Review History of the WIT Project - Esprit Project 25677

Abstract

This document retraces the history of the review process of the Esprit project 25677, i.e. WIT, granted by the European Commission (EC) DGIII/F7 and INFSO C/3. As one of the reviewers appointed by the EC, I had the privilege to follow the development of the project and to formulate whenever and wherever possible recommendations to contribute to the smooth and high quality execution of the project objectives. The WIT project aims at the development of a low cost and co-operative platform supporting the handling of product models in the furniture and wood industries across the internet and integrating product data with the business logic of the partners. This will be achieved through the development and deployment of co-operative and Internet based tools to facilitate knowledge sharing across all business entities of the wood/ furniture sector. This paper is made of several documents, i.e. three major review meetings, the last and final happening in Athens on the 22 nd Sept. 2000, the previous ones in Brussels on the 19 th November 1999 and 4 th Nov. 1998 and of one document produced by the consortium, dealing with the relationship between WIT and the STEP standard. Each individual review follows the same structure, i.e. Conformity of Work to the Technical Annex, Project Management and Co-ordination, Relationship with other relevant projects, Activities related to standards, Plans for Industrial Exploitation of Results, Update of Synopsis, Synthesis of Reviewers' Technical Comments. All documents compiled in this Review History of the WIT Project had a public status and can therefore be freely published or used as long as proper quotes are made. They should serve new consortiums gathering together to better understand how the review processes were conducted and to be better prepared for such events.

Full text

Review History of the WIT Project
Esprit project 25677
Patrice POYET
CSTB, BP 209, 06904 Sophia-Antipolis, France
Reviewer on behalf of the European Commission
Abstract
This document retraces the history of the review process of the Esprit project 25677, i.e. WIT, granted by the
European Commission (EC) DGIII/F7 and INFSO C/3. As one of the reviewers appointed by the EC, I had the
privilege to follow the development of the project and to formulate whenever and wherever possible
recommendations to contribute to the smooth and high quality execution of the project objectives. The WIT
project aims at the development of a low cost and co-operative platform supporting the handling of product
models in the furniture and wood industries across the internet and integrating product data with the business
logic of the partners. This will be achieved through the development and deployment of co-operative and Internet
based tools to facilitate knowledge sharing across all business entities of the wood/ furniture sector. This paper is
made of several documents, i.e. three major review meetings, the last and final happening in Athens on the 22nd
Sept. 2000, the previous ones in Brussels on the 19th November 1999 and 4th Nov. 1998 and of one document
produced by the consortium, dealing with the relationship between WIT and the STEP standard. Each individual
review follows the same structure, i.e. Conformity of Work to the Technical Annex, Project Management and Co-
ordination, Relationship with other relevant projects, Activities related to standards, Plans for Industrial
Exploitation of Results, Update of Synopsis, Synthesis of Reviewers’ Technical Comments. All documents
compiled in this Review History of the WIT Project had a public status and can therefore be freely published or
used as long as proper quotes are made. They should serve new consortiums gathering together to better
understand how the review processes were conducted and to be better prepared for such events.
Review on the 22nd September 2000, Athens
Introduction
It is reminded that the WIT project aims at the development of a low cost and co-operative platform supporting
the handling of product models in the furniture and wood industries across the internet and integrating product
data with the business logic of the partners. Costas Paleologos acting on behalf of Ronald Mackay as project
officer, requested that an overview of the previous recommendations be made and that the consortium would
comment on actions taken on these. It is reminded that two major recommendations had to be addressed. One
dealt with the implementation plans, addressing issues like the user guide(s) to help people through the use of the
system, ease of deployment, etc. whereas the other, the strongest point made during the previous review, was
about the exploitation issue, including steps taken to formalise the IPR issue(s), market analysis and
segmentation, etc. Nikos Sakkas says that we should go through the proposed agenda designed to answer most of
these questions, but suggests as well making a brief presentation of the status on these matters. As far as the first
point is concerned, it is acknowledged that the documentation is still difficult to use, that the user guide has been
reduced to a 25 pages document which describes the functionality of the client tool (which should be connected
to the F1 key), and the project leader does not believe that the situation is entirely adequate, but stresses that
efforts have been made towards the development of a more sophisticated help system. With regards to the ease of
the communication between the server and the client, Nikos Sakkas does not see a point there, and believes that
this must be pretty invisible to the user. With regards to the second and major recommendation made during the

previous review, the exploitation plan, Nikos Sakkas acknowledges that it has not been delivered timely, but
asserts that the work is available today. The partners are expected to present their plans during the review day, it
is claimed that some synergies have been identified, but the exploitation will not happen at the whole consortium
level.
Conformity of Work to the Technical Annex
The WIT project aims at the development of a low cost and co-operative platform supporting the handling of
product models in the furniture and wood industries across the internet and integrating product data with the
business logic of the partners.
Overall, the project is in line with the aforementioned objective and considered by the Project Leader (Planet) as
having been completed at 90% on a technical basis (i.e. one of the seven applications envisaged has not been
terminated), the WIT server is reported implemented, only minor parts of the clients have not been finished,
content has been developed to a satisfactory stage (graphics, models, etc.) for the two user companies involved
(i.e. Shelman, Resoflex).
The deliverables to be expected at this stage were the following:
1. D 2.3 / D3.2 WIT Software (implementation) ;
2. D 4.1 WIT Evaluation (which has been undertaken under different angles)
3. D 5.2 WIT Dissemination (combines different aspects, including the scientific communications to
conferences and magazines, plus the leaflet) ;
4. D 5.1 WIT Exploitation (claimed to include mostly individual exploitation plans of at least some of the
partners of the consortium, but not available for the review day).
For each of these deliverables, an assessment has been made. The review report will successively address aspects
linked to the WIT implementation, the WIT evaluation. The WIT dissemination and exploitation plans are being
covered in a separate section of the report, further to the receiving of additional material sent by the consortium
parties after the review.
WIT implementation
The WIT implementation covers the server implementation (with the tools deployment and related activities), the
client implementation, the web implementation, related demonstrations, and miscellaneous matters. It is reminded
that the WIT architecture supposes that the user has to deal with some piece of client software to resort to a WIT
service with proper product databases. The WIT client accesses the WIT server to go and search for products.
The WIT-N server knows about the WIT servers around and knows about the WIT clients requesting services
and matches requests and services available. Contact from the web browser is established with the WIT-N server
(a WIT client is downloaded), the WIT client is more a Business to Business type of application. Any product
structure from a supplier can be used and can be mapped to the data model being used in the WIT architecture.
The project has focused its efforts on two suppliers (i.e. Shelman, Resoflex).
The presentation of the Server implementation covers the two implementation strategies followed by the project,
i.e. JDBC/SQL and STEP based and illustrates the work reported in the corresponding deliverable. Two server
implementations have been made, one using a JDBC server with a connection to a database server (MS SQL,
plus a directory server and a web server in Greece, i.e. WIT-N) and a STEP version in Norway plus a ftp server
(graphics) in Germany. A presentation of the major programming classes made available by the software
implementation is undertaken, the four major layers being the "Remote classes", “Net classes", "User interfaces",
and "Client data classes".
As far as EPM's implementation of the WIT server is concerned, i.e. a STEP based implementation resorting to
an early Java binding + SDAI + dictionary models (WIT server data model), EMP report that they have focused
on providing performance on the server side. EPM mention that a specific set of functionality has been developed
including (query and quotation mechanisms, order processing, collaboration management, data integration, etc.)
and that the database management resorts to dedicated WIT objects with particular properties, etc. It is

