Third Working Group on
SIPAM System Design Development
Iraklio, GREECE 4th – 7th April, 1995

Denis LACROIX - Othmen BEJI
Mikos PAPANDROULAKIS - Bruno MENU

TPITH ΣYNANTHΣH THΣ OMAΔAΣ
EPΓAΣIAΣ TOY ΣI ΠAM ΓIA
BEATIΩEH TOY ΠPOΓPAMMATOΣ

FAO/MEDERAP/SIPAM
MEDiterranean Regional Aquaculture Project
System of Information for the Promotion of Aquaculture in the Mediterranean

Abstract

SIPAM project ends now the first phase of experiment as this thied working group meeting may be considered as the milestone of the adoption of the prototype.

Eight countries attended this meeting (Cyprus, Croatia, Egypt, France, Greece, Spain, Turkey and Tunisia) during which the final approval was given to the structure of the various databases, the choice of parameters and the definitions of all table elements.

The two lectures allowed to better evaluate first, the difficulty of data management with several levels of collection for several types of end-users (J. Bostock, Stirling University) and second, the minimum legal frame that has to be discussed and precised in order to clarify SIPAM legal status for future scientific and commercial relations (L. Hefferman, FAO Legal Office).

As the development of SIPAM network remains dependent on external financial supports, the search for permanent (countries and/or intitutions) and start-up support (European Union) is a necessity. If a newsletter is not justified at the moment, a yearly map of aquaculture was considered to be a valuable sign of SIPAM usefulness.

It clearly appeared that the interest of SIPAM is directly related with the combination of two elements : quantity and quality of data. Both have to be now tested at the real scale of an operational system through exhaustive data entry. Both can be improved by the demand of exacting end-users.

The role of GFCM merges to he now essential in several fields : constitution of SIPAM coordination committee, approval of the budget and support proposals, agreement of legal requirements. Its relay of MEDRAP II should become effective at the next GFCM meeting (MAY 1995) which would ensure a better responsibility endorsement of the development of SIPAM by all actors.

Résumé

Le troisième groupe de travail du project SIPAM peut être condieéré comme la clôture de la phase expérimentale du project cur il s'est terminé par l'adoption définitive du prototype du logiciel. Les huit pays présents (Chypre, Croatie, Egypte, Espagne, France, Grèce, Tunisie et Turquie) ont donné leur accord final à éléments des tables référence.

Les deux conféfrences données au cours de cette réunion ont permis d'une part, de mieux évaluer la difficulté de gérer des données complexes à partir de différents niveaux de collecte et pour différents types d'utilisateurs (J. Bostock, de l'Université de Striling), et d'autre part, de [précise le cadre légal indispensable á proposer pour le SIPAM afin de lui permettre de mener á bien sa double mission scientifique et commercial (L. Hefferman, bureau juridique de l'OAA).

Comme le développement du réseau SIPAM reste dépendant des sources de financement extérieures, la recherche de financements permanents (Estats et/ou institutions) ou limités dans le temps (aides européennes) est une nécessité.

Par ailleurs, si la diffusion d'une lettre de liaison est consiéréc comme prématurée, la préparation d'une carte annuelle de l'aquaculture en Méditerranée a été décidée en raision de sa double valeur d'utilité et d'image.

Il est également clairement apparu que l'intérét du réseau SIPAM est directement lié à la conjoncture de deux éléments clefs: la quantité et la qualité de l'information. Ces deux éléments doivent maintenant être testés en vraie grandeur sur un système désormais operationnel par l'entrée de toutes les infoarmtions disponibles pour chaque banque de données dans chaque pays. Les deux éléments pourront aussi être améliorés en fonction de la demande des utilisateurs finaux.

Le rôle du CGPM est désormais essentiel: il intervient dans la création du Comité de Coordination du SIPAM, dans l'approbation du budget et des propositions de financement, dans l'accord des propositions de cardre juridique à l'action du SIPAM. Sa prise de relais du SIPAM devrait être effective à l'occasion du prochain conseil (Mai 95) ce qui permettra une meilleure prise de responsabilité du développment du SIPAM par tous ceux qui en sont les acteurs.

NB: From page drawing is the logo of IMBC (courtesy of A. Eleftheriou, IMBC)

I. Introduction

The third working group meeting on SIPAM system design development organised by MEDRAP II project in collaboration with FAO (FIRI) was held from the 4th to the 7th of April, 1995 in the Institute of Marine Biology of Crete (IMBC) centre of Iraklio (Greece, isle of Crete). This meeting was attended by representatives of SIPAM from eight countries: Croatia, Cyprus, Egypt, France, Greece, Spain, Tunisia and Turkey.

Were also present:

Annex 1 gives the list of participants and corresponding address.

The meeting was opened by Mr. A. Eleftheriou, Director of the IMBC. He welcomed all participants and underlined the fact that IMBC has already started to run a database program, named “Aquabank” with the financial support of the European Union and gained some experience of data collection and processing. He expressed his wishes that Greece and especially IMBC would bring a valuable support to SIPAM.

As SIPAM data manager, D. Lacroix thanked M. Eleftheriou for hosting SIPAM working group and, as Mr. Coppola's arrival was expected to be very soon (delay due to strike on Athens airport), D. Lacroix proposed to start working by a brief summary of the present situation of SIPAM. Thsi seemed all the more justrified as SIPAM group welcomed a new member, Spain, represented by Dr. Ignacio Arnal from the spanish Institute of Oceanography (IEO) and as Mr. Bruno Menu was representing France instead of Mr. Ch. de l a Pornélie. He presented to all participants what had been done since the previous meeting in Nantes. The planned workplan was globally achieved although numerous improvements had to be integrated into the 2nd version of SIPAM before distribution to the countries. Therefore, this meeting in Greece may be considered in charge of the final adoption of the prototype before exhaustive data entry in each country as planned in Nantes report.

The officials of the meeting were then designated as follows:

The preliminary agenda was adopted after some minor modifications (see annex 2).

II. SIPAM System Development

II. 1 Situation in the pilot countries

It was proposed to start examining the situation in the pilot countries in order to benefit of a good knowledge of the problems and constrainsts that could have hampered the use of SIPAM software.

Croatia :

Mr. Vlado Dadic met some problems in the installation of the software. He had limited time to test it with real data except for production which database is now complete. Some colleagues biologists experimented its use and found it “friendly”. “modify”, “delete” and “search” functions showed to be inconvenient as too many references are required to obtain an answer.

