Cover Page

B.3.3 Project management

Project management is an important task that must cover day-to-day management of the project and ensure efficient running of work packages and their coordination. All the management structures will be covered by the Consortium Agreement, together with financial issues, contingency and property rights.
In order to achieve this task with success, in addition to the Work Package Leaders (WPL) and the Project Coordinator (PC) some trans-partners units will be created. These units are the Project Management committee (PMC), the technical Coordinator and the External Liaison and Quality Manager (ELQM).

B.3.3.a Project Management Committee

Each of the partners should provide a manager that, together with the project coordinator, will build the PMC. PMC members will be in constant relation (emails, phone, document exchanges) and will meet on a regular basis (aprox. every 6 months). These meetings will in general coincide with the technical meetings and reviews. Any specific problem, however, will be dealt by particular actions of the PMC.
The PMC is responsible for:

• 
  Supporting the PC to fulfil obligations toward the EC
  
• 
  Coordination of activities within and between WPLs
  
• 
  Risk management and enacting corrective actions if problems arise
  
• 
  Ensuring that all work meets the technical and functional requirements
  
• 
  Reviewing and proposing budget transfers in accordance with the contract and the implementation plan
  
• 
  Proposing and deciding on necessary changes in work sharing, budget and participants
  

B.3.3.b External Liaison and Quality Manager

The ELQM will coordinate the external communication of the Project and be responsible of the Quality. He/She will be accountable to the PMC.
The ELQM is responsible for:

• 
  Review and assessment of the quality of the work
  
• 
  Deciding on press releases and joint publications
  
• 
  Agreeing on procedures and policies in accordance with the dissemination plans
  

The ELQM will also be aware of dissemination and exploitation activities and IPR settings.

B.3.3.c Technical Coordinator

The technical coordinator will have the role of supervising all technical tasks in such a way that they lead to the organic achievement of project goals within the planned budget and the agreed deadline. S/he will also be in charge of software quality control and will take part to all architectural meetings, irrespective of partner involvement in that particular WP/task.

B.3.3.d Project Coordinator

The PC will act on behalf of the project as a whole and provide a permanent secretariat. PC has the responsibility of representing the project to the European Commission, of overseeing the execution of the project and, in particular, the timing of the milestones and deliverables, and preparing and leading the annual reviews. The PC will act as the point of contact between the European Commission (EC) and the consortium and will attempt to solve any conflicts among partners, coordinating all the work package leaders.
The functions of the project coordinator are:

• 
  Responsibility for ensuring the scientific success of the project
  
• 
  Monitoring the overall project planning and reporting from participants to provide continuous review and assessment of project results and progress to achieve the objectives
  
• 
  Preparing contract with the EC
  
• 
  Assisting WPLs and make sure than the communication among them proceeds smoothly to ensure a successful integration of the various parts and components of the project
  
• 
  Collecting from the partners the cost and other statements for submission to the EC
  
• 
  Preparing the reports and documents required by the EC, with support from the partners
  
• 
  Ensuring delivery of all deliverables specified in the contract or those required by the EC for the corresponding reviews and audits.
  
• 
  Preparing the board meetings and decisions of the board, chairing the board meetings.
  
• 
  Appointing a Project Administrative Assistant who is responsible for the non-scientific aspects of the project and supports the project coordinator in day-to-day business. This person is responsible for handling daily administrative issues and is also the contact point for media and the public. Is responsible for:
  
  • 
    Dealing with financial and administrative issues
    
  • 
    Handling legal and contractual issues
    
  • 
    Monitoring and updating project schedule and workplans
    
  • 
    Organization of project meetings, workshops and reviews meetings
    
  • 
    Preparation of press releases, project presentations and flyer
    

B.3.3.e Work Package Leaders

Every work package is assigned to a WPL who has the responsibility to ensure that all intended tasks in their work package are completed on time, including:

• 
  Sending the expected deliverables to the PC, and meeting the specified milestones.
  
• 
  Taking technical decisions concerning their work package when those decisions do not interact with other work packages.
  

B.3.3.f Conflict Resolution

Adequate records of the nature of the conflict will be kept for reporting purposes and for planning future activities. The PC will organize a conflict resolution meeting within 30 days following the reception of a written request transmitted by any of the project partners. Requests for meetings must include hints for solutions.

Step 1: Discussion and Dialogue

The preferred method for dealing with conflict is open discussion and dialogue by the conflicting parties.

Step 2: Mediation at the WP level

Parties whose disputes are not resolved in step 1 may utilize mediation through the appropriate WPL. Here, the aim is to facilitate communication between disputants with the goal of assisting them to reach a resolution. The PC must be informed as early as possible of the0 nature and status of the conflict. If the conflict is not resolved at this stage then it moves to step 3.

