Intercept: Proposal for infoDev activity

Data and data flows

Reviewer 1: There is little mention of the role of metadata and the essential feedback loop which this service will facilitate through its relationships between in-country data sources and international information networks.

It is perhaps useful to comment on the concept of "metadata" in relation to the needs of sustainable development. The concept was first developed in response to the need to reconcile data sets of different agencies or derived from different scientific surveys using different methodologies, or the challenge of bibliographic searches across (library) information systems based on different classification philosophies. Technical solutions have been developed to bridge across data sets. In this sense, metadata has an important role to play in facilitating searches for data in support of development processes. We are familiar with the GILS and GELOS programmes and take account of the types of standards they are seeking to promulgate.

The internet is providing another type of popular "meta-information" structure which UIA and WCMC are experimenting with in a sophisticated manner in the use of hard and soft links at the data element level. And as discussed above, the structured knowledge bases of these organizations have significant metadata characteristics. These will feature in the forthcoming on-line version, currently in beta-test mode. The scope will continue to be developed.

However, the focus of this project is not on "data" as such, and to a significant extent it is not on "information" – or "meta-information" as is conventionally understood. Our project is concerned much more with "knowledge" as "meta-information", or preferably with "meta-knowledge" and the "patterns that connect". This is consistent with our interpretation of the theme of the World Bank conference on Global Knowledge (Toronto, 1997).

The concern of this project with knowledge is evident in the concern with how complex patterns are to be detected and presented with computer assistance -- beyond the capacity of standard statistical graph packages. It is interesting that there is a burgeoning investment in "information visualization" software products. We contend, such presentations need to hold politically significant operational issues such as conflicting opinions (of different constituencies and factions) of the significance of information and on how data should be interpreted and rated as relevant or irrelevant. Policy-makers require a context that interrelates such conflicting perspectives. It is to this end that the project is focused.

Two-way flows

Reviewer 1: Promotion of consistent, two-ways flows of information / knowledge are essential if data quality is to improve.

As noted, this feature has been built into the on-line system that will shortly move out of its test phase. A significant part of the project will focus on the protocols and editorial logistics of managing user comments and facilitating access at appropriate levels of accreditation -- without alienating those whose views whom some may consider irrelevant or out-dated.

Reviewer 1: Again, this will require the successful reconciliation of commercial with non-profit concerns.

This is certainly the case. It is obvious in the transaction costs associated with licensing certain types of software. It is more obvious where a commercial service imposes a significant cost on users of data over which it has a monopoly or holds under copyright.

Commercial and non-profit data

It is useful to acknowledge that a situation may well emerge in which much information of significance to development may be held by commercial services whose priority is to ensure a maximum return on investment inconsistent with the needs of developing countries. This is a well-established pattern.

This project proposes to address these challenges in three ways:

1) Where appropriate, referring users on (via hyperlink) to commercial services holding such information, and leaving it to them to follow that path and to negotiate the access transaction. However, as suggested above, partnership arrangements may be explored with certain 1services to reduce the financial and administrative hassle of this procedure for the user.

2) Developing the strengths of a contextual information system across the full range of human endeavour and concern; this is to offer policy-makers access to larger patterns that cannot be detected through specialized, restricted systems that are invariably not open to constant self-renewal.

3) Engaging the power of the internet as a public knowledge base, enhancing its interactive characteristics and receiving new materials into a structured but flexible knowledge architecture. In this way, non-profit concerns can have access to content improved through their own inputs and efforts, made feasible because there is a framework and process to use to do so.

Reluctance of data providers

Reviewer 2: The greatest problem in developing functional information networks (as WCMC knows all too well) is the reluctance of data providers to provide data.

There are various reasons that "data providers" may be unwilling to release the information, and we are trying to address all of these. Firstly we need to move towards a true network approach, where the data provider can provide access to his data without it being sent elsewhere. Secondly we need to ensure better information within the database on source and quality of information. Thirdly we need better accountability of how the information is being used.

Data maintenance and quality

Reviewer 2: The constant information problem in the developing world is that the baseline information is not being kept up to date and is not available at a fine-enough scale - these fundamental problems are not addressed in any way under this project (and nor could they as it is a completely different scale or problem). It is worthwhile to note that this problem of up-to-date and appropriately-scaled information has plagued WCMC efforts to introduce similar types of tools (at a much simpler level) at the World Bank - there was little demand for a product like the Biodiversity Map Library even in an institution that would theoretically need that kind of information every day...

The reviewer acknowledges that this project does not have a principal objective of improving data management wholesale. As made clear in the proposal and, if not there, certainly above, this project is concerned with providing a context for data and information, knowledge structuring rather than information management per se. However, we would add that within the domains of the partners who contribute information to this project – UIA and WCMC – every effort is made to keep information up to date, using a range of scales and hierarchies commensurate with the accuracy and specificity of the information.

One principal focus of the INFO2000 project is to design systems and processes that can handle incommensurate information. We believe that non-standard information is a fact of life and that people can and must be enabled to use mixed information. It is currently unrealistic to expect up to date information from the highest to the lowest scale. But taking the reviewer’s example, we are not sure why there is a problem providing the age and scale of the information are known.

Information can only be up to date and at a scale for local use when they are maintained adequately at the local level. In many parts of the world they are currently not. The first point is about information services, the second is about capacity building. What we need to do is to build the link between the two. This WCMC and DA are doing as part of their other work.