Cyprus :

Mrs. Daphne Stephanou received the diskettes on early March. No major problem was encountered neither in the installation nor in the use of the software, but the programme has to be tested with more entries since there was not enough time available to do so. Although some refinement seems necessary especially concerning the “search”, “modify” and “delete” functions. She highlighted the fact that the laws and regulations database should be tested also by specialised lawyers and not only by biologists because those options are not user friendly. She complained about the lack of time between receiving SIPAM 2 software and reporting at the Crete meeting. In addition, some functions were not found so friendly to use although a better interface with user would be useful.

Egypt :

Mr. Magd El Baweb underlined the fact that nay software import in Egypt was controlled by customs and this filtering procedure caused and important delay in the transmission of document. Thus, he asked to use preferably the FAO pouch, much more rapid and reliable. When loading the software, he found that the content of the diskette № 5 was not available. He appreciated this new version, much more colloquial than the previous one and although some definitions remain uncertain (brackish/marine/fresh waters), he took advantage of this software data collection in Egypt and the quality control of those data.

Consequently, he was not able to present standardised and precise data on aquaculture production in Egypt.

France :

On behalf of Ch. de la Pomélie, Bruno Menu reported that no particular problem was met in the use of the software. Few bugs appeared notably in “programmes” database. The lack of time had unfortunately limited the test.

Greece :

Although limited time could be spared for the test of the software, Nikos Papandroulakis noted with great interest the potential of this program especially as an international complement of the existing national network called Aquabank (see annex 4). He noticed that some table lists were not complete (country for example).

Spain :

Mr. Ignacio Arnal was happy to participate as Spain representative to this SIPAM working group. As the adherence of Spain to this network was recent, he had received through CIHEAM the copy of the diskette only one day before leaving Madrid to go to Greece. So he only had time to check that installation was easy and to estimate that requested data represent an important and specific effort for Spain as detailed and reliable data about all sectors of aquaculture are not available at the moment.

Tunisia :

Mr. Zouhaier Saghrouni underlined that he benefitted of the proximity of SIPAM office in the Ministry of Agriculture in Tunis. He noted that some species reared in aquaculture were not found in the “product” table such as “sea bream” and but were easily solved with the help of the programmer Mahjoub Zitoun.

Turkey :

Miss Oya Ersan met no problem when installing and using the software except with diskette № 4 in which a virus ws detected. She complained about the impossibility to pass though some areas without giving a data or select an answer in table. As numberous data are missing in Turkey, the fulfillment of a complete record in any database is not easy. Anyhow, she greatly appreciated the opportunity that gives SIPAM to start an exhaustive data collection in all fields related to aquaculture in Tukey.

II.2 Situation in the SIPAM Regional Centre

The SIPAM regional centre work has progressed in accordance with the workplan established by the two working groups of Nantes (Nov./Dec.94). The team of Tunis has introduced the modifications and the improvements in accordance with the recommendations of the working group.

Seven diskettes representing the second version of the SIPAM software including executable programmes, their auxiliaries, the datafiles to run the sustem (and on installation procedure guide) were sent to all the SIPAM representatives. A user's manual “pre-alpha release” was prepared in Tunis and was distributed to the participants of Iraklion SIPAM working group. This document collects all the standard definitions, tables used by the software and the database with the different screens and the selected Directories.

In order to support the project, some equipments were purchased by the SIPAM regional centre such as:

-   1 Modem Robotics 14,4

-   1 “lecteur” of laser disk.

Concerning Internet connection a preliminary contract was established in Tunis with the head office (IRSIT) who will connect the regional centre to Internet. E-mail facilities and FTP (file transfer protocol) procedures were discussed. The contract was sent to FAO Hq (system designer) for advice. In Crete, the decision was to continue the discussing with IRIST, in order to find the best way to cooperate. Other possibilities in cooperation with FAO representation in Tunis and another FAO project were suggested and will be explored in order to share installation cost of Internet.

The “pathology” database had been prepared by SIPAM team from the basic model of “SIPAM” pathology database. Checking and improvement will be done by FAO expert before transfer to SIPAM module.

II.3 Overview from SIPAM system designer

a) Methodology

The value of the methodology is directly linked with two key-elements: quantity and quality of data;

This implies that three elements are essential in the process of networking:

-   the country which commitment has to be sincere and concrete: financial, technical and administrative support of he national couterpart;

-   the national representative in charge of the collect and the management of all data related to aquaculture in his country. Its exactness and intellectual honesty are crucial;

-   the end-useres who are also the sources of data who have to be convinced that the quality of what thye are expecting from the network is related to the quality of the data they give to the national representative. For several reasons (political, financial, economic,…) they are not so often disposed to deliver real data.

b) Operational frame

It can be summarised by the following figure:

c) Technical aspects

Several aspects have to be clarified as many participants asked precise questions on technical aspects:

• “What to do when data are unprecise?”. The answer is simplification. The essential task at the moment is the demonstration that SIPAM can run. Therefore, if precise checking of data is not possible, its record under a more general item has to be found. Refinement will be undertaken in the future.

• “What to do when some data are not available in a general package of precise data?”. The answer is estimation with a precise rule which has to be explained and justified. Later, it will be possible to adjust the estimations.

• “What to do when the acronym doesn't exist?”. The acronym is a compulsory key in two tables (suppliers and institutions). Hence, it has to be filled. It is proposed to use a pseudo-code to be defied according to countries suggestions.

• “What is the frequency of data collection?”.

  It is obvious that data collection frequency depends on the type of data. Most of the databases have to be updated on a yearly basis ("Experts, Institutions, Suppliers, etc…). “Production” could be updated every half-year but it is untimely to ask fo this constraint now. Participants were asked to prepare propositions to be examined during the next meeting. Later on, when data about marketing and prices will become available, a monthly frequency could become necessary.

• “Is it possible to add specific databases in the national module?”.

  Each country has the possibility to replace an usual file by another one or even a complete package through the “BAT” procedure (command “call” in DOS). For example, France can replace “suppliers” by “Aqualog” after checking the copyright and the interface procedure. Similarly, a national system for bibliography can be installed as a package. Interface routines with “Procite” regional bibliography system will be installed later.

  New references can be added in the national program such as new species when needed. They will be screened and added if justified in the regional module. Additional footnotes would help to select relevant new references.