questioned by the review team whether the corresponding EXPRESS data model underlying EPM's
implementation could be made publicly available. The answer from EPM is positive and further to the review, it
is agreed that the models should be loaded on the WIT web site.
The demonstration of the Client implementation addresses the "Project manager" (enables the creation and the
description of a project), the "Suppliers" (provides access to the product manufacturers in a timely manner
through a tree-like representation), the "Product query" (the user gives the characteristics of the products he is
interested in and gets a list of the available products) - in fact two access modes are offered, i.e. "list-view" and
an "icon-view"). Further capabilities include "My catalogue" (to keep records of "nice" items, sort of "favourites"
that would be kept on your own disk), "Space planner" (sketch and handling of furnished rooms), "Quotation and
order" (this is where you do the business; ask for a detailed offer, etc.) "Contact" (keeps track of an address
book). From what is demonstrated, the Client goes beyond what was sketched during the previous review and
provides a convincing picture of what an end-user will get access to. Furthermore, questions which had been
raised during the previous review with respect to the perceived necessity to have developed an "ad-hoc" 3D
viewer starting from the Java 3D environment are now answered by the very convincing demonstration of the
"Space Planer" providing for a satisfactory editing and selecting of rooms, furniture, avatars, etc. The tool is well
integrated and convincing.
The quotation and order processes are then demonstrated. Nikos Sakkas mentions that the identification
mechanisms used throughout these processes are still pretty much rudimentary. The review team insists on the
importance of the business model at that stage and does not see the technology as a major bottleneck. Questions
such as: are we dealing with virtual market places?, who is going to be allowed to connect ?, e.g. trusted resellers
connecting to the final product manufacturers, final end-users, others ? appear to be more important to be
addressed than the specifics of the authentication mechanisms. From there, discussion between the review team
and the project about virtual market places arise such as: how to integrate the WIT Client with the new ways of
making business allowed by virtual places?, etc.. Of course in a sense it's going to be company specific but at the
same time virtual market places will cross companies' boundaries and will require that each of the company think
of how they are going to integrate and co-ordinate old and new ways of making business (from quotations down
to product management and shipment, etc.) and also think of the way they are going to maximise the business
they can derive from new ways of interacting.
End users argue that the 3D display (pretty much independent of the platform) of the furniture and the integration
with the product manufacturer’s databases is most worthwhile. Shelman is not an end-product company (semi-
finished components) and would like to see the project continue. Linking to the companies' PPS or ERP remains
and open question and often companies have to go through a significant business re-engineering to accommodate
these new technologies.
Finally, IPR protection strategies through watermarking of 3D models are presented. At least two major
objectives are reported, i.e. general labelling trying to enforce the property rights attached to the images and
mastering better collaborative environments / processes by means of dedicated watermarks identifying the
process stages and knowing where the "objects" (and their representations) stand in this process. Industrial issues
are mentioned, including a possible agreement defining licensing between Fraunhofer-IGD and MediaSec. The
review team suggests that a workshop be scheduled with the end-users to check the kind of usage / requirements /
expectations they have with respect to the technology. The benefits to the WIT project can/must be maximised by
not limiting this to a sole technological effort.
Costas Paleologos asks about the system requirements (i.e. hardware). It is answered that a good basic PC
system, with Open GL, plus Java 3D extensions of the standard Java environment are required. Overall, the
review team congratulates the consortium for the progress made with respect to the quality of the software
implementation and integration achieved.
WIT evaluation
The consortium has engaged in a broad range of assessment procedures. Worth being listed, the following
evaluation means have been used by the consortium:
1. Self-assessment ;
2. Graphics evaluation ;

3. User request(s) evaluation ;
4. Third party assessment ;
5. Assessment of the tools deployed.
Overall it can be thought that the partners have tried to make a fair assessment of where the project stands.
Before going into more details, the initial assumptions underlying the assessment can be reminded as:
1. Being business focused ;
2. Making use of data modelling techniques for data and knowledge representation and management (e.g.
STEP and SQL) and ;
3. Resorting to graphics for the man/machine interface.
The major conclusions appearing from the documents available and from the presentations made at the review
are the following:
A) Consortium / Self-assessment
It is observed that no software packaging mechanism is yet available. The consortium seems to be engaged on
setting up improved procedures as appears from actions taken with respect to exploitation.
B) Graphics evaluation
Java-3D appears in the end as a wise choice and the development has been successful. The cost of 3D models
with limited reuse of 3D data has been questioned but the interest for the user is pretty compelling and supports
attractive (man/machine) interaction mechanisms. It is mentioned that 2D textures have been hand crafted, are
small, seamless, and equally scaled. Nevertheless, the parties in charge of this matter remind that graphic
authoring requires professional support. It is unclear what further steps the consortium will take with respect to
the involvement of professionals in this area.
C) User request(s) evaluation
It is reported that steps to perform "User request(s) evaluation" had been implemented but will not be
demonstrated during the review day. Outcomes from these have been taken into account. Most of the users'
comments were related to the incompleteness of the database. Nevertheless, and unfortunate as it is, some cannot
be alleviated due to system requirements and constraints (256 MO). The consortium argues that they cannot be
taken into account as these were not included into the requirements
D) Third party assessment
Issues like internationalisation, friendliness, standards, internet graphics, etc. have been addressed under the third
party assessment.
E) Assessment of the tools
Among the various issues related to the assessment of the tools, logistics and security is considered too simple,
and the password based approach not even covering the WIT-N problematic. Authentication by certificates
issued from trusted organisations is still to be addressed. Security by encrypting the Rmi calls is under
consideration.
Project Management and Co-ordination
Considering the material available and the presentation given during the review day, it can be assessed that the
project management function has been properly performed by the co-ordinating partner even though a number of
deviations have occurred.
Among the most noticeable deviations, it is worth mentioning the delayed production of the leaflet and the
postponement of a workshop to be scheduled before the review meeting, which was not considered possible due
to the non-availability of the marketing material at least 45 days before the meeting. Furthermore, minor parts of
the client application have been left in a standstill and development not followed up due to the departure of some
of the main software developers (from the Fraunhofer institute as understood).
Furthermore, it is suggested that the IPR clause addressing the availability of the code be clarified (eventually
including end-users).

