Use Case: Which DBLP to use

There are no fewer than five RDF conversions [1] of the authoritative DBLP XML dump [2]. (If you look at the list carefully, there are 4 -- there's one more newborn conversion on my hard drive waiting to be published.)

One hasn't been updated since 2007 [3] and defers to two others in its comments:

• [4] http://datahub.io/dataset/rkb-explorer-dblp
• [5] http://datahub.io/dataset/l3s-dblp

both of which describe the the latest 2001 paper that I'm doing an analysis with.

The fourth [6] provides a description and some code [7] about how theirs was created, but their final dump file and URIs 404 and they stopped updating in 2008.

The end of my story actually fizzles out, since [4] and [5] provide what I need, so I can hopefully reuse theirs instead of needing to reproduce their results from scratch. But that fact remains that four efforts were made to transform one dataset, which seems to oppose the "convergence story" that semantic web usual cling on to.

Are there some recurring factors in this very simple example that we can use to highlight some of the issues that we think are important for the community?

Regards, Tim

p.s. I'm still trying to figure out if I can help co-chair...

[1] http://datahub.io/dataset?q=DBLP [2] http://datahub.io/dataset/dblp [3] http://datahub.io/dataset/fu-berlin-dblp [4] http://datahub.io/dataset/rkb-explorer-dblp [5] http://datahub.io/dataset/l3s-dblp [6] http://datahub.io/dataset/sweto-dblp [7] http://knoesis.wright.edu/library/ontologies/swetodblp/