  The responsibility for the selection will be assumed by the SIPAM coordinator under the control of the coordination committee with a regular open discussion during the two yearly workshops.

• “How to finance the development of SIPAM after MEDRAP's end?”.

  It has to be clear that even if SIPAM is supported by donors for some additional years, the goal is to become financially autonomous. As previous networks (Infopesca, Infosamak, etc…) have met some difficulties in balancing their budget, SIPAM must integrate the demand of potential end-users since the beginning. All those end-users such as banks, engineering officers, investors are ready to pay for good quality data but they want value added information.

• “How to gain time and efficiency in data transfer?”. The answer is INTERNET. Up to new only five country representatives are connected: Croatia, France, Greece (IMBC), Spain and Turkey, FAO Hq will be soon connected to all regional centres. The regional office in Tunis should be soon connected also which will facilitate exchange of information and data especially when the File Transfer Protocole will be installed. However, E-mail ad FTP remain only telecommunication tools which utility depends first on the relevancy of the message.

II.4 Finalisation of definitions and tables

It was agreed by the participants to adopt the definitions of Aqua.stat (FAO software) and to complete what is missing. concerning “pathlogy” and “Legal aspects”, participants are requested to wait for final modification by FAO office in order to fit with FAO usual standards.

a) Environment

1. Aquaculture is the farming of aquatic organisms, including fish, malluses, crustaceans and aquatic plants. Farming implies some form of intervention in the rearing process to enhance production, such as regular stocking, feeding, protection from predators, etc. farming also implies individual or corporate ownership of the stock being cultivated. For statistical purposes, aquatic organisms which are harvested by an individual or corporate body which has owned them throught their rearing period contribute to aquaculture, while aquatic ogranisms which are exploitable by the public as a common property resource, with or without appropriate licences, are the harvest of fisheries.

2. By freshwater environment, it is understood the cultivation of aquatic organisms where the end product is reised in freshwater; earlier stages of the life cycle of these species may be spent in brackish or marine waters.

3. By marine environment, it is understood that the cultivation of the end product takes place in sea water, earlier, stages in the life cycle of these aquatic organisms may be spent in brackishwater or freshwater.

4. By brackishwater environment, it is understood the cultivation of aqutic organisms where the end product is raised in brackishwater; earlier stages of the life cycle of these species mays be spend in fresh or marine waters. Brackishwates are characterised by large seasonal fluctuations in salinity. If these conditions do not exist or have no effect on cultural practices, please record production under either “freshwater culture” of “mariculture”.

b) Type of product:

1. Flesh: Animal product destinated to direct human consumption.

2. Seed: Living organism at various steps of development used for grow-out in aquaculture farm and natural waters. This class included breeders and eggs.

3. Derivated product: Product or sub-product obtained from cultured organisms for industrial purposes or jewellery but never for direct human consumption.

4. Ornamental: Living organisms (mostly exotic fishes) used for ornamental purposes only (aquaria).

c) Structure of culture

1. Ponds and tanks: artificial units constructed above or below ground level capable of holding and interchanging water. Rate of exchange of water is usaually low, i.e. not exceeding 10 changes per day.

2. Enclosures and pens: Waer area confined by net, mesh and other barrieers allowing uncontrolled water interchange and distingueshed by the fact that enclosures occupy the full water column between substrate and surface, pens and enclosures will generally enclose a relatively large volume of water.

3. Cages: open or covered enclosed structures constructed with net, mesh or any porous material allowing natural water interchange. These structures may be floating, suspended, or fixed to the substrate but still permitting water interchange from below.

4. Raceways and silos: artificial units constructed above or below ground level capable of high rates of water interchange in excess of 20 changes per day.

5. Dam/reservoirs: semi-permanent of seasonal enclosures formed by imprervious man-made barriers and appropriate natural features.

6. Suspension culture: artificial device for the production of macro-algae or molluses, equipped with systems for floating and anchoring. Cultured organisms can be exposed temporarily to sun and air. The main characteristic of this type of culture is the benefit of the tri-dimenstional space of the water column.

7. Bottom culture: system of culture for the production of algae and molluse on the ground. Special equipment as net set or liners can be used to facilitate manipulations.

8. Fish and rice culture: it is a pond used initially for the culture of rice and in which fishes are added for grow-out purpose.

   Concerning the tables used by the software, few modifications has been introduced according to the suggestions of the working group.

III. Methodology of data collection (Stirling University experience)

John Bostock is project officer at Stirling University and especially in charge of the follow-up of data collection at the institute of aquaculture. A summary of his paper is presented in annex 5.

IV. observations on the legal aspects of SIPAM.

IV.1 lecture from FAO legal office (Liz Hefferman).

The lecture is sammarised in annex 6.

The discussions were centred on three topics: the general legal framework; the copyright implications; and the legislative database.

IV. 2 Main comments on FAO Legal Office lecture

After the communication of Miss Liz Hefferman, three points were discussed by the participants :

• Legal general framework

The mandate of GFCM of SIPAM is given. Some problems remain about the situation of countries which are not members of GFCM and members of MEDRAP II project. They have eithr to ask to become member of GECM before entering SIPAM or aks to FAO a special authorisation to become directly member of SIPAM. The commercial aspect of SIPAM has to be also taken in account in the latter of agreement between the different parts involved in SIPAM.

• The copyright of SIPAM

The main question remains who has the “property” of any SIPAM product? It was mentioned that three is few general laws concerning the copyright at an international level and recogising the intellectual protection of databases and computer softwares for a certain duration (20 years).

As Tunisia is the hosting country of SIPAM, the tunisian laws concerning this protection must be harmonised with the laws of other countries members of SIPAM in order to adopt the same licence for the use of SIPAM.

Waiting for a final decision, L. Heffernan suggested to secure the copyright of SIPAM by “@ SIPAM, year” and to transfer to the countries the right to prosecute locally illegal end-using.

• Legal situation of SIPAM

Proposals were discussed and agreed by the participants:

-   the legal situation of SIPAM has to be clarified;

-   under the umbrella of GFCM, letters of agreement have to be signed between the GFCM and the tunisian government and between GFCM and other member countries.

Concerning the SIPAM database called “laws an regulations”, some discussions has taken place especially about the classification of the legislation, the guidelines for keywords and official or non-official abstract of the laws and the language to be used for this purpose. A proposition will be submitted soon by the Legal Office after checking the existing database.