Activities related to standards
No activities were reported
Plans for Industrial Exploitation of Results
As of the day of the review, the exploitation strategies have been limited to individual statements made by some
of the companies. Presentations made have covered, though partially, the following items:
1. Exploitation framework ;
2. Partners plans ;
3. IPRs;
4. Publications.
Before drawing a synthesis of the situation as perceived by the review team and mentioning the requests made
during the review day as follow up actions to be undertaken by the consortium to clarify the exploitation
strategies, it is worth mentioning the situation as presented by each partner with respect to their involvement.
Planet exploitation
Planet considers WIT as a delivery of a value adding service (VAS) that is enabled by means of a software
toolkit. The major efforts toward setting up such a VAS can be listed - as understood by Planet - as:
1. Structuring and re-using data ;
2. Capturing and modelling graphical data ;
3. Setting up and populating a WIT database ;
4. Setting up a WIT server environment ;
5. Enabling the WIT-n level.
Planet report that they do not wish to be involved in:
1. the service part of WIT, especially in graphics related issues, quoting ZGDV as the best possible partner
for this matter ;
2. Maintaining WIT code, e.g. participating in major code upgrades.
Planet identifies two basic exploitation modalities and these are discussed in terms of anticipated advantages and
disadvantages, i.e.:
1. Company focused delivery (personalised WIT delivery, esthetical conformance / integration with
existing company's sites, etc.);
2. Sector wide delivery, i.e. "deep" furniture portals.
Planet indicate that a possible objective would be the set-up of a WIT based furniture portal in Greece (sector
wide but regional approach) in the next two years attracting 40/50 manufacturers and trading companies, to be
further extended to other places in the world. Building the right alliance is seen as critical. Planet / Ernst &Young
are consulting organisations, Shelman is a leader in Greece in the furniture industry, but Planet considers that a
major software partner is still missing.
ZGVD exploitation
ZGVD remarks that the German furniture IT market is crowded. Different technical approaches (e.g. CAD
driven, marketing driven, etc.) and solutions for different sub-markets (e.g. office, kitchen, etc.) have emerged.
Nevertheless, ZGVD consider that there are good chances for:
1. Customised editions of the WIT client tools (e.g. catalogue oriented, etc.);
2. Browser based solutions (e.g. small applets, no software download, on-line shopping) ;
3. Add-on components for catalogue shops (e.g. large distributors, CD-ROM).

ZGVD perceives that the most promising approach is to come up with a small plug-in, offering a limited service
atop of published and public electronic catalogues.
EPM tech
EPM remind us that they offer STEP data warehouses for decision design, support and operation. EPM mention
that they will focus on the building sector and will pay attention to the IAI and TC59/SC13/WG6 (framework for
object oriented information exchange). Among the functionality of interest are:
1. Interfaces to libraries for building design, building services and facility management ;
2. ERP and e-procurement solutions ;
3. Other catalogue projects.
Emphasis is placed on data management and little is said by EPM with respect to the particular application area
addressed by the WIT project, i.e. the furniture industry.
Further collective contributions
These individual plans have left during the review the feeling of a somewhat uncoordinated perspective on
exploitation. Furthermore, the review team has discussed and asked whether these exploitation frames miss the
possibility of creating virtual market places of benefit to a number of companies, resellers, customers, etc. The
discussion ensuing lead to request that additional thoughts be put into devising how the various partners could
pull together to come up with a more co-ordinated approach to exploitation. From there, the review team has
received additional documents on the 6th of October (as requested per the Project Officer) which demonstrate that
a much thorough analysis has been performed, both at the level of the individual companies involved and more
interestingly by groups of partners.
The executive summary and business plan of WIT-N.GR, was prepared by Planet- E&Y S.A. in collaboration
with ZGDV. It is identified that <<the long term business target is the set- up and operation of sector focused
portals that may support a “many to many” communication profile and enable accessing, 3D viewing, using and
eventually purchasing products from any number of enabled suppliers. This five year plan represents a first
instance of this spectrum. It targets the furniture manufacturing and trading sector. The core toolset, called WIT-
N (Wood & Furniture Information Technologies Network), is now in its beta version, and has by design
integrated both “many to many” communications and rich data delivery and management, including advanced 3D
graphics. >> Furthermore, << the capital required to initiate operations is estimated to approximately 500 K
Euro. This includes operational expenses of the first year as well as some working capital. Potential investors
have been identified and could commit capital by holding shares. The equity capital of the new company should
be around 500 K Euro. The exact participation of Planet – Ernst & Young S.A in this capital as well as the
eventual participation of ZGDV in it is a matter to soon decide.
The business plan for WIT-N Portugal will be implemented by a new company being created by Alfamicro. The
equity of this company will be shared by the main developers of WIT. Negotiations have started with ZGDV
(Rainer Malkewitz), with Planet – Ernst & Young (Nikos Sakkas) and with the Alfamicro developers. Alfamicro
report that they << already reached an agreement with Rainer Maltewitz and with two of the Alfamicro
developers (Luis Viseu and Luis Reis). Nikos Sakkas from Planet – Ernst & Young has also given his agreement
of principle >>. Alfamicro will carry out discussions with venture capitalists in order to ensure the initial
investment required by the project. The minimum required capital is estimated in the document released by
Alfamicro according to various business projections and is evaluated so as to cover first year (2001) expenses
(934 K Euro). As a whole, and including the working capital, the required venture capital is estimated to about
1.100 K Euro.
The end-users have equally produced exploitation perspectives and positioned these with respect to their current
business models. Shelman report that as a raw material rather than a furniture manufacturer the 3D representation
capabilities of WIT and the related tools (such as the WIT Space Planner) can have only a limited direct use for
Shelman. They can however have a significant impact, if used by clients of Shelman. Exploitation of WIT
comprises the following activities:

1. A thorough take up of the existing WIT Server/ semi-finished elements within the company itself.
2. An internal investigation into the possibility to provide extensions or new WIT servers for some or all of
the rest of the Shelman factories and product ranges. Decisions on this should depend on the overall
evaluation of the existing WIT Server especially from the marketing people involved. They should be
reached within 2001.
3. A value chain level investigation into the possibility to provide dual type of WIT Servers, merging
products from various companies. This opportunity is proposed to the consortium as something to
seriously look in and also as something that should perhaps be stressed as a possibility during the WIT
workshop event, whenever this takes place.
On the contrary, Resoflex is positioned in the complete cycle of raw materials, manufacturing, sales and customer
services. They sell their furniture products as an “extended product” which goes up to the point of supplying the
product with a team of local workers which may perform any tasks that are associated with customer service.
Resoflex see WIT as a product capable to help the visualisation across the supply chain and helping the different
users to understand in real time product views, availability, costs, etc. Resoflex intends to exploit WIT once the
prototype becomes a product.
Synthesis of Reviewers’ Technical Comments
The review team has congratulated the consortium for the work done since the previous review. Challenging
concepts and systems have been developed. The partners are urged to devise means to capitalise further on their
investments. Technical work is considered satisfactory, the review team appreciate the management effort,
however a need arise to strengthen the exploitation strategy, look at how the documents will be improved (cost,
revenue, layers, who does what, etc.) having a good picture of the environment and each consortium partner to
have a positioning in the exploitation picture (who pays what and gets what).
From there, the review team has expressed the following expectations:
1. First draft of the updated exploitation plan by 6th Oct ;
2. Package of the software by 6th Oct ;
3. Project manager must ensure that commission receive the last progress report by the 15th Oct ;
4. Final report, to be written after the completion of the project, within a month of the completion of the
project.
At the time of this writing (15th Oct.), the first two and major requests have been satisfied. The Project Officer
will have the duty of ensuring that the two remaining actions be closed on time.
Review on 11th November 99, Brussels
Introduction
It is reminded that the WIT project aims at the development of a low cost and co-operative platform supporting
the handling of product models in the furniture and wood industries across the internet and integrating product
data with the business logic of the partners. Ronald Mackay reminds the project that the review time should be
considered as a means to get advice and support from the review team, knowing that the project "belongs" to the
companies involved, and the Commission expects that the work is being done for the contractors' benefit.
Since the last review a lot has happened and the objective of the meeting are mainly to assess whether the
Consortium is on track, whether it is likely to achieve the results promised, with emphasis on exploitation. The
agenda of the review is presented by Nikos Sakas including: Project Overview, Implementation related issues,
Exploitation related issues and final Discussion. It is mentioned that some deviations have appeared since last
review, including a report on prototypes failure and request for extending the implementation task from month 18

