How do JGI Data Release Policy and Bleeding Edge Bioinformatics Fit Together?

Apart from technological novelties, our previous commentary on Quip, Minia, SlimGene and Titus Brown’s paper on Scaling Metagenome is interesting for another reason. You can get the sense by reading a discussion from Titus Brown’s blog. Professor Brown and colleages posted their papers at arxiv.org six months before it got accepted in a mainstream journal, and then the authors of Minia leveraged on the supposedly unpublished paper, and ‘unpublished’ an improved approach.

Every time we download a genome from JGI, it makes us go through a long list of agreements. An example of the standard agreement is posted here. Here is an extract -

JGI policy is that early release should aid the progress of science. By accessing these data, you agree not to publish any articles containing analyses of genes or genomic data on a whole genome or chromosome scale prior to publication by JGI and/or its collaborators of a comprehensive genome analysis (Reserved Analyses). Reserved analyses include the identification of complete (whole genome) sets of genomic features such as genes, gene families, regulatory elements, repeat structures, GC content, or any other genome feature, and whole-genome- or chromosome- scale comparisons with other species.

How is ‘publish’ defined in the fast-moving era of arxiv.org, blogs and twitter? Am I ‘publishing’ something by placing a preprint on arxiv.org? Is writing on a blog about a set of interesting genes based on large-scale analysis of a JGI sequence banned? Or is the agreement limited to mainstream journals directly or indirectly funded by bankrupt governments?