V. Development programme

V.1 Immediate tasks

The immediate task to be achieved as soon as possible are sumarised in the following table.

Mr. Coppola underlined that the first priority is data entry. Even if small modifications are introduced later, the structure of the software will not change.

V.2 Workplan

A workplan for the next 3 months had been discussed and approved:

Concerning the adherence of new members in SIPAM, Mr. Akrout stressed that precise conditions have to be given in order to “filter” candidates suchs as:

-   ad hoc computer equipment (minimum power and memory requirements, softwares availability, E-mail connection through INTERNET, etc…;

-   qualified and available personnel;

-   existing data network about aquaculture in the country;

-   will be play an active role in the network.

V.3 Budget

Aquaculture Committee Meetings

For the regular meetings of the GFCM Aquaculture Committee, it is proposed to hold one meeting per year, the duration of which would not exceed three days. FAO would cover the attendance of the SIPAM Regional Coordinator and two/three FAO headquarters staff, depending on the agenda. The participation of country representatives would be covered by the countries themselves as in the case for other Committees of the GFCM. The annual cost of the Aquaculture Committee Meeting is estimated at US$ 3500–5000.

Participation in Management Meetings of EAM. TECAM and SELAM

The GFCM Aquaculture Committee Technical Secretary should also participate, as requested in the constitutional documents of the three networks, in the annual management/coordination meeting of the other three networks, TECAM, SELAM, and EAM for discussion of the progress achieved and for finalisation of the annual work programme. These are meeting with a duration not exceeding three days. This participation would imply an annual estimated cost of US$ 3000.

SIPAM Management Meetings

The SIPAM Management Meeting, would normally involve the participation of the SIPAM Regional Coordinator from the Tunis office, two staff members from FAO Headquarters, and two experts from the region participating in the national SIPAM work and attending on a rotational basis. This meeting would have a workshop characteristic to work on the new data bases structure and in the finalisation of the programme of work, and would imply a cost to the organisation estimated at US$ 4000 per year. Similar to the coordination/management meetings of the other three networks the SIPAM management meeting will discuss the progress achieved and adopt a programme of work for the following year. The duration of this meeting should also not exceed three days.

SIPAM Scientific Activities

In addition to the SIPAM Management Meeting it is considered convenient to organise, at least for the first years, one scientific activity per year for SIPAM. This activity would tate the form of a seminar of 2–3 days duration in which the work of the network will also be presented to new potential users. The FAO could cover the attendance of up to 8 participants from the countries already linked or to be connected to the network, but the attendance would be open to more participants which will have to cover their own attendance cost from various sources. These seminars should also provide elements for the design of the future work programme of the SIPAM network. The estimated cost of this activity to FAO would be about US$ 8000 per year, and it is expected that the institutions hosting these seminars will contribute partly to the cost.

SIPAM Expansion

In the initial biennium it will also be necessary to consider the expansion of the network from the present 8 coutries to include the maximum number of countries belonging to the GFCM. The travel required to brief the new countries joining the SIPAM will have an estimated cost of US$ 3000 per year, but will be limited to the initial biennium after which it is expected to have the definitive structure of the network in place.

Summary of Expenditures

The total cost of the activities indicated above, excluding optional activities (see below) to the GFCM Secretariat would thus be comprised between US$ 21500 and US$ 23000 distributed in the following way:

If the SIPAM Management Meeting could be held in conjunction with the meeting of the GFCM Aquaculture Committee there would be savings estimated at about US$ 3000, bringing down the estimated funding requirements to US$ 18500–20000 which would decrease to US$ 15500–17000 after the first biennium. In principle, and resources permitting, the above mentioned costs could be met by the FAO Secretariat, through FAO Regular Programme funding. It is however expected that in the future it should be possible to finance at least the scientific activities of the network form voluntary contributions of the member countries or from revenues generated by the network activities through the sale of information generated by the SIPAM network.

Administrative Support

The Aquaculture Committee and the SIPAM network would be administratively supported by the GFCM Secretariat, and assisted by a Technical Secretary and by the technical staff in charge of fisheries information systems development from the FAO Fisheries Resources and Environment Division. The reports prepared for the various meetings of the Aquaculture Committee and the SIPAM Network Management Meetings would be edited and published by the FAO in the GFCM series. The operation of the SIPAM Regional Centre will be financially supported by the Tunisian Government, for what are the standard costs as indicated in the exchanged letters of agreement between the Tunisian Government and the FAO.

Possible other activities to be considered in the future

-   Training of national SIPAM new operators. (one workshop/year).

-   Specific training for a better use of computerised networks in cooperation with FAO office. (one workshop/year).

-   Editory of specific maps and synthesis book about aquaculture in the Mediterranean.

-   Experts missions to study opportunities for the improvement of the system.

-   Development of relations with other regional information networks (SIPAL, INFOFISH, INFOSAMAK, GLOBEFISH…).

-   Development of specialised aquaculture databases (TECAM, SELAM, EAM).

-   Development of paying services for external clients and study on the demand in this sector.

It is clear that the extended budget is not yet funded and that contributions from other sources will be necessary. The preparation of the budget will be one of the responsibilites of the GFCM through the committee of aquaculture. This is all the more justified as there is a good opportunity to present a support proposed to the European Union in the next months. This proposal can be done either through an international organisation (GFCM) or through a multi-lateral project associating northern and southern countries. MM. Arnal and Dadic were asked to get in touch with CIHEAM and PAP/RAC respectively in order to prepare a common request to the Europeau Union. Indeed, M. Coppola remined that CIHEAM has asked SIPAM to be the official link between the four network. This implies the involvement of SIPAM in the preparation of the proposal to the European Union. Mr. Akrout suggested that Ifremer could support the request to the European Union by a part-time person who could coordinate the preparation of programs to be financed. Mrs. Kentouri stressed the urgency of this action for deadlines are soon.

V.4 Potential actions

a-Newsletter

The usefulness of a newsletter has been recognised since the first version of the project document for an information system (1991); it has been confirmed in the final SIPAM project document (1993) and during the last steering committee meeting of Beymelek (1994).

Recently, CIHEAM and PAP/RAC representatives confirmed that a usual link between the specialised networks as a newsletter was compulsory in order to avoid the exclusiveness of computerised data and to facilitate the interface with non-computerised systems.