to month 23. Of course, this leads to question to what degree WIT has been delivering the value it has been
claiming and with which contingency plans for what has not been done.
Implementation is reported complete at around 70 to 80%. It is though acknowledged that reaching 100% may
take a lot in SW development. Software integration needs to be advanced and gets now a high priority with
possible completion in 3 to 4 months. This might be optimistic and contingency plans are to be suggested.
Difficulties are reported and are arising from the diversity of teams with integration overhead certainly bigger
than what was anticipated by the Consortium. A better way of working should be found. The review team asks
about the evaluation of the current completion and suggests that the difficulties encountered should be reported in
a "lesson learnt" sort of document.
As far as the deliverables are concerned, it is reported that D2.3/D3.2 - concerning software development - are
still due, that D3.1 covers the Implementation Plan (at the users side), D3.3 should be considered as a
documentation manual (i.e. a user guide as how to use the client tool, plus a technical guide strictly speaking) and
finally D5.1 is the Marketing Plan (whereby partners plans are being elaborated but the global strategy is still
lacking). It is suggested to pay attention to due dates as far as the deliverables are concerned and to inform the
review team of any further delays that might happen.
Deviations are mentioned, and the project is running into a small delay, reported to be between 4 to 5 months,
and the project manager reports that the delay is not expected to increase further. The project officer stresses that
delaying some of the project's results might not be a major problem as long as the overall budget is respected,
given that no extra funds will be provided by the European Commission. Slight modifications to the project plan
might be necessary and a greater "level of detail" is required for the period ahead.
The way ahead for the consortium is summarised, and the following actions are mentioned:
1. Continuous software amendments (IGD)
2. Training of users (ZGDV)
3. Product leaflet (PL)
4. IPR agreement (EPM)
5. Commercial fair participation (AM) Milestone where the value is visible
6. Refinement of exploitation plans (PL)
As a summary of the current situation, it is felt that good and novel concepts are being built, that implementation
is on track but that a more professional outlook should be targeted, that participation into an international
exhibition should be used to display the value of the project and that exploitation plans should be devised at the
consortium level, but at the same time not threatening the flexibility of the partners, by balancing views.
Conformity of Work to the Technical Annex
The 3 business layers of the WIT software architecture are now well advanced, i.e. WIT-N, WIT server and WIT
client. WIT-N to WIT client has been allocated to only one partner and has reached a satisfactory stage of
completion whereas the production of contents to be delivered by the WIT server to the WIT client(s) is still
lagging behind expectations. An efficient and easy communication between the client and the server is still not
reached.
The WIT architecture supposes that the user has to deal with some piece of client software to resort to a WIT
service with proper product databases. The WIT client accesses the WIT server to go and search for products.
The WIT-N server knows about the WIT servers around and knows about the WIT clients requesting services
and matches requests and services available. Contact from the web browser is established with the WIT-N server
(a WIT client is downloaded), the WIT client is more a Business to Business type of application. Any product
structure from a supplier can be used and can be mapped to the data model being used in the WIT architecture.
The WIT client's parts have now reached a good level of description and completion, including the "Project
manager" (create change, delete), "Suppliers" (search through the WIT-N), "Product query" (search by supplier
with the query sent to the whole product catalogue to get an excerpt), "My catalogue" (to keep records of "nice"
items, sort of "favourites" that would be kept on your own disk), "Space planner" (sketch and handling of
furnished rooms), "Quotation and order" (this is where you do the business, ask for a detailed offer, etc.)
"Contact" (keeps track of an address book). A demonstration of the WIT client facility is now available.

Installshield has been used to facilitate the installation of the downloaded WIT client (checking for the existence
of a Java virtual machine on the system), etc.. Then the WIT client can be launched and opened.
"Project manager" - starts with the project manager which gathers sort of general information, and contains a list
of rooms, for each notes can be taken …
"Suppliers" - based on the LDAP technology, with some interest external to the consortium expressed in the way
this technology can be used to manage sort of yellow page information (normally used to handle net resources).
A WIT-N directory is available with Portugal and Greece regions with particular companies (e.g. Resoflex,
Shelman) etc. These are "configuration data", a browser is created with organisation level and downwards
includes the parts level. Then from the browser the user can go and search for server contents, including groups,
families and types (answer is where the user should go in terms of country and company). Search for server(s) is
based on (product groups, families and types) or text or project requirements.
"Product query" - (supplier, use type, product type, item type, characteristics, selected values) this is hosted on
the WIT-N server and is downloaded through configuration files when the server is contacted.
"My catalogue" - The local catalogue is populated over time, type of use (e.g. office furniture), with various
suppliers and then according to (product line, 3D model, item). For each 3D model displayed, the user may have
various items with different properties such as colour, etc.
"Space planner" - From what has been demonstrated, the review team believes that a lot of work remains to be
done with respect to this component. From what was presented, it is noted that the systems always does snapping
and does not appear to be a real 2D editor, it has no satisfactory 3D display, seems to be bugged (Java
exceptions), and one can question the reason to redevelop this kind of tool and why not reusing an available
library or plug-in from a software vendor.
"Quotation and order" - Rooms in the project are displayed on the left hand side of the user interface, on the right
side is displayed the furniture in the room, plus quantities of the product(s), and product(s) origin can be
highlighted by some selection mechanism, queries can be made (such as estimated price), then an order can be
generated. It is noted that the process looks a bit remote from the way people work today and does not study how
a migration process from the current way of working to a new way of working could be made. It is suggested that
the electronic commerce directive be studied, as conducting business requires a clear identification of the
contracting partners (where are the partners located) including VAT regime, etc. It is also suggested that it be
clarified when does this transaction become a legally binding contract, as sometimes massive loss of credibility
incurs from refusing to honour orders. Various procedural aspects must be clarified, including when does the
confirmation of the order happen, how is the degree of credit to be checked with information attached to known
client, etc.
"Contact" - last panel dealing with conference partner and conference date, etc.. The demonstration is there not
convincing in any respect (combo boxes not working, etc).
The structure of the WIT-N server has now progressed enough to be presented and is organised in a layered
manner (4) with two WIT-N servers proposed, one from PL the other from EPM. The four major layers contain
the "Remote classes", “Net classes", "User interfaces", "Client data classes". The WIT server is available at
http://wit.planet.gr/witserver/ and http://wit.planet.gr/witserver/donwload. On the server side, three strong points
have to be stressed, among which the mapping of structure(s) existing into different companies with the structure
supported by WIT (product groups, families and types are the basic elements of the WIT structure), and the way
the system relates the 3D model with the real product by means of "parametrisation" mechanisms (including
colours, finishes, etc). Entities are grouped into four sets: "Catalogue and characteristics", "Models and
visualisations", "Company and user profile", "Requests logs".
The WIT server from EPM is presented including:
Overview: WIT client has to integrate a WIT server interface to communicate through RMI with the WIT server.
Th WIT database interface can be implemented by various means (SQL implementation + JDBC+SQL database)
or STEP implementation + Java early binding + SDAI + dictionary models (WIT server data model) and
Resoflex Data Model (picture)
Functionality: query, quotation, order, collaboration (arrange net-meetings)

