Recommendations

In this toolkit, we have argued for clear, consistent, readable and accessible URIs: Cool URIs. Cool URIs don't necessarily assume that the resource being represented is fixed, stable or permanent, only that the URI persists, identifying a resource (or more likely a collection of resources) that we acknowledge may change like the weather.

By proposing our ideal model, we are not expecting each institution to redesign the content of their websites, but rather we have highlighted where the institutions in our study could redesign and remap an improved underlying URI structure. This is an important distinction: the design of an institutional URI structure has little to do with the resources as are they are presented to users in their web browsers. Website design is not the issue here. URI design is. Although we are calling for a much more imaginative, experimental and ambitious use of the .ac.uk domain, if planned well, work on institutional URI structures could take place without anyone actually noticing thanks to HTTP 3xx status codes. This alone would improve the opportunities for data aggregation and re-use and the development of services built on simple APIs.

An abstract space that is perfectly organised relies on the agreement and co-ordination of a number of technologies (e.g. networking, web servers, browsers). In our physical spaces, technologies work together so as to ensure the appearance and experience of order and continuity. The same applies to the design and development of our cyber spaces when there is an appreciation of how each technology can conceal the complexity of the other. For example, web servers can hide file extensions and seamlessly map one resource onto another, URIs can be constructed according to a grammar that is meaningful over time, browsers can integrate search features into the location bar and so on. The abstract space of the URI should be thought about and designed so that it is not only perfectly organised but persistent (trustworthy, reliable) and offers us a constructive grammar of signs which can be understood by human and machine.

The Learning Landscapes project asserted that the three fundamental qualities of good design are efficiency, effectiveness and expression. These are communicative attributes, describing an ideal space that is shaped by a number of recommended principles that are broadly applicable to cyberspace, too. When thinking of university cyberspace for the Linking You project, those principles on page 46-7 of the Learning Landscapes Final Report may be reformulated as:

In essence, much of this could be broadly understood as 'digital literacy', where the learning landcyberscape is designed to engender capable, confident and critical individuals engaged in research, teaching and learning, so that they are active producers of their own social world. It feels like the work required to really understand and take advantage of our new university space-time has hardly begun.

In addition to these principles, the Linking You project has its own recommendations, too.

1. Shared vocabulary

The first is the need for a shared vocabulary - one that is understood and used by academic and managerial stakeholders within universities and across the sector. One of the key aims of the Learning Landscapes project was to "provide a clearly understood vocabulary within which the future development of the University can be articulated, in order to better inform the design of the built environment of higher education." This same approach is common across many other domains, not least that of web architecture, and our comparison of 40 institutional websites has identified where there is already agreement around a shared vocabulary and where there is divergence. In short, we'd like to see the data model in our toolkit be developed into an ontology for the HEI sector. We've made a start with our model, but we'd like to see a national working group that includes academics, professional services and students.

2. data.ac.uk

Second, we'd like to see this proposed vocabulary inform and be informed by the work being done on data.ac.uk, ideally adopting the ontology as the definitive model for organising data.ac.uk. This would not only demonstrate that the ontology has formal backing but also encourage contributors to data.ac.uk to use the ontology in their own institutions.

3. UK HEI data objects

Third, we think there's mileage in scoping out and prototyping an open directory of UK HEI data objects e.g. courses, people, research centres, locations, etc. This could act as a halfway house for institutional web managers not able or willing to contribute to data.ac.uk but do wish to publish their URI endpoints for the convenience of developers and researchers. Our comparative study of 40 institutions is a start (and we intend to add to it), but something more sustainable needs to be put in place.

4. Migration

Fourth, attention needs to be given to the way institutions transition to a shared ontology for the sector. Research needs to be done that examines and recommends strategies for migrating from existing and legacy URI structures to a model of best practice. HTTP 3xx status codes are at the heart of this.

5. Further study into institutional websites

Fifth, we think that the work of the Linking You project could be extended to look at the use of other areas of institutional websites. While we concentrated on our 'corporate' website, we recognise that this is just one aspect of university cyberspace and more needs to be done to understand what makes up the whole. This could be tied in with a study of institutional use of CMS products, which would give us a better appreciation of the technical barriers and opportunities for change that are available to HEIs.

6. Understand organisation challenges

Sixth, to complement the technical focus of our recommendations, we need to better understand the organisational challenges involved, too. We therefore recommend that work is undertake to survey, interview and case study web managers' and other stakeholder attitudes around the use and 'value of URIs'.

7. Complete our study of institutions' websites

Finally, a couple of extensions we've given ourselves on this project: One is to further our study of 40 institutions to complete the national picture of over 100 HEI websites. We began the project by thinking we'd compare our own URI structures with a couple of other institutions, then we increased this to looking at 20. When we hit 20, we realised that 40 wouldn't be too difficult to do and might give us more confidence in our proposed model. Now, having hit the deadline of the Linking You project, we're still keen to do more and intend to complete a national comparative study before the end of the academic year. Secondly, we also intend to reformat this toolkit as a Briefing Paper and send one to stakeholders in every institution along with a poster of the proposed model to stick on their walls.

Discuss This

Comments powered by Disqus