However, M. Coppola presented three remarks:

-   since a FAO aquaculture newsletter already exists, no more resources can be provided by FAO Headquarters;

-   it is untimely to launch this bulletin as nu financial resources are secured at the moment and the worst position would be to stop edition after 2 to 3 issues:

-   the priority is clearly on software improvement and data entry, not on spreading the products of SIPAM on a large scale. SIPAM has to demonstrate first its efficiency and usefulness. Later, time will come for the diffusion of SIPAM system by other ways, like a newsletter. Consequently, it was decided to postpone the realisation of this newsletter until favourable conditions were combined.

b-Map

The project of a yearly map of aquaculture production in the Mediterranean was first presented at the Nantes working group and later approved by the last steering committee meeting of MAEDRAP. As it represents an excellent symbol of a common interest cooperation among mediterranean countries, this project was approved without restriction. M. Arnal suggested to prepare in the same time a small booklet detailing the data and their sources, which could be distributed with the map.

VI. Conclusions and recommendations

The first conclusion of this working group is that SIPAM software prototype may now be considered as adopted by all the member countries. This was obtained through two convergent ways:

-   this project could be useful to all countries as it had been discussed and tested by the countries themselves, although it still need further improvement before final spreading;

-   the agreement, of all participants to the content and the parameter definitions of all the databases is now clearly established as a stabilised basis for standardised data exchange and process.

The second conclusion is related tot he basic element of this network: data. Its accuracy and reliability depend first on the people who are in charege of search, collection, and checking before final recording in the right databse. Methodology efficiency depends on the combination of two key-elements : quantity and quality of data.

Concerning the legal aspects, the status of SIPAM must be clarified, especially what is concerning the protection of software, databases content and copyright. Letters of agreement have to be signed by the different parts involved in this project (GFCM, Tunisian Government, FAO). Those commitments have to consider also the future commercial aspects of SIPAM activity.

The last conclusion concerns the perennity of this system. As SIPAM member countries and FAO may be considered as “genitors” of this “newly born baby”, they have the responsibility to give it the maximum chance of survival. Hence, the devlopment of this network has to be oriented to the demand of final user who could also provide funds. The cost of the development of SIPAM has to be secured on the long term which implies to designate a small team for the preparation of a precise finance request to the European Community which submittance could be done either through GFCM or through a group of mediterranean countreis. At the moment, an ad hoc group of four representaives (Croatia, France, Greece, Spain) has to prepare a request to the European Union with the support of Ifremer.

Recommendations are consequences of those observations. Exhaustive data entry is now urgent-Procedures for collection or estimation of missing data have to be started. The organisation of a permanent part or full-time national team is necessary in order to secure the steadiness and the professionnality of the participation of each country tot he regional network.

Similarly, the permanent regional centre team has to be secured by the bosting enuntry (Tunisia), with a clear commitment in the charge of the operating cost of SIPAM as defined in FAO proposal (letter of agrement dated January, 1995).

The change from the experimental phase to the growth phase is crucial. It is now under the responsibility of each participant to make this change a successful metamorphosis, not a crisis.

Annex 1

Third Working Group on SIPAM
System Design Development
Irakilo - Greece, 4th – 7th April, 1995

List of Participants

Annex 2

Agenda

Annex 3

Report on the visit to Selonda fish farm
Friday 7th of April, 1995
by Bruno Menu

General presentation

Selonda produced 30 T of sea bass and sea bream in 1987, 70 T un 1988 and 120 T in 1990. Since 1990, Selonda group developed new hatcheries and expanded to 8 new grow-out sites. In 1994, fry production reached 20 million fingerlings and production totalised 1.100 T of commercial size fishes.

This farm is located in a narrow valley ending in a well sheltered hay on the left side of the road between Corinth and Epidaurus. The hatchery is producing 5 millions fry per year for all grwo-out sites of the group. The local capacity of the site is 500 T of production per year.

Hatchery

Its capacity is 5 millions fry/year (1–2 g). Two marine water supply sources are available: one from a drill (16–21°C), re-oxygenated and one from the sea (600–700 m³/km), filtered at 50 microns and UV sterilised.

Breeders allow egg supply from november to June for sea bass and all the year round for sea bream larval rearing in managed in one single hall (40 tanks of 1,5 m³ and 10 tanks of 10 m³) Artemia (8 tanks of 2,2 m³) Rotifers (14 tanks f 1 m³) and algal inoculums are produced in 3 small rooms close to the hall. Algal mass production is done in an external hot-house which can be cooled by bottom circulating water (10 cm depth).

Pregrowth phase

Fingerlings are weaned in larval rearing tanks and start pre-growth phase in 22 big circular tanks located under open side halls until they reached the average weight of 1,5 g.

Grow-out cages

A large aluminium shuttle equipped with a crane allows to transport personnel, feed and equipment to the cages. 3 separate cages lines are producing with one small transfer set of cages. Cage structure is in galvanised metal with plastic floats (15 × 15 × 6m). Water depth is about 12 m. Feed is distributed automatically by air flow through pipes connected to one general silo, ashore. Additional manual distribution allow to control stock quality and behaviour.

Operations are limited to initial sorting, net change and final harvest. The site is not exposed to waves or to algal blooms. The only risk is net sealing by excess of fouling which can ends in anoxia.

Processing hall

After 16 to 18 months, fishes are processed in a specialised hall which seems to be rudimentary when considered the total of production to be processed per year.

Personnel: 35 persons are working on this site (hatchery + 500 T). Selonda group totalises 200 people.

Problems and future development

The main biological problem is infections pathology on sea bass (95% of the problems according to the pathologist). Selonda foresees to start soon a big capacity process unit (4,000 T/year - 3 millions USD). New species (Pagrus, Sole) are or will be tested in order to diversify production and above all, to avoid potential problems linked with a single client as it is with Italy nowadays.

Annex 4

AQUABANK
Data Bank of Aquaculture Farms of Marine Species in Greece
by Nikos Papandroulakis

The project AQUABANK is operational since March 1994. It was financed by STRIDE programme. The available information is concerning the state of development of the Greek Aquaculture. The information system is developed in two parts. The main data base located in Iraklion which is available on-line, and a microcomputer based system containing summary information.

The main information system is developed on a relational data base (UNIX, ORACLE), is multi-user and the access to the information is through a local Ethernet network, packet switched data network HellasPAC or public telephone system. Through the network the information is available to anyone interested. It is also possible retrieval of secondary information like the growth rate of the reared fish according to the location and the season of the rearing as well as the time for reaching the marketable size according to the location of the farm. The users, divided in three categoires, have different access privileges.