Example
Finally, watermarks in 3D geometry have been implemented (3D Studio Max). Grouping of "entities" is
preserved by the process. The review team suggests that it is hard to understand what this brings to the project
and in what it differs from the standard usage of an off-the-shelf tool. Questions of positioning of these activities
in this project are asked and answers claim that even though the images might be public one might wish to know
from where they come. Further clarification would certainly be welcome.
As far as the deliverables are concerned, the following were received ahead of the review meeting:
- the Fourth Periodic Progress report (Mar 99 – Sep 99)
- D 3.1 (Pilot Trials – Implementation Plan)
- D 3.3 (WIT Documentation Manual)
- D 5.1 (WIT Exploitation strategy)
All three deliverables were well presented and of good quality. They should be accepted as they stand.
Project Management and Co-ordination
The material presented in the Periodic Progress report suggests that the project has been well run at least in an
administrative sense. However, since the last review 12 months ago, it is clear that certain deviations have
occurred in the project schedules, which have brought about slippage in relation to the overall project plan. On
the surface, this would appear to contradict the assertion on p6 that ‘from a management point of view, the
project has run smoothly.’
Activities related to standards
No activities were reported
Plans for Industrial Exploitation of Results
Discussion on exploitation has happened with external bodies, looking for partners who can provide some added
value to the partners' products. Such companies have been approached, and on-going discussions are happening.
The project manager reports that WIT does not have a software vendor in the consortium providing for an easy
path to the market for the WIT results. This leads to the fact that alliances might be necessary. IPR is not felt as
just a legal framework; there must be some major restructuring to any standard IPR document. It is suggested that
a clear picture of what belongs to who be gained and reported in a module by module approach.
From what is being presented, the exploitation strategies are considered very fuzzy by the review team and the
corresponding economic model for running a WIT business is pretty unclear. It is suggested to come up with a
document in a very soon future to further detail plans in that respect.
Furthermore, Deliverable 5.1 raises a number of issues and questions which are summarised below:
1. The lack of an agreed IPR policy at this stage of the project is a cause for serious concern. Immediate
steps should be taken by the partners to overcome this fundamental difficulty
2. Likewise, the fact that there exist no terms for co-operation as yet is a fundamental omission in the
exploitation strategy. In order to ensure the success in the market place of the final WIT product,
strenuous efforts should be made by the consortium to reach agreement on these issues
3. This reviewer is keen to establish how the economic case for the exploitation of the product has been
affected by the rapid technological advancements that the world has seen even during the lifetime of the
project. This refers to, in particular the use of the internet as an enabling technology in the supply chain.
How has this affected how the consortium will differentiate their product in the marketplace next year,
as opposed to when the product was conceived? In short, how have the market dynamics impacted on
the potential success of the exploitation strategy?

4. The issue of the internationalisation of WIT should be addressed and planned for. The conclusion
offered in the text of D 5.1 that: ‘ WIT-N can address a significant global market by offering good value
for a low price’ is not substantiated in the report. In fact, the evidence presented in the report seems
contradictory on this point. Firstly, the pricing policy as outlined in D 5.1 seems heavy for a product of
this nature (how many SMEs will be able to afford the amounts quoted for a product that is not yet a
proven standard?). Secondly, the market study undertaken is, at best, thin as it is based only on data
from Greece and Portugal. Thirdly, if WIT- N is as is stated on p7 ‘..global by design’, then it is
contradictory to state that the product could ‘…..suffer serious market reach limitations because of the
use of only the English language.’
Clearly, the consortium requires to address the issue of market research, market reach and pricing policy, with a
view to achieving an integrated approach to the issue of exploitation, which taken as a whole seems to be
uncoordinated at present.
Update of Synopsis
Not necessary.
Plans for Dissemination of Results/Web site
A list of activities and publications would be welcome.
Synthesis of Reviewers’ Technical Comments
It is felt by the review team that:
The project has made progress;
The technical solutions taken are somehow supported;
The project is following its work-plan reasonably well, except for some delays mainly coming from integration
issues.
A major concern is the unclear route to exploitation: The consortium has to work rapidly and to define exactly:
1. the future offer(s): A product, services or a mixture;
2. the partners involved and their exact roles;
3. the markets to be addressed;
4. the need for additional partners in marketing and software support;
5. and as a result the detailed plan including participation in fairs needs to be developed
Furthermore, the ‘owner’ of the code and responsibilities regarding updates, future language versions etc. have to
be precisely defined. Unclear are also currently several legal questions. One of the reviewers offers help
regarding the organisational and legal issues with respect to close contacts with the VIVE project which is
addressing the creation, operational aspects and shutdown of Virtual Enterprises.
The demonstrations were quite interesting and the review team looks forward to seeing WIT in 'full' action and
operated by a real commercial user. Thus it is strongly recommended holding the final review at Shelman's as
originally planned.
Nevertheless, the consortium has clearly not made the degree of progress during the last year which was
anticipated at the time of the second review meeting. Whether this is due to difficulties within the consortium, or
because unforeseen technical difficulties have been encountered is difficult to accurately judge from the
deliverables alone. Whatever the reasons, it remains clear that the success of the WIT project is dependent on all
members of the consortium adopting a highly focussed and truly integrated approach to the achievement of the
objectives set out in the revised project plan issued in the fourth periodic progress report. All members of the

consortium are encouraged by this reviewer to address the issues highlighted above and to maximise their efforts
towards the success of the project in this final period.
Review on 4th Nov. 1998, Brussels.
Introduction
The objectives of the WIT project have been confirmed during the second review meeting as aiming at the
development of a low cost and co-operative platform supporting the handling of product models in the furniture
and wood industries across the internet and integrating product data with the business logic of the partners. In
that respect, the WIT project proposes to cover a seldom addressed issue, i.e. the business relevance of the
technologies starting from day one of the project, this being dealt with by the identification of the corresponding
strategic framework. Some deviations have appeared since last review, including a report on prototypes failure
and request for extending the implementation task from month 18 to month 23.
Progress assessment by workpackages
Deliverables being due at that time have produced, this includes PPR, D1.3., D 2.1., D 2.2. All due deliverables
have been produced and released on time and demonstrate good quality. They are well informed and the work
done is in line with the technical annex. They also demonstrate that good co-operation is happening for most of
the consortium. In that respect, it is suggested that the four deliverables be accepted. It is recommended that
glossaries be added to the documents to help grasp abbreviations.
Well documented presentations were given during the review day and are summarised hereafter:
Overview of the project was given by PLANET.
The following key points have been reported:
An unambiguous understanding of the user requirements has been gained. This is presented as a prerequisite to
the exploitation path. The consortium faces rapidly changing environments and technologies (e.g. JAVA, etc.),
but this is perceived as keeping the partners close to the market. It is reported that the project must start on
implementation at the time we are now. There are still design issues pending but the project should start
embarking on implementation and furthermore think about the prerequisites about installing the system.
Some problems have been reported by the project management including the fact that the project has been unable
to define a vision with respect to implementing STEP in the WIT project. The consortium has tried to find a way
to position STEP in the project, but beyond demonstrating goodwill this did not advance much further in a very
useful manner. This is perceived as a threat to the progress of the project. The management reports that the
project should cope with STEP in a way that does not obstruct the core development and see whether one could
find a means to make the system compliant with a STEP AP in the future would one exist.
Short overview reporting of deliverable D1.3 has been made including a description of the Business scenarios, a
first strategy for STEP, and an understanding the project genericity Short overview reporting of deliverable D 2.1
has been made including a description IDEF0 process models, WIT functionality tables, WIT registration.
Results comprise a descriptive presentation of a three layer business platform Short overview reporting of
deliverable D 2.2. has been made including a description of the WIT system design. Work done covers extensive
UML modelling and lead to produce the class diagrams which are the “most” significant view of the software
system. Also given is a preliminary version of a UI implementation. Claims are made for readiness for full scale
implementation
Summary of work which has taken place in WP5 has also been reported, including two web sites (client / server
registration), various papers (International Journal of Information Management):

