The "Encyclopedia of Life" project is an ambitious concept to build an Encyclopedia of species information on the web, rather like WikiSpecies but hopefully with a bit more sophistication. Long before the EoL project was announced, I had been wondering what such an Encyclopedia might look like, and how could it be constructed in a way that scales to the size of the task. The result is something I've called mySpecies, (in homage to Rod Page's iSpecies), and is explained in a bit more detail below.

mySpace + iSpecies = mySpecies

mySpace

The user is at the centre of every successful Web 2.0 product, and the metropolis of mySpace is a perfect example of this. With as series of simple tools, mySpace provides a place to generate and share web content with consummate ease. The result is a chaotic explosion of content that might otherwise have withered in obscurity (sometimes deservedly), were it not for the social structures in place to support and develop the site. Like many of its Web 2.0 forerunners and predecessors, mySpace enables and engages users by providing the fabric of a community that sustains and enriches itself. By placing the users needs first, these Web 2.0 sites scale to challenges that initially seem insurmountable, whether that be the construction of a cosmic compendium of knowledge like Wikipedia, the million channel TV network of youTube or a billion page photo album called Flickr. However, this collaborative approach is not the only way of amassing knowledge on-line.

iSpecies

Without holding a single webpage, iSpecies provides as much information as a good Internet encyclopaedia, but it did not take an army of users to write this website. Instead of housing static pages about different species, iSpecies compiles a dynamic profile of a selected organisms by linking to molecular, taxonomic, and other sites to generate an aggregated synthesis of results that appear on a single web page. However, interacting with iSpecies is a decidedly one-sided affair. It is usually not clear where search results have come from, whether they are comprehensive, or whether they are accurate. More to the point, if you find an error, or are aware of something iSpecies missed, there is nothing you can do about it. You cannot correct errors in iSpecies, and iSpecies does not take requests – it is just a search engine after all. In spite of this, iSpecies regularly handles over 1,000 queries a day, more than a year after it was first announced. But what if iSpecies did accept requests. What if it did allow us to make corrections, and better yet, provide a space to add our data (my data) when I had something to say about a taxon. This is the concept behind mySpecies, a user edited and dynamically updated Encyclopaedia of Life that is seeded from content already on the Web.

mySpecies

mySpecies is a fusion of the principles behind the user-generated content of mySpace and the search engine aggregation of iSpecies. By seeding pages with dynamically generated content from some of the most comprehensive life science databases and websites online (Fig.), it should be possible to generate the conspectus of an Encyclopaedia of Life at the push of a button. uBio’s taxon name thesaurus would provide the initial index against which these pages would be built, and these pages would be available to search and browse as part of the site, with content cached and pages periodically rebuilt. What distinguishes mySpecies from an enhanced version of the current iSpecies is that these pages can also act as a stub to be edited like a Wiki by any user. Users can take ownership of a page by creating their own version of a stub on which they can add, edit or supplement its contents with their own data. They can control who (if anybody) is entitled to edit their version of the page and redefine the context of a page around the original taxon. However, unlike Wikipedia, other users can reuse edited stubs as a new seed to customize their versions of the same page, or go back to the original unedited stub and work from there. Thus multiple pages that are conceptually linked around the same taxon can be created. Users may link to any other page or elements within a page version, effectively building a hierarchy of pages that relate to their bespoke needs, with uBio playing a key role of reindexing edited content and aiding page navigation. The dynamic elements (feeds) serving content to an edited page would remain dynamic, ensuring content remains up to date for these elements, and the user would be able to define custom feeds from other sources serving data in a compatible protocol that relate to their needs.

Stub pages would inevitably contain errors just like many of the edited pages. This is the price paid for the scale that can be instantly achieved with this approach. The signal to noise ratio would be controlled by filters, views, recommendations, links and all manner of statistics that ensure the best pages are the most consistently ranked and thus more discoverable than those of marginal interest. These statistics would also provide the incentive to discourage users from theft of content without adding value to it, since the new page versions would have less views, links and recommendations, and thus be less discoverable through the search and navigation interface. Visualizations of pages representing different user needs would be needed to ensure the site is consistently navigable. These need not necessarily be confined to traditional hierarchical classifications of taxonomic groups. For example, they might represent how users have linked or borrowed content from each other’s pages or reflect real world aggregations of living things, rather than monophyletic or traditional taxonomic groupings, such as the animals in a zoo, bird species visiting a feeder or the parasites of a host.