The microcomputer information system is developed based on the data of the main system. It is developed on a user-friendly environment (Windows) and does not require special knowledge. The system provides gathered information and gives an overall picture of the present situation of the Greek aquaculture. The system is available to everybody and will be renewed every year with new data arising from the main system.

The data collection was done with questionnaires sent to farmers after informing them for the purpose of the project, and also with personal interviews from specialised personnel. 75 farms covering almost 60% of the total national production and all different types (vertical integrated, on-growing, big or small), were presented on the first data collection. Also, the geographical distribution of the farms was representative of the total. The check of the data was done in two ways:

1. through the questionnaire itself with cross check evaluation of the data.
2. by the local check at the farms.

Another part of the information existing in the database was retrieved at the institute from row data that the producer provided.

The further data collection will take place on regular base (4–5 months) from 10 farms selected from the accuracy of the data they have provided the structure, from the size and also from the geographical location.

Final objective of the project is the integrated recording of the situation of the Greek aquaculture, the consultancy services and the long distance training to the farmers, with the use of new technologies, something that it is essential for the further development of the aquaculture branch.

Institute of Marine Biology of Crete
P.O. Box 2214, 71003 Iraklio, Cret, GREECE
Ph. + 30 81 242022. Fax. 241882
E-mail: [email protected].

Annex 5

Stirling University Experience
by J. Bostock

Introduction

Stirling Aquaculture are the consultancy and project management group within the Institute of Aquaculture at the University of Stirling. We have one foot in the Mediterranean as one of our key projects at present is Malta Mariculture, a 500 tonne per annum sea bream farm which we were instrumental in setting up and for which we now provide technical management. We are hoping to set up a marine fish hatchery in Malta over the next year.

We are also involved in many other projects in Europe, Africa, Asia, North and South America, some of which have required data collection and the use of databases.

The main problems linked with data management are the following :

• we are processors of information - constantly receiving new information which may be assessed, developed and used to advise clients;

• large quantities of information are received, but is mostly unstructured and only valuable if is is accurate and readily accessible;

• key information is often unavailable or is out of date.

databases do not solve all these problems, but they can help overcome the problem of not being able to locate data, and they provide a tool to help summarise and analyse information. Database development and maintenance is expensive however, as most data is time dependent or time sensitive. Any attempt to capture it into a database requires the database to be maintained, otherwise the value of the database may rapidly decline.

A. The role of databases at Stirling.

Most of the work at the institute of Aquaculture is contract base, with relatively short time horizons. This, and perhaps the fact that we have no specific computing section has discouraged the establishment of large databases. Databases are seen as a tool to use when there is a very specific job to be done. Most are very small by the standards of SIPAM Typcial examples are as follows:

As an index to paper-bases information

This of course is partly covered by databases such as ASFA, AGRIS and the University library catalogue. However, these do not cover the many reports, newsletters, magazines and industry brochures that we receive. Key information about each publication or article (e.g. keyworks and reference) are entered into a database to help with subsequent retrieval. Another database keeps track of the many CVs (Resumés) we keep. Again, rather than enter full details about each person, we record only name and keywords on subject and geographical area. A search on say “shrimp” and “India” will give the names of people with expertise in shrimp and experience in India - we can then go the paper CV for more information.

For simple information storage and retrieval

For some types of information (such as names and addresses) databases are the easiest and most flexible form of storage. Most of these use a single data table and are focused - e.g. UK fish farms or Institute Newsletter mailing list, or equipment suppliers. Most are single table databases, although the equipment suppliers database uses three tables to link products and suppliers in many-to many relationship.

For data analysis

Databases are sometimes used to help analyse information gathered during a one-off survey. Examples include water quality surveys, evaluation of training courses, or surveys to gain information on fish farm practices databases have the advantage here of providing a better interface than spreadsheets for data entry and checking, and allowing greater volumes of data tobe input. Modern packages can also carry out much of the required analysis. Otherwise it is simple to export the data to a spreadsheet or statistics package for additional analysis.

For project management

In a simple way, databases are sometimes used to track project, progress, for instance carrying out a questionnaire survey. Each record contains the contact detail for the person to be contacted. Logical fields track the progress. Memo fields allow specific comments to be included, such as notes to phone back at a specific time. The progress of the project can quickly be summarised, and delaysaddresses.

The most complex database developed within Stirling Aquaculture also falls within this category, althgough is also for data storage and analysis. This is a database setup to help manage a fish farming extension program in Malawi.

All of these databases are PC based, although we potentially have access to Networked application, including a Unix bases Oracle serve. Currently we are using Microsoft Access under Windows, with earlier databases developed in dBase III + or IV.

B- Problem with databases.

Databases are often not used because of problems such as:

-   most are not suitable for unstructured information;

-   unless they are very well planned, difficulties quickly arise as users wish to enter data where there is no apporpriate field, or where a numeric value is required, but only qualitative information is available;

-   definitions are misubderstood so data is mis-entered;

-   relational databases can be difficult for end-users to understand and set up;

-   the likely end result is considered to be not worth the investment in time and effort required to design and construct the database.

C- Experience of data collection in Scotland and in Malawi

Wherever possible, Stirling Aquaculture relies on secondary data from all types of publication. Where primary data collection is required this is minimised by asking what do we really to know in order to answer the questions that have been asked? A strategy for answering those is then worked out. Sometimes this involved questionnaires. Most of the time a small sample of people are approached, and the questionnaire is completed via a telephone interview (conversation) or even a personal visit.

Using this personal approach it is possible to achieve an 80–90% response rate. Obviously, obtaining commercially sensitive information in this way can be very difficult, although much depends on the attitude of individual companies, or even individuals within those companies. Developing out of hours social contacts with people in the industry can often be the only way of hearing what is really going on.

For surveys of this kind, budget and practical considerations often outweigh statistical ones. Since up to 70% of the Scottish salmon industry is owned by a small handful of companies, targeting these plus a few smaller operators can be sufficient to give a good coverage.

Government agencies potentially have en advantage in data collection, in that companies can be put under a legal obligation to provide appropriate information to the department concerned. For instance, in Scotland there is statutory requirement for the industry to provide information on production and disease to the ministry concerned.