1. Sector-Nets and Hyper Chain management at the ESPRIT IiM Conference Goeteburg;
2. The rise of the Hyper Chain at EMMS, Bordeaux 98, etc.
Furthermore, business contacts have been established in Germany, Greece and Portugal. Contacts made with
ESCN. Moreover, T5.1 covered a market watch where a monitoring of market developments lead to take notice
of various events including a closure of Cosmo (setback for VRML), Space planning tools – EAI versus Java 3D,
to recommend a shift from Microsoft tools to Symantec Café and not use MS J++ any longer (re-orientation), to
envisage the usage of Internet Directories and LDAP (MS Active Directory has not yet appeared on the market,
the only product is the one from Netscape).
A progress report was also delivered. PPR1 stated goal is to finalise a concept for a data model and a plan for
STEP. Difficulties were expressed for converging for a common data model for all user companies (required for
a long term exploitation of the project results) The project reported that it succeeded in designing the WIT S/W
platform, but that it failed in the work out some early prototype and demonstration of core functionality. Question
of why prototypes failed was briefly addressed and it was reported that the project had proved unable:
1. to define and deliver a “core functionality”
2. to set up and populate local database (including graphical information)
3. set up web access to this database and demonstrate core processes (sending / receiving queries)
Prototypes are now considered as having absolute user priority, they are within the consortium capacity and are
considered as necessary project requirements (for soundly based project exploitation)
Suggested modifications arise from the fact that even though resources are adequate, time schedule is to be
changed to reflect concurrent engineering. WP2 should be running until month 24 instead of 18 and deliver a first
version of full WIT software by month 24.
Exploitation strategy should address the following questions as reported by PLANET:
1. WHO – A new organisation
2. WHAT – More WIT clients and WIT servers
3. WHEN – Core functionality in place
4. HOW – demos, leaflets, alliances
The STEP Issue is overviewed by the project management for its past and future implications on the project. Past
activities related to STEP have been frustrating has EPM is reported as largely out of the project by the
management and has shown little involvement and is presented by PLANET as having a different view on the
project’s orientation. Furthermore, no contact, no monitoring and no co-operation with FUNSTEP is reported this
leading to some degree of confusion. As far as an eventual future usage of STEP is concerned, PLANET asked
the question whether a STEP implementation will take place?. PLANET reported that it should not be at the
expense of, or delaying the required development at the user sites and would it take place it should be without
invalidating basic business and software design assumptions. Furthermore a close co-ordination with FUNSTEP
would be welcomed.
Presentation of “High level functionality” by ALPHAMICRO
A) General Introduction
Provide low cost, friendly and co-operative tools with general objectives. Present an “ad-hoc” diagram showing
major roles from forest exploitation, down to sawmills, wood panels, furniture, and finally direct
sales/retailers/distribution network. Three types of users are identified (direct salesman at the manufacturing site /
architect-“prescriber” / cyber user). They have analysed the full scale of an office furniture people (including
beyond RESOFLEX, architects, interior designers, etc.). Major steps are:
1. pre-design definition
2. design briefing
3. design
4. installation
5. occupation

B) Functional activity model of a generalised furniture project model
Presentation of the background and work achieved during the execution of T2.1, including the IDEF0 diagrams
reflecting the business organisation and implications on the S/W requirements. The work seems to have mainly
been performed by ALPHAMICRO and it is unclear what the WIT partnership has brought in at that stage.
PLANET’s role - the largest contributing partner (8 man/month) should be better exposed at that stage.
Low Level WIT functionality (UML) by PLANET
Presentation of the WIT S/W Design is made. Immediate implementation should follow. Common and
unambiguous understanding of the SW functionality has been reached. Clear and balanced split of the
development work can result from S/W design. Development of a documentation tool is an on-going process.
Software design is of a paramount importance without focusing on automatic code generation from the class
diagrams developed. The overall capacity of the UML language has not been used and PLANET has focused on
use case / component / class diagram and sequence diagrams (time dimension) for the logical view.
The following efforts are reported:
Case (main and detailed view)
Component view
Logical view
Presentation tier (server and client view). Instead of doing any design, have preferred to directly implement
something.
Application tier, core WIT functionality
Database tier
WIT DM and Data base schema
Directory services
PLANET describe the architectural choices made in terms of having either a centrally administered or better a
semi-centralised/distributed option (WIT servers are centrally administered and WIT clients managed by servers,
but are empowered to reconsider their server partners in the context of a business relationship). Finally WIT-N
maintains a server registry so that clients may chose where to register also knowing which functionality is being
offered by the server (and must provide assistance to servers for internet enabling their product data bases). The
hyper chain principle is exposed whereby a radical expansion of co-operation happens when clients communicate
heavily with their providers and when clients can continuously reconsider their partners (extended reach)
The various use cases are being described, including WIT client use cases, WIT server use cases, and WIT-N use
cases.
The component view describes the software packages.
The logical view provides for an in depth modelling of the WIT software platform. It is said that the logical view
should provide for a reasonably detailed description of the software platform but not delving into implementation
considerations such as which methods are private or public, etc..
Wit.client-classes (p. 39, D2.2.) To Present a user interface, Configure the environment, Select the servers,
Produce the request objects,, Produce events, Accommodate server responses.
Wit.server-classes (p. 42, D2.2.)
Wit.network-classes (p. 42, D2.2.)
Directory Services: Should be implemented by means of emerging internet standards such as LDAP technology.
LDAP supports a number of standard LDAP object classes but WIT would require some extension both to server
side classes (WIT_top; WIT_person…), and to server side attributes such as prod uct_use; Product_Type;
BaseMaterial; Surface_Material, etc.. Two modes of communication are possible, including getting URLs for
which a page is the appropriate answer to the request and cases where the entire WIT tree is being delivered.

