Mapping Hacks

I enjoyed this talk at FOSS4G, World Digital Library: Designing a Multi-lingual Geographic Search Interface. It was given by an Human-Computer Interaction specialist who offered a snarky commentary on search interfaces designed by programmers for geographers.

Considering the insurmountable opportunities, we come up with some “pretty darn flighty ideas” about how spatial/temporal data search interfaces might work. A key theme is ‘narrowing’ focus and trying not to present too many options at once. He had a short list of tips for keeping things simpler:

• Avoid overlap (visual, cartographic, functional)
• If there’s a timeline and a map, the two should always synchronise, each reflect canges in the other straight away
• Preview info at a location with a thumbnail pointing to it, with very brief description
• We see a lot of “panel for selection options on the left, map on the right, timeline below the map on the right” interfaces
• Have a ‘title bar’ rather than a select list for gazetteer-driven query

And it’s got to be beautiful; especially if it is part-brochureware, intended to gain or renew the interests of funders. The World Digital Library is a project founded by UNESCO and the Library of Congress in association with a network of national libraries worldwide; but this development and promotion phase was funded by a $3M grant from Google, Inc.

I’m interested in multi-lingual storage and query, which often seems overlooked in software, but didn’t figure out much new. I learn from the WDL discussion paper that “In order to qualify as credibly “multilingual,” the WDL interface should be offered in the six official UN languages (Arabic, Chinese, English, French, Russian, and Spanish), plus Portuguese.”

Other statements in this paper got my goat, with regards metadata, data domain models, and structured data search. The highlights:

Search should be through a simple Google-type search box. Advanced search capabilities also will be available. … Metadata schemas and their formats and structures thus vary from one stage to the next, generally increasing in complexity … The architecture will adopt commonly used standards and best practices … This characteristic will enable effective search within and between diverse collections in ways that heterogeneous information sources could not afford. Uniformity of formats and structures within the repository will enable global search.

I could go on about the overpromise of “data harmonisation”, and how this looks a bit like the National Mapping Agencies legislation their kind of process onto the many smaller agencies producing and maintaining geographic data, in a way which is not culturally sustainable or technologically necessary - but I will save it for another time.

One of the project’s initiators talked of “how to create metadata (the online equivalent of cataloging) and the interoperability that can create a unified and usable online library that is multimedial and transcultural” as the biggest challenge for the project, after copyright and other rights assurance. I look forward to seeing more of their work on this, on multilingual metadata for partially spatial public domain data stores.

To me, this fine system architecture diagram for a distributed repository, showing the the central system synchronising with a network of mirrors and processors - this system architecture diagram is worth N thousand words.

Posted in data, book, metadata, search, libraries |

You can follow any responses to this entry through the RSS 2.0 feed. Trackback from your own site.