Data collection on the extension project in Malawi is more formalised than in Scotland. Again a series of forms provide the basic framework for the data collection. These concern things like the registration of new farmers, registration of new ponds, stocking and harvesting activities, visits of the extension agent and pond management practices. Each of these forms follow the database data entry screens to assist with inputting the data to the computer. The forms are completed by the Extension Agents when they visit farmer. They have regular training sessions to ensure they understand how to collect the data and why it is needed.

In order to gather information on pond inputs, Extension Agents often use a sampling technique asking farmers to recall what they have put into the pond over the past few days. Asking farmers to give a good idea of general patterns has rarely been found to give accurate data. Another method which is successful with some farmers is to provide them with a notebook and show them a diagrammatic method of recording pond inputs.

The incentives for farmers to provide data are centred around the advice and occasional assistance with stocking and harvesting that they receive from the Extension Agent. In Malawi Fish Farming Extension Agents are not involved with enforcing fisheries regulations or collecting taxes, so reasonable cooperation can be established.

D- Data verification

This can be one of the most difficult and expensive aspects of database construction and maintenance. Data is only useful if it is accurate. Problems include :

-   missing data;

-   errors due to inaccurate data entry,

-   inaccuracies in the original;

-   inaccuracies due to changes since the data was entered.

There are various methods used at Stirling to try to limit these problems. Missing data must usually be tackled by looking for other sources of the data. This often involves identifying the person who may know the answer and specifically asking them that question. Otherwise it may require personal observation or measurement if very important. Errors during data entry can be checked by a second person comparing entered database records with the original paper forms. Overcoming inaccuracies in the original data can be the hardest task. In this case, the most satisfactory approach is often for a subject area specialist to review the data to spot data which does not fit in with expectations. This may be automated to some extent by applying a simple statistical analysis to the data. If this reveals valuas which are well above or below the norm these can be checked with the source, or with alternative sources. Occasionally there are other published sources of information against which the data may be compared. More sophisticated analysis along these lines may use correlation, or regression or grouping of datas appropriate to identify data that does not fall into normal ranges. It may not be necessary to update all information in the same time interval. Market prices for instance need to be almost constantly updated, whereas data on fish farming companies may only need to be checked once per year.

Eliminating all errors from a database is probably almost impossible, and becomes more increasingly expensive (Probably exponentially) as the goal of nil errors is approached. In every case there will be a trade-off between acceptable cost and acceptable error rate.

E Directions for the future

I intimated at the start that databases are not something with most users at the Institute feel comfortable with, or use to their full potential. However, there is considerable interest now in the Internet World Wide Web. For those not familiar with this technology it is a way of linking documents and databases located on many computers around he world but providing a seamless interface where it is not necessary always to know where any specific peice of information is stored. Free text information or more structured databases are maintained locally, but accessed globaly. People at be Institute are suddenly interested in making data available, and sing this system - something which has never really happened with previous database systems.

The advantage of this approach is that data is maintained locally where there is the best chance that will be accuite and up to date. The disadvantages is that there is much less structure and search mechanisms are still fairly crude. This may chang over the next few years with the introduction of more sophisticated search programmes which will trawl the Internet for answers to more specific information.

There are also serious quetions about data security which need to be addressed when dealing with commercially (or politically sensitive data. However, rapid progress is also being made of this front with secure data transfer mechanisms already with us, and the possibilities of credit and transactions to be carried out just around the corne.

The World Wide Web is not offered as an alternative to the types of databases discussed above, or the SIPAM, but as a new means of sharing information in a way that is robust and can be readily searched and the data integrity sustained. Careful interpretation of information on the World Wide Web will always be required, and verification may still be needed. Like journal articles and newsletters, it will never provide access to everything. However, it is a welcome steptowards helping us with more general information needs.

Examples of information that may be available through the World Wide Web from Stirling University Institute of Aquaculture includes:

-   staff directory and resumés;

-   institute facilities;

-   institute services;

-   current research programmes;

-   course prospectus;

-   institute newsletter articles;

-   short scientific papers on current research;

-   library of images for use in aquaculture extension;

-   short articles to provide information on commonly asket questions about aquaculture development.

Annex 6

Observations on the Legal Aspects of SIPAM
by Liz Heffernan, FAO Legal Office

A- General Legal Framework

1. It was noted that the institutional structure of is currently in a state of transition. With the termination of the MEDRAP II project during the current year, responsibility for the coordination of SIPAM will pass to the GFCM. It is envisaged that SIPAM will come under the operation of the proposed GFCM Committee for Aquaculture Development. Three and related matters will be discussed at the next meeting of the GFCM in May 1995.

2. The legal framework is anticipated to comprise a number of elements. In the first place, agreement among GFCM members will be required regarding the foundation and operation of SIPAM. This will reflect consensus on the objectives and characteristics of SIPAM, and will involve defining such matters as its institutional frameworks and the rights and obligations of the various participants. In trun, it will serve to clarify certain other legal issues, such as the question of copyright ownership.

3. As regards the contribution of the working group, the importance of making a clear presentation to the GFCM of the objective,s characterictics and operation of SIPAM was emphasised. It was also noted that while SIPAM is currently operating solely in the government/public interest sphere, extension to the private/commercial sphere would involve additional legal considerations. The general question of securing financial support for the continued operation of SIPAM was raise, although recognised as matter outside the mandate of the working group.

4. Finally, some comments were made concerning the status of GECM as a body established by an agreement under Article XIV of the FAO constitution. In particular, it was observed that the GFCM does not possess sufficient autonomy to enter into agreements with countries that are not GECM members and that any such arrangements must be concluded by the FAO Director General. This issue may arise in the context of the participation in SIPAM of states which are not GFCM members, such as the current participants, Croatia and Portugal, which are members of MEDRAP but not of the GECM.

5. The legal framework will also comprise an arrangement between the FAO Director General and the tunisian government, which has formally offered to act as host country to SIPAM. This offer is currently the subject of discussions between the Director General and that Government. The agreement ultimately concluded is likely to cover such practical considerations as the physical housing of SIPAM, maintenance, staffing and basic expenditure.

6. Thirdly, it will be necessary to seek legal advice as to the implications of the operation of SIPAM under Tunisian law, e.g., to ascertain wether there are registration or other formalities and to assess the application of Tunisian intellectual property laws.