Client implementation can be done with Sun JNDI (generalised naming and directory service interface). This
leads to the description of the wit.directory-classes both on the wit.directory.client and wit.directory.server sides.
WIT 3-tier db architecture. Common data model on the server side is presented as a major requirement for the
success of the project. Additional properties should be made available rather on the client side (such as those
described into the ALPHAMICRO specification document).
Collaboration Issues (FHG/IGD)
Described p. 21, 23, 25, 27 of D2.2. in terms of functionality to be supported. Use NetMeeting for video-
conferencing being presented as a kind of industry standard. The WIT Collaboration Class-diagrams are
presented and have been developed quite independently from the remaining parts of the Software Design
Specification.
Space Planner and Catalogue User Interface Design Issues (ZGDV)
Requirements about the user interface are expressed and must include:
Main 2D and 3D viewing area
View for collection of items
Control widgets
2D view for camera navigation
Presentation of the Model/View/Controller Paradigm (Krasner/Pope 88) as a basic approach. Then various
classes are described (plus actions applying on these objects) and implications are considered at the UI level, e.g.
for the model level we have item, tray, catalogue, room, project, etc. some of these objects being persistent, for
the controller we have catalogue, tray, space planner and at the view level we have single item, room2D and
room3D for instance. The project is the highest level of persistent object on a WIT client and serves for the
tendering / ordering processes.
Security Issues (FHG/IGD)
Watermarking is designed to address copyright issues. The WIT client is only involved when it exploits
watermarks coming along with material downloaded from the web. Different examples are provided such as
watermarking of bitmaps (Syscop robust against JPG compression) or geometry (TVR, Tetrahedral Volume
Ratio) pictures. Various watermarking systems are presented, such as Syscop (textures), TVR (3D geometry) or
Geomark (3D geometry). Presentation is made of secure and authenticated communication channels and
reference is made to SSL (Secured Socket Layer). Then, implications on the protocols to be secured are
presented including HHTP, RMI, UDP, etc. Possibilities are to perform HTTP over SSL secured connections. A
number of drawback do exist, including no transactional awareness (sessions as continuation of previous ones).
The User Focus (SHELMAN/ RESOFLEX) 1/2
A global market with an increased share for exports for companies like SHELMAN. Requirements are expressed
to have the basic functionality of the project running as soon as possible, to have the product databases populated
and to have a vision on the running costs of exploiting such a system plus a description of the technical
requirements to deploy and administer such a system (i.e. hardware, software, and skills necessary). RESOFLEX
(Pt) was purchased by a group which also bought in Germany a wood panel producer and the company stresses
the importance of supporting the entire value chain. As a general comment, the end-users are still at that stage a
little bit distant from the technological endeavours the project focuses on. They have little understanding of the
technologies in terms of how they would/could impact their business, seem to be loosely associated to the
execution and little visibility has been provided to their involvement into the definition of the end-users
requirements.

Management, collaboration
The management is capable and committed. Partners have complementary roles and good synergies are
demonstrated. Collaboration is efficient and progress - as expected and planned in the TA - has been evidenced.
Nevertheless, there are a number of difficulties appearing in the consortium. Some are of technical while others
are organisational. It is reported by the project management that technical matters related to STEP have been
addressed by resorting to external assistance and EPM has been blamed for not participating enough in the
project, including being absent to the first WIT/ESCN workshop. Further to that, and hoping to clarify the usage
of STEP within the WIT project, a subcontract has been placed to ProSTEP with the following tasks:
Analyse user requirements of WIT related to the enabled WIT STEP data model;
Implement a representation of the WIT data model using STEP methodology (EXPRESS-G);
Carry out a comparison of WIT data model and the FunSTEP data model in order to evaluate whether suitability
of FunSTEP data model for WIT.
The report from ProSTEP will come in December and in the meantime the implementation tasks will have been
clarified.
Exploitation and dissemination issues
Exploitation is a stated objective from the start and the business environment of the project is well analysed in
D1.1. End user companies have a realistic and evident commitment to exploiting the project results.
Dissemination has been encouraged properly by the project management, including a web site, papers in various
places including ESPRIT IiM, etc. A presentation on the topic is made by PLANET:
1. Exploitation is envisaged on short and longer terms basis
2. Create awareness to install clients ;
3. Minimal functionality (05/99) ;
4. Full functionality (end of the project).
Presentation is made of the various levels of exploitation:
1. User level (support WIT companies and their value chains);
2. Value chain (extend the value chains addressed);
3. Hyper chain (install and support at an hyper chain level).
Still a number of critical issues are being listed such as:
Quality and cost of business graphics over internet appears to be still one of the major questions. Think of a new
organisation to undertake exploitation such as WIT-N (must know the chain value and be computer literate);
Towards WIT-N:
Business alliances must be forged;
Clarify partner roles in WIT-N;
IPR document.
The global feeling is rather frustrating at that stage as the project is standing at the specification stage - no
product whatsoever being available - and moreover no partner or individual company seems committed to
exploiting straightforwardly the eventual product(s) which could be delivered would the project be successful.
The notion of the WIT-N is a new business model and it would be beneficial testing that idea with potential
customers to see whether they would subscribe to such a service. It could take a few years before a service like
this would make money.

STEP and WIT (PLANET/ALPHAMICRO)
Will WIT export / Import data from other applications such as ERPs systems (e.g. SAP/BaaN)
Standard CAD/CAM applications, or custom applications (would require a STEP AP in the sector) ?
PLANET expresses that:
1. STEP is not a core technology for WIT (common data model);
2. STEP may ensure a connectivity to CAD/CAM systems ;
ALPHAMICRO say that STEP would make sense if a standard would exist (in the furniture industry). They tried
to find very much about FunSTEP which were reluctant to release data. ALPHAMICRO says that the consortium
is aware of applications for projects like NIIIP or others. The consortium linked to the ESCN initiative and WIT
got involved and they organised a STEP workshop in September and concluded that an EXPRESS model could
be developed. ProSTEP will deliver in December a “STEP enabled” data model and WIT is very much service
oriented whereas FunSTEP is much more geometry oriented.
Conclusion
The review is positive and it is suggested that the deliverables be accepted. The consortium is asked to clarify its
position with respect to the usage and deployment of the STEP technology.
This includes:
Clarify the role of EPM;
Clarify implementation plan and schedule and roles of the partners;
Description of the pilots.
Next review end of May, beginning of June.

STEP positioning in the WIT project
STEP as a standard
STEP is a revolutionary concept that aims at making industrial data, of the broadest possible range, open and
available to any requesting application. In this sense, it represents a significant attempt to break away from the
attitudes of the proprietary world, where application providers more or less “tied up” their customers to their
particular- proprietary data model, rendering it useless to other applications. Clearly, the situation where a
particular item- product in an enterprise is modelled in a variety of ways to serve the needs of equally varying
applications is alarmingly common. This represents a waste of resources, leads to confusion and ambiguities and
inherently prevents any re- use of existing data models.
However, no standard (STEP included) can eventually establish itself on purely technical merits. There can be
only one and very clear metric of success, i.e. the degree to which the technology is adopted in the market. As an
analogue, let someone try to imagine himself in the 19th century with all today telecommunication facilities at his
possession. What would be their use? absolutely none, as long as he could make no use of them. Just as the
significance of telecommunications in everyday life depends on the degree the technology is deployed, in exactly
the same sense the success of STEP will be judged by the degree of its proliferation.
There has been some success cases in STEP, in its deployment in high technology sectors such as aerospace/
automotive/ shipping industry. There are also some ongoing exercises in other sectors of the economy. However,
right at this moment STEP cannot claim any significant penetration, beyond the very early experimentation, in
the more conventional and less high- tech sectors of the economy.
The main tenet is that only market forces can determine the outcome of an evolving technology. The maturity of
the technology, the promises it may righteously deliver are simply not enough. One more example: the Wankel
internal combustion engine represented a technology breakthrough as it was characterised by a much simpler
design compared to its Diesel/ Otto competitors. It would have very well established itself in the market at least
as a viable alternative. It never did, perhaps because nobody was willing to discard years of R&D and massive
investments in facilities just for the sake of a promising concept.
In the case of STEP things may be even worse as there are many big players in the IT market that would be
alarmed by the success of the technology as they would lose the “proprietary tie up” option. IT is a technology
where the promise of standards and openness has been many times raised, just to be in the very next second
disregarded by its initial proponents. There even is a view around, according to which differentiation (see:
disregard to standards) and proprietary approaches is what keeps the world moving. This just provides a
ideological alibi to what we have witnessed in so many instances. The latest surge of Internet and the battle
between Microsoft/ Sun/ Netscape is perhaps the freshest justification of this symptom.
Where can STEP be used in WIT?
WIT will build a platform to help conduct over Internet business, integrated with design. It focuses on the wood/
furniture sector but it clearly also aspires at developing tools of a high re- use potential. This is what we call “the
genericity issue” and is treated in another part in this document (D 1.3).
WIT abides by the client- server paradigm, although we believe that this relationship, taken at the Internet level,
will redefine many concepts technical and business alike. We especially expect this new environment to have a
significant impact on the very definition of the value chain (annex to D 1.1, paper submitted for publication and
presentation at the IiM- 98 conference in Sweden).
Simply put, WIT- clients using a very lean environment will have access to services provided by the WIT-
servers. For these services to materialise there are two basic requirements
1. To develop the toolset (WIT API) implementing the service functionality and to download or install it at
the WIT client level, and

