Evolving Bits

On March 14, 2008, the website R-phylo.org and the Special Interest Group (SIG) mailing list R-sig-phylo launched. Or, more precisely, their existence was announced on several channels (EvolDir, Ecolog) hopefully reaching as many as possible of the community that we have created the site for: developers and users of phylogenetic and comparative methods in R. R is the open-source clone of S/S-plus, and a powerful statistics platform and programming language at the same time. As Samantha Price, a postdoc at NESCent and one of the driving forces behind the site, writes in the announcement:

The wiki currently contains information on analyses and packages available in R and provides step-by-step tutorials for those just getting started.  We strongly encourage new users, experienced users and developers to both use and modify this resource to create a thriving comparative methods in R community.       

The site itself has actually been up since several weeks. After the documentation on how to do comparative analyses in R, which was created at the Comparative Methods in R Hackathon in December, was moved over, Sam Prince, Brian Sidlauskas, and Brian O’Meara, all postdocs at NESCent, worked on polishing up the initial content, determined to formally launch the community on the Ides of March. Sam is also the one who created the logo - I wish my own logo designs were only half as good!

The site right now is a wiki. The hope is that it is not only for, but eventually also by the community. Everyone can create an account for themselves and edit or create content (anonymous edits are disabled at present, though). And yes, in case you were wondering, despite the uncommon style (or ‘skin‘ as it’s called) it is indeed a MediaWiki. The URL layout of the site (i.e., putting the wiki under /wiki) allows to add other components to the site later, such as a blog (any volunteer writers?), or static content, or services.  

The community mailing list, which we decided to make a special interest group (SIG) hosted on the same server as other R Project SIGs, has actually been up even longer, namely since the beginning of the year. The interesting thing, and a strong indication that we’ve really hit a need here, is that seemingly by word of mouth (or web sleuthing?) a number of scientists from various institutions have discovered and subscribed to the list in the meantime, and two discussions involving some of the preeminent experts on the topics already took place without any formal kick-off. I hope that’s a good sign for this becoming a lively and self-sustaining community over time.  

Aside from signifying the beginnings of the first resource of this kind for this community, the launch of the site is also noteworthy from another perspective. We are using it as a pilot for moving towards decentralized, user-centric, globally unique user identity and authentication, in this case using OpenID. We all know the problem of proliferating user account/password combinations too well. For R-Phylo, using the “Login with OpenID” link in the upper right corner you can use an existing identity of yours (provided it’s an OpenID) as login and authenticate against your OpenID provider. For example, I (re-)use my claimID identity.

I’ll be blogging separately about our ideas and the experiences we make employing this technology on a larger basis; global user identity is becoming increasingly important for us not only to control the number of user accounts we ask our numerous scientific visitors to create when in all likelihood they already have more than their share of identities when they come to us, but also to lay the basis for authenticating data access and querying authorization across databases and digital data repositories.