Step 3: Referral to the PMC

Conflicts should be resolved unanimously by the PMC provided each concerned partners is represented. If the problem is not satisfactorily dealt with in a timely fashion by the PMC then it moves to step 4.

Step 4: External Resolution

In the case of a failure to reach an agreement, as a last resort the conflict may be referred to and finally adjudicated for resolution by a courts of law having jurisdiction in the matter.

B.3.3.g Organizational and change management

Each site manager will be appointed since the beginning of the project on the basis of managerial and technical skills. Should a change intervene in the organization, the partner will take the responsibility of replacing her/him with personnel with analogous skills both from the technical and managerial point of view.
Should for any reason an organization withdraws from the consortium the PMC will immediately try to identify an organization with analogous skills. In case this is impossible, the consortium has been built by taking into consideration that certain roles and competencies are partially overlapping. Major overlaps are built on pairs of XEROX/CELI for NLP, UNITN/UVA for MT and algorithm configuration, OD/GONET for system integration and log analysis, UBER/BAL for digital libraries and user evaluation.

B.3.4 Security, privacy, inclusiveness, interoperability; standards and open source

The GALATEAS architecture is based on web services, so it is intrinsically interconnected. Web services will be based on SOAP (currently maintained by the W3C) and all language oriented web services will be delivered via UIMA (http://incubator.apache.org/uima/: Unstructured Information Management Architecture) platform. UIMA enables applications to be decomposed into components (for example "language identification" => "language specific segmentation" => "sentence boundary detection" => "entity detection (person/place names etc.)"). Each component implements interfaces defined by the framework and provides self-describing metadata via XML descriptor files. The framework manages these components and the data flow between them. Components are written in Java or C++; the data that flows between components is designed for efficient mapping between these languages.

UIMA additionally provides capabilities to wrap components as network services, and can scale to very large volumes by replicating processing pipelines over a cluster of networked nodes.

Apache UIMA is an Apache-licensed open source implementation of the UIMA specification. That specification is, in turn, being developed concurrently by a technical committee within OASIS , a standards organization.

All UIMA services are based on type systems i.e. declarative statements of components capabilities and their semantics. GALATEAS will use as much as possible available type systems. Whenever this will be impossible, we will try at least to reuse part of them, and we will obviously publish the modified type system on all UIMA related repository.

Concerning log analysis GALATEAS will map possibly proprietary log format to the W3C “standard” Extended Log File Format (http://www.w3.org/TR/WD-logfile.html).

Concerning privacy, there is an issue related to the fact that GALATEAS will handle query logs where potentially private information might be stored. However, the large majority of log formats hide the information about the username, and the only key for connecting several action by the same user is the anonymous session identifier. If, in the course of the project, some non anonymous query log format will appear, the consortium will apply some standard anonymization technique at customer side. These techniques rely on the fact that transaction logs are highly structured objects and the user information can always be detected by deterministic methods.
Finally, concerning open source, as we stated in several sections, all software produced by GALATEAS using programs originally available under an open source license will be licensed under an open source schema, and the development process for that software will be based on some public open source platform. We expect that this, together with isolation of proprietary modules under web services, will leverage scientific and commercial take up.

B.3.5 Resources to be committed

The main financial effort documented in form A3 concerns personnel and indirect costs: here are average costs which have been used for the computation of the break down table:

	costs
XRCE	6945
BAL	6943
C0ELI	6741
GON	6700
UBER	4132
OD	7027
UNITN	5683
UVA	5278

The project is funded 50%. Academic bodies have already verified the possibility of involving internal personnel in GALATEAS activities as a form of co-funding. Private companies will invest their own resources in order to carry on all activities in which they are involved. From a first examination of their balance sheets it emerges that they can allocate enough money to provide reliable co-funding for the three years long project. Moreover, we expect that part of the co-funding will be derived by revenues originated by the exploratory commercial activities during the third project year.

In general we considered that 6.000 € per partner should cover travel expense for the three years duration. Xerox and CELI, being the project coordinator and the technical coordinator respectively, have an extra budget of 2.000 € to cover additionally required travel. Moreover CELI and OD are assigned globally 18.000 € additional other costs. These are meant to cover reliable hardware acquisition for both off-line processing and web services provision, as well as the costs of bandwidth allocation.
Xerox has subcontracting for 18.000 Euros. These will be used to by two months consultancy of Dr. Ulrich Kampffmeyer, who is currently a consultant for Europeana 1.0 as experienced specialist. Thanks to this consultancy we expect to strengthen even more the liaisons with European digital libraries efforts.