2. To set- up, at the WIT- server level, the product databases, which will provide the data, required for the
applications.
For the WIT API we have already selected the constituent technologies. These will be Internet based standards
and respective technologies such as http/ html, Java/ Java-beans, VRML, LDAP, agent technologies, security and
collaboration standards and possibly component distribution technologies (CORBA, DCOM, etc.)
It is exactly in the issue of the product database where the STEP option comes in. For the product databases there
are essentially two options:
1. To build them based on the usual tools/ environments (Microsoft SQL,Oracle, etc.) which will be
accessed by JDBC, to be deployed on the server machine, as to not undermine the “thin” client
requirement. OLE DB technology can be used to access data that can typically not stored efficiently in a
relational data base such as document, image and other object formats.
2. To use STEP for the design and population of the databases.
The pro’s and con’s of the two approaches with regard to the product database set-up are largely highlighted by
the discussion in the preceding paragraph. However, in the next paragraph we will go in more detail on the
impact that the adoption of STEP might eventually have in WIT.
The impact of STEP on WIT
In the envisaged environment a WIT- client should be able through his WIT- API to systematically access and
retrieve data from any number of enabled and registered WIT servers, according to the following Figure 6.1.
PRODUCT
DATABASE
2
WIT API
PRODUCT
DATABASE
n
PRODUCT
DATABASE
1
Figure 1 - WIT Client interfacing a number of product databases hosted on WIT- servers
There is no reason to suggest that in terms of this conversation of the WIT- client with the WIT- servers any
benefit can arise from adopting STEP to model the databases.
The only benefit that can realistically be anticipated from the use of STEP in WIT has to do with the possible re-
use of the data in the product databases by other applications.
How can such a situation evolve?
1. First, if the same business entity that deploys the product database wishes to use the same data in other
applications it has in place. The probability such a instance may arise may vary greatly. However, the

internal re- use of data on a STEP product data server should not by default be considered a trivial task.
And, from the side of the business entity (and not the technology purity) there is no clear reason to
suggest that the STEP data base demonstrates an advantage compared to any other proprietary database,
unless of course the requesting application has already STEP interfaces. Only in such a case would there
be a clear advantage for the STEP option.
2. Second, an external entity could require access to the product database for other, besides the WIT API,
services. This is something that more and more should be expected to occur in the future. But again, the
advantage of STEP would arise only if this external body required STEP data, i.e., his accommodating
application was STEP enabled.
In summary the conclusions are as follows:
1. STEP has no advantage to present to the deployment of the WIT- API per se. The WIT- client has no
interest whatsoever on the way the data he accesses are modelled. The WIT- server, if it were only to
serve the Internet WIT clients and nothing more, would have no reason to implement STEP. In such an
“isolated” environment standards lose their sense and only add complexity, cost and possibly frustration.
2. STEP can be very promising in a situation where the product databases are used also by other
applications and these, be they internal or external, have to reap a clear benefit from accessing STEP
data as to any other proprietary data. This can only occur if they themselves are STEP- enabled.
Otherwise, again no advantage from STEP can be realised.
What underlies the second conclusion is what has already been stressed in the previous paragraph. STEP will
have some impact on WIT only to the extent it proliferates as a technology in wood/ furniture, in which case it is
reasonable to expect many applications deployed in the wood/ furniture sector to request STEP data. In broad
terms, such applications may be:
1. Standard business applications for enterprise resource planning systems (e.g. SAP/ BaaN, etc.) and
standard CAD/ CAM systems used for detailed design and manufacturing;
2. Locally developed applications;
3. Applications installed at other entities within the wood/ furniture sector;
4. Applications installed at other entities within a sector that conducts day to day business with
wood/ furniture. Such a case is the construction industry, that is a prominent client of the wood/
furniture sector:
5. In case (1) the potential of STEP for WIT relates directly to the degree STEP is adopted by leading S/W
manufacturers such as SAP/ Baan, etc. There is clearly little evidence in support of such a claim.
6. In case (2) no locally developed application should be expected to be STEP enabled; so there is
practically no benefit to reap in this direction.
7. In cases (3) and (4) the potential usefulness is completely connected to the adoption of STEP as a
concept by wood/ furniture and by the construction industry. Practically speaking, it is only this
perspective that may create some STEP impact for WIT. Especially, because there is on- going work in
STEP modelling in both sectors, aiming at specifying respective STEP application protocols.
The position of STEP in WIT
In the above paragraphs an attempt has been made to clearly identify what WIT should and should not expect
from STEP. The main conclusion is that STEP can unleash its potential only if it is adopted as a standard
primarily in wood/ furniture and secondarily in the construction industry. To a lesser extent, perhaps because
there are no such indications yet, WIT might gain from STEP if STEP is widely adopted by major ERP and
CAD/ CAM S/W manufacturers.

WIT wishes to develop a close to market product in two years’ time from today. Not only because the work will
be by then coming to its end. But also because major technology building blocks of WIT such as VRML, LDAP,
etc. will, with a large possibility, be by then mature technologies. More bandwidth may also be available, that
will render the whole exercise truly feasible. However, there is enough uncertainty in these issues themselves.
Going exclusively the STEP way is equal in saying that we will not be close to the market. This nobody wishes.
On the other side WIT does not wish to be left aside of the environment that may evolve should STEP eventually
be adopted in wood/ furniture. Such a development would soon render the WIT tool set obsolete by disabling it
from integration in this new environment. The attempts especially by the Fun STEP project to standardise by ISO
a STEP protocol for wood/ furniture are especially critical in our dual approach, vis a vis STEP. Although a
standard by ISO is still far from a standard in the market, it is nevertheless the first step. And as long as there are
attempts to make this step, WIT must keep a close look on them.
Conclusions
In summary WIT will pursue a dual approach, using both JDBC to interface the WIT API with common
relational databases as well as STEP- “wrapping it“ in a way it can also address STEP data bases. It will
moreover monitor closely the efforts of the Fun STEP consortium. WIT considers the move towards an ISO
standard of a wood/ furniture STEP protocol as a requirement in order WIT’s involvement with STEP not to
become a purely academic exercise.