B- Copyright

7. A brief outline of the law of copyright was presented. Copyright remains a matter which is primarily regulated by national law so that the nature and extent of protection afforded varies from country to country. There have been considerable efforts at harmonisation of copyright laws at an international level principally through WIPO and UNESCO. Bilateral arrangemenjts and, more particularly, multilateral agreements such as the Berne Convertion ande the Universal Copyright Convention (UCC) provide for reciprocal protection and impose minimum requirements, thereby reducing the impact of disparities in national laws. It was noted that most of the SIPAM countries are parties to one or other of these two conventions.

8. The diversity of national copyright laws is particulaarly apparent in respect of new technologies such as computers and databases. Although the application of copyright principles to computer software and databases was slow to emerge, it is steadily gaining acceptance. It is the subject of a provisions in the Agreement on Trade Related Aspects of Intellectual Property concluded in the Uruguay Round of the GATT and of an EC Draft Directive whcih seks to harmonise the nationals laws of the EC member states on the protectionj of databases. At a natinal level, most copyright laws either specifically or implicity regulate databases.

9. In general terms, copyright law gives the owner of work the exclusive right to copy and distribute the work as well as the right to develop and produce derivative works, for a specific period of time. Thus, once a work is protected by copyright, the permission of the copyright owners is required in order to engage in any of these activities.

10. Multilateral arrangements suchas the Berne Convention and the UCC set out certain minimum reguirements or sdgeneral principles on copyright These requirements relate to matters such as defining the ype of work which warrants protection, determining the country of origin of a particular works and establishing the duration of copyright protection.

    Copyright in databases derives fro the general protection afforded to “literary” works. Although under the Berne Convention, copyright is created automaticallty and without the need to complete any formalities, the UCC practice is generally favoured, whereby copyright is established by the use of the copyrights symbol, the name of the copyright owner and the year of first publication, e.g. “c copyright FAO 1995. All right reserved”.

11. The need to define the relevance of copyright considerations to SIPAM was identified. It was recognised that a number of factors may limit or exclude the application of copyright principles. For example, the public interest/non-profit character of SIPAM, the non-confidential nature of the data, ad the limited circulation of SIPAM data to the public, reduce the need for copyright protection. A further consideration is whether the dat (and if so how mmuch of thae data) is actually suseeptible to copyright. Generally, data must be sufficiently “original”, having regard to its content and skill or labout involved, to warrant protection. Thus, statistical, bibliographical and similar data generally fall outside the protection of copyright.

12. A distinction was drawn betwen three copyright issues which may arise in respect of SIPAM. First consideration must be given to the database itself i.e., the access of software. In the case of SIPAM, this software was devised by FAO and is protected by FAO copyright.

13. A second concern relates tod the data supplied to SIPAM. This data may be supplied at a national level to the national centres or, alternatively, may be supplied by national, regional or international sources to the Regional Centre. For most part, it is unlikely that this data will be protected by copyright. However, if it is so plrotected, then it should not be input or stored in the SIPAM, system without the prior authorisation of the copyright owner. Such permission may involved the payment of a fee and the inclusion of a copyright acknowledgement with the data. Thus, any requests for permission to use a third party copyright should be corrdinated through the Reginal Centre. This is necessary given the possibility of pre-existing arrangemetns between FAO and third parties regarding the use of copyrighted data which may be supplied to SIPAM.

14. The third copyright issue relates to the data which is actually compiled on the database, i.e., SIPAM data. There is considerable support for the view that the acts of selecting, compiling and storing data on a database result in a work which may attract copyright protection, independent of any rights in the origina data. Cleraly, this issue is more complicataed than the other two, not least because the data is actually compiled at national and regional levels. In this regard, it is suggested that regional rather than national selection and compilation should be deemed to give rise to SIPAM data. It is this SIPAM data which should be the subject of distinct copyright protection. The identity of the copyright owner must avait the conclusion of the general legal framework for SIPAM, as discussed above. However, given the co-ordinationg function of GFCM, SIPAM data is likely to come under FAO copyright.

15. The need for copyright protection depends in part upon the number and profiles of potential users of the SIPAM database. In this regard, the importance of the role of the national centres in identifying and co-ordinating the use of SIPAM data was emphasised. However, it was also recalled that the prinipal objective of FAO copyright protection is not economic per se, but rather is to secure the widest possible dissemination of the data.

16. Some consideration was given to poteciton future applications of SIPAM data. Commercial uses would involve a change of emphasis in the practical and legal arrangements for SIPAM and, for examplen might ultimately require the introduction of standardise licensingarrangemets. However, there was agreement that any such uses are not contemplated at the current time. It was also noted that copyright protection might also extend to any derivative products produced or devgeloped from SIPAM data.

17. In light of the above, it was recognised that a precise and comprehensive accoundt of the implications of copyright protection for SIPAM would require analysis of the copyright, laws of each of theparticipating countries. Similary, since certain data supplied to SIPAM originates at a national level, restraints on its use for SIPAM purposes will be a matter of for the respective national laws of the SIPAM countries. In addition, any such revgiew would require consideration of other relevant areas of national law, such as contract and criminal law, as well as identification of any special provisions or waivers operating in favour of international organisations or non-profit, activities. As more immediate and practical firt step, it was suggested that advice be sought regarding the copyright laws of Tunisia, the location of the Regional Centre, which for copyright purposes, will be deemed to be the “country of origin” of SIPAM data.

C. Legislative Database

18. A brief overview was presented of the operation of the Fisheries Law Database at the FAO Legal Office. It is still in a transitional phase and is estimated to the fully operational by early 1996.

19. The database is exclusively legislative in character and covers wide-ranging forms of legislation in substantive areas of fisheries law. Aside from the general benefits stemming from the operation of the database, the Legal Office is using the system to develop a system of classification of laws. It offers afar more comprehensive and precise classification of the data than the traditional methods. At the current time, the data is drawn exclusively from within the FAO archives and FAO archives and FAO sources. The keyworks for the fisheries database have been selected and are accompanied by guidelines which provide the user with definitions of keywords.

20. As regards each individual record, the data is provided in English, French of Spanish. Information is presented regarding the type of text or legislation and its legislative history. A brief outline is provided in the form of a summary. For selected pieces of legislation, a more extensive abstract, prepared by legal experts, is available. Although not yet operational, linkage is anticipated, so that the entire legislative history of a piece of legislation will be available on a single page.

21. Concerns were expressed regarding the role of the national centre in dealing with legal data and, in particular, in drafting summaries of national legislation. It was suggested that this latter function will require the use of legal expertise. In response to question it was confirmed that the FAO fiheries law database also covers EC legislation.