B.3.6 Dissemination / Use of results

B.3.6.a General Dissemination activities

The main targets for dissemination are:

• 
  Digital Libraries and OPAC-running libraries as stand-alone organizations.
  
• 
  Aggregation of Digital libraries and OPAC-running libraries accessible by a unique entry point.
  
• 
  The community of users of resident search engines.
  
• 
  Scientific Community of both the NLP and Digital Libraries sector.
  

In practical terms, dissemination activities will be focused on the following major types of objectives:

• 
  Awareness raising
  
• 
  Knowledge and Experience Transfers
  

In order to achieve these objectives the channels will be

• 
  New social networks channels: activities on Facebook and Linkedin
  
• 
  Standard channels of relevant scientific communities (e.g. presentations at conferences, workshops, articles in scientific journals, etc.). Impact of the Consortium work will be disseminated in the two scientific communities involved in the Project (NLP community and above all Digital Libraries /Search Engines community).
  
• 
  Participation to evaluation experiments such as CLEF.
  
• 
  Project Web site: besides been indexed by all major search engines the project web site will be the object of a campaign for acquiring a better visibility on the net. Such a campaign will be implemented with the following methodologies:
  
  • 
    Definition of keywords related with the official Web site content;
    
  • 
    Definition of search statements of potential interest for all the Web users, based on the pre-defined keywords
    
  • 
    Use of a specific application (a product of GoNetwork s.r.l.) for the generation of an high number of Web pages optimised for all the major search-engines (Search Engine Optimization).
    
• 
  Service Web Site: after the first year of project a web site will be launched with the goal of describing the services provided by GALATEAS partners, not the project itself.
  
• 
  European Commission Dissemination Services: Production of a video in English (e.g. flash ) publicizing GALATEAS objectives, target users, technologies used and services offered. This video will contain a demo of the services offered by GALATEAS. It will emphasize the impact both on end user satisfaction and ROI by potential customers. This video will be given to the Commission services for its own dissemination and awareness activities during and after project completion. It can be used as the “Public end-of-project showcase” as required in annex X.
  
• 
  Presence in major fair trades in the field of IT support to (digital) libraries.
  
• 
  Production of marketing material such as brochures and PowerPoint presentations (since the very beginning of the project) and commercial flyers (since the first year of activity)
  
• 
  Publication in IT management oriented reviews.
  
• 
  Press release (2 press release, in correspondence of the launch of each fee-sustained service)
  

B.3.6.b Assessment of Dissemination activities

Dissemination of components available under open source will be assessed by the number of downloads (popularity) as published by all major open source development platforms. We expect at least 150 independent downloads by the end of the project. In order to increase that number, the GALATEAS consortium might organize GALATEAS-based evaluation experiments (possibly in the context of CLEF)
As for general dissemination, the following methods of assessment will be adopted:

00

• 
  Academic dissemination: we estimate that academic dissemination can be fairly monitored by counting the number of published papers and talk to conferences. In general we consider that an average of ten items per year is satisfactory.
  
• 
  Web dissemination. For access to web sites the consortium will internally publish statistics of access to web site. On the basis of the results provided by Google AdWords, we have computed that a web site such as the GALATEAS project web site should receive at least 160 new visits per months. The ambitious goal is to increase these access by an order of magnitude thus reaching about 1500/2000 new visits per month by the end of the project. This result will be achieved both by explicit advertising and by a careful indexing campaign by major search engines (GoNetwork).
  
• 
  Networking. The number of contacts cannot be considered in itself a success indicator for a valid networking action. Adoption of the GALATEAS services by third parties (irrespective of commercial conditions) is a better indicator.
  
• 
  Fair trades: For talks the number of auditors represents a quite reliable quantitative indicator. For “stand presence” the number of registered contacts per day will be used.
  

1 In general to limit sparseness of data we will require that customers of the LangLog service will provide a classification system with no more than 500 categories and a depth inferior to four.

2 Of course the information provider must also transmit the association map between IDs of digital objects and its classification system.

3 In most cases this translation might be even more relevant than the ones provided by standard translation dictionaries. Indeed, non-specialist translation dictionaries tend to model everyday language, where the "language" of library record is different in the sense that I) it might be specialized for a certain domain; II) it is more formal than everyday language. Therefore we expect user translation to be more pertinent and free from senses that are present in the common language but unlikely to occur in a library catalogue.

4 All figures in this section are from International library statistics: trends and commentary based on the Libecon data.by David Fuegi Martin Jennings, 30 June, 2004, projected on the basis of sampled growth in expenditure.

Version of 5 February 2010 agreed with the EC page  /