In order to ensure that research service output remains findable to users both internal and external, this proposal describes principles and supporting methods to underpin future development of our web services. This is a ‘target model’ rather than a prescriptive solution or a roadmap for development.
Couple editorial strategy with website content strategy.
Build pages for both users and search engines: every page is a landing page.
Be clear about the difference between content formats and ‘products’.
Clearly express what our ‘products’ are: why would people choose one over another?
Provide multiple routes to content.
Design for accessibility: for people and for machines.
Allow for discovered user journeys.
Ensure that each concept has one description in order to group related information.
Continue to support shop fronts for the two libraries and POST.
Design for collaboration between Library sections and between research services.
Credit researchers by name and by link. Always with informed consent.
Maximise the value across the lifecycle of our content.
Design URLs to provide findable content.
Maximise search engine value by exploiting aggregations that build over time.
Integrate with the wider web.
Deliver the ability for our content to be consumed beyond the website.
Design for transparency and accountability.
Establishing a shared domain enables us to publish content allowing for future collaboration between research service teams.
Allow for aggregations of research output - whether by topic, geographical extent, author or format.
Aggregations provide landing pages for search engines and users, allowing the full breadth of our content to be exposed.
Maintaining shop fronts for the individual research services allows for different branding guidelines and tone of voice.
Maximising routes to content to serve a wider audience in a way that’s scalable, consistent and can be abstracted away from manual curation. How much we link to content shows the web - and search engines - how we assign value to it.
We should see the website as a graph of interrelated information, rather than as a navigation tree or silo. Focus should be given to how pages relate to each other.
Allow authors to aggregate their output to build more entry points, more links to content and link to their research elsewhere.
Understand and document editorial strategy. Information architecture is emergent from content strategy. Content strategy is emergent from editorial strategy.
Design URLs to maximise persistence, consistency and predictability.
Focus on optimising for search engine indexing, making internal search easier to implement.
Concentrate efforts on things that improve both internal and external search. Driving information architecture through information management and maximising link density to content will help with this.
Openly licence our content so it can travel freely across the web.
Provide data feeds so third parties can re-use our content.
Provide event information in a format that can be read in calendars.
Publish data used to write research briefings as non-proprietary files.
A work in progress, target sitemap for the House of Commons Library, the House of Lords Library and POST:
Interface classes stitching these models together to follow.
Robert Brook