Schlagwortarchiv für: trust and confidentiality

ADA-EDA: A distributed system supporting the development of agriculture 4.0 under the constraints of trust and confidentiality.

How should confidential information be transmitted between information users in the economy and more generally in society? The paper of our author, Alain Sandoz, examines a set of conditions under which this question can be consistently answered.

The need to manage and to regulate the transmission of confidential information arises in the presence of two types of actors: on one side, information providers (e.g. farmers) and on the other, information consumers (e.g. public or private database operators that collect and manage information on farms).

We assume an information provider trusts two or more information consumers with some identical piece of confidential information. The information is confidential in the sense that it is sensitive and the information provider wants to control its dissemination. Under what conditions can the first consumer to receive that information be allowed to transmit it to others that request it? If this is possible, the information provider will not have to repeatedly send the information to different consumers (sending information can be a burden). Moreover, globally the information in the system might end up being more consistent. In some complex situations the information can be composite, each piece being trusted to some consumer, and the whole to some other. Situations of this type occur traditionally in medicine, in education, in industry, and in agriculture for example. The paper focuses on the latter, and more specifically Swiss agriculture where this is currently a hot topic.

Introduction

Swiss farmers produce goods and services. In general terms, goods are edibles or other products grown and processed under well-defined conditions (respectful of animal welfare, organic, etc.). Services can be the contribution of farmers to the environment (bio-diversity), to the landscape, or to the preservation of cultural heritage, for example.

A farm is a business and a farmer is an entrepreneur. A farmer can increase potential revenues by registering to labels (which bring a premium on products), by making agreements with buyers, and by receiving payments from the Federal government and/or cantons in exchange for services.

Some of these measures are defined in the law, others in contracts, others still by actors in the agro-food market. In order to manage and to remunerate the flow of goods and services, information has to be provided along the associated logistic and value chains. This information mostly concerns the farmers and their property and is sensitive: indications on quality, prices, quantities, results of controls by the contractual partner on the truthfulness of provided information, etc. are confidential. In this case, information is mostly collected, processed, and stored in the form of structured digital data, so that at any time several datasets coexist in different databases that partially and sometimes redundantly describe different aspects of a farmer’s property and business.

As an example, consider a Swiss farmer who produces wheat under some specific label, say organic (Bio Suisse) or integrated production (IP-SUISSE) . The farmer supplies the label organisation with data in order for the latter to plan production, distribution, and controls, and to charge the farmer for its services (such as marketing campaigns related to the label, that support the product’s price on the market). If the label organisation orders the farm to be independently controlled, some of this data will be collected again together with other attributes by the controlling organisation. The farmer also supplies the canton with similar data, in order for the canton to decide on the amount of direct payments this culture entitles the farmer to receive. Again an on-farm control is possible (collecting more of the same data) and in any case the canton will transmit the dataset to the Confederation, which will then manage its copy on its own. The data might additionally be supplied to the wheat producers’ professional organisation (which defends wheat producers political interests). The farmer might also produce maize, vegetables, forage, different breeds of animals, milk, etc. For each type of production, different sets of data will be required from the farmer by different actors.

We call the farmer an information provider (noted IP below). The different actors or organisations that require data from the farmer are called information consumers (noted IC). So information providers must often repeatedly make the same information available to different ICs, possibly in different forms and at different times of the year.

Information consumers store information in databases in the form of structured digital data. Unless they consent to a collective effort (like the cantons and the Confederation in the example above), there is no reason nor any means for different consumers of the same information to use identical data structures.

To summarize, each IC requests information from the IP in the form of structured data on some predefined authenticated electronic channel (front-end). For the IC, this has three positive effects: (1) the IP is identified and authenticated on the IC’s system; (2) the information is delivered in a usable digital format and values can be validated (or invalidated) using application logic; (3) the procedure provides a technical means for the IC to register from the IPs their commitment to the information they have delivered. Information transmission and delivery is fully traceable.

In this situation, ICs have no interest to change their information provisioning procedure, even though new demands on new information related to the farmer increase over time. New digitally supported production tools also supply the farmer with an increasing quantity of data, in turn generating new information demands from ICs.

With the increase in the number of ICs and of digital sources over the past 20 years, the problem of data transmission from farms to all types of ICs has become acute in Swiss agriculture.

This situation has led to the creation of the ADA-EDA project. ADA stands for “Agrar Daten Austausch”. EDA (Echange de Données Agricoles) stands for the French translation of the German acronym ADA.

The paper describes the solution that ADA proposes to the problem of data delivery to multiple ICs. Not surprisingly, the approach can be generalized to other application domains under specific conditions and requirements.


Full Article

You can read the full article here.


Hintergrund

Wie geht man erfolgreich mit hoher begrifflicher Komplexität um, die wächst? Wie bringt man die hoheitliche Datenverwaltung mit der kommerziellen Datenbewirtschaftung zusammen, so dass die gesetzlichen Vorgaben und Bedürfnisse von beiden erfüllt werden?  Wie baut man ein robustes adaptives System für das kontrollierte Weitergeben von Informationen? Das sind die drei Schlüsselfragen, die sich im Projekt «ADA» der Agrosolution AG stellen. ADA baut eine Datenaustauschplattform für die Landwirtschaft und wird von Alain Sandoz geleitet, einem der führenden Schweizer Spezialisten für strategische IT-Innovationen. Ein Team rund um Prof. Dr. Reinhard Riedl (Projektverantwortung) und Prof. Dr. Alessia Neuroni (Projektleitung) begleitet das Projekt wissenschaftlich und erarbeitet die Spezifikation der Lösung. Diese wird als Free Open Source Software zur Verfügung gestellt. Der Beitrag entstand in Rahmen der Zusammenarbeit mit der Berner Fachhochschule. Das BFH-Team beteiligte sich als Sparringpartner in der Erstellung des technischen Beitrags.

 

Creative Commons LicenceCreate PDF

Ähnliche Beiträge