the
specification of the SGML and database systems (WP tasks 3.1 and 4.1).
This specification should be commenced before the Paris meeting.

4.30-5.45
The last part of the meeting was devoted to specifics: what next?  First,
the partners each described their understanding of what they expected to do
in the next few months, and what resources (people!) they expected to have
to help with this work:
DMU/CTA
PR explained that Adrian Welsh, present at the meeting, would be Project
Officer at DMU.  He will be working around 1.5 days a week on the project.
His immediate responsibility will be to set up the project web site.  This
website will have four different parts:
1. A section open to the whole world.  This will be our public face.
2. A section open to associate partners as well as ourselves, but NOT the
whole world.  This will be used for discussion of the developing standard,
dissemination of software etc used by the project, 'in house' testing of
the prototype servers, etc.
3. A section open only to ourselves.  This will be used to hold copies of
the project handbook, internal project reports and the like, as
appropriate.
4. A section open only to each partner.  This will hold the administrative
forms which have to be completed by each project partner, for each two
month management report and ten month progress report.  At present, PR has
an immense spreadsheet which details the tasks each partner is to do on
each task of each WP.  This will be extracted for each partner, converted
into HTML, and put on each partner's site.  Also, at the BEGINNING of each
two month period we will put up for each partner what resources (according
to the Project Plan) they should be using over the next two months on what
package: ie who should be working how many days on what.
AW will set up the security systems for these: these will probably be
password based.  We are investigating setting up our own List-Serv.  AW
will do this too, should we have one.  AW will also be responsible for day
to day supervision of the two other staff DMU expects to employ on the
project.  One of these will be an administrative assistant, around 2 days a
week, to keep track of the various project paperwork.  We do not expect
this person to start until about two months from now.  The other person
will be a student programmer, who we will recruit through our links with
the Department of Computing Science in DMU.  We have one such person
already working, who will spend occasional days on the project in the next
few months, but we expect the bulk of the work to be done by a new computer
science student, beginning on 1 July and then working full time on the
project.
DMU's immediate tasks, then, will be to set up the web server, set about
recruiting an administrative assistant, and start looking at the fields for
the preliminary standard.
AMI
AMI has been successful in winning a grant from the University of
Copenhagen for cataloguing.  Thus, AMI expects to be able to recruit two
cataloguers, one of whom will be paid by MASTER, and a third person to
supervise the cataloguers and paid (in part or whole?) by MASTER.  A first
task is to translate the entire existing catalogue into English (this task
will not be funded by MASTER).  This translated catalogue will then form
the basis of the MASTER records.  In addition, it is planned to recatalogue
the entire collection in the near future.
OU
OU has a key role in the early work packages.  First, OU must lead the
preparation of the preliminary standard, and so must commence work on this.
Second, OU must prepare SGML tools for the record making WP.  This will
require specification and evaluation of software.  OU has money for a half
time project officer over the whole life of the project.  LB said that he
would like to appoint someone full time, by combining the MASTER work with
another grant.  He is currently awaiting news of grant applications, but
does not expect to know the result of these till Easter.  PR expressed
concern that this could mean OU would not have anyone in place until the
summer.  The key WP 2 relies on both the standard and SGML software being
ready in preliminary form by the summer -- they must be ready before
September, when the workshop preceding the first round of testing is to be
held.  LB thought this could be met, even if no-one is appointed until the
summer.
KB
KB is aiming to make a minimum of 300 manuscript records, and perhaps as
many as 1500, using the MASTER system.  KB is most interested in the SGML
solution.  These manuscript records are to be made in English, and will be
descriptions of manuscripts containing Dutch language from around the world
(? correct me AK!).  KB will contribute to the discussion of the
preliminary standard.
NLP
NLP plan to develop the MASTER records in coordination with their
long-running Memoria Mundi project.  NLP has many catalogues, in many
formats, containing many disparate kinds of information.  The long term aim
must be to link all these together.  NLP will concentrate on use of first
level records to give basic access points to each manuscript description,
and to images of the manuscripts as part of their long-standing
digitization program.  There are some 6500 manuscripts in the NLP, and many
of these have no manuscript description at all.  NLP are interested in
using the database implementation of the standard, to be developed at IRHT.
IRHT
IRHT intends to catalogue the 250 mss in Orleans as the base for their
contribution to MASTER.  In the next months, IRHT will begin work on a
specification for the database implementation of the standard and
contribute to discussion of the standard itself.

The next meeting
It was agreed that the next meeting should be a meeting of all partners,
and should run for two days.  It should cover the following areas:
1. Discussion of the preliminary standard.  We should aim, very soon after
this meeting, to have a document ready for submission to the TEI workgroup
2. Specification of the functionality of the SGML and Database tools to be
used for input of records to the standard.
3. The consortium agreement should be agreed at this meeting, if it has not
already been agreed.
4. A draft of the Project Handbook will be ready by this meeting.  This
handbook will be put on the 'partners only' section of the web site.
As discussed above, there was a strong feeling in the meeting that this
next meeting must be for the MASTER partners only.  Much of the meeting
will be the full group, meeting as the Standards committee and as the
Technical committee.  Smaller groups should then be designated to carry on
the standards and technical committee work, and report to the full
partners.
IRHT offered to host the meeting, in Paris, on 26 and 27 February.

***NOTE NEW EMAILS: peter.robinson@dmu.ac.uk *********
*****************OR: peter.robinson@english.oxford.ac.uk********************

Director, Centre for Technology and the Arts
Faculty of Humanities and Social Sciences
De Montfort University, The Gateway, Leicester LE1 5XY, UK
Phone +44 (0)116 250 6349, fax 257 7250

The Canterbury Tales Project
Centre for Socio-legal Studies, Wolfson College
Linton Road, Oxford OX2 6UD
Phone +44 (0)1865 284245, fax 284221


Digital library website: http://www.dlib.dmu.ac.uk
Canterbury Tales Project Website: http://www.shef.ac.uk/uni/projects/ctp/
Collate website:


