Hi Dan, I saw your GWU posting on C4L last week and it sounds like you have
a great shop. I hope you are enjoying it.

I really appreciated the brief conversation I had with you at CNI last
month. The SHARE plans are starting to come together with a bit more
specificity, and just last week we learned we could be on a very fast track
for some grant funding. While this is terrific for the project, it does
mean we have to make some decisions without as much clarity as I'd like.
I'm wondering if I can impose on you again for some thoughts on these
decisions? If so, read on.

We now have a project
plan<https://docs.google.com/document/d/1-7zHFvuOmgxv_RYCS8VSE7-3G9XYkUzflw-w1gsI-0E/edit>for
the notification system I described to you in December. While this is
still pretty much an internal document, feel free to look it over if you
like. Essentially, we want to build a system that captures research release
events from a variety of systems from CrossRef and CHORUS to VIVO and local
repositories to systems like RePORT and the Open Science Framework. These
events would represent publications, but also data sets or presentations or
other outputs of the research process. The goal is to build a digest of
these events that can be subscribed to (like "following" on Twitter) or
queried by stakeholders such as offices of sponsored research at
universities to managers of funding agencies (federal and private) to
publishers themselves on the hunt for future authors and editors.

Even though we are still defining the scope of this work and trying to come
to grips with how much is already being done by some of the players, we are
also trying to wrap this up in the form of a grant proposal this week. For
the purposes of this grant, I've asked that we plan for a team of one
development project manager and two full time programmers. We may have a
site in NYC or Virginia willing to host this team, or we may want to allow
the team to be distributed around the country in an effort to broaden our
available pool.

In particular, I'd love to get a project manager on board ASAP, since this
person could help sort through some of the existing projects and open
source codebases to determine from what foundation we would best launch the
development effort.

I believe we will have to start this small, focused on a partner or two and
what we can harvest, how we can structure these "events" and store them for
retrieval and push notifications, then iterate as we add partners and build
toward the full notification system. The SHARE steering group supports all
the code we write being open source (there was advocacy for the Educational
Community License <http://opensource.org/licenses/ECL-2.0>, which I am not
familiar with, but is an offshoot of Apache) and any data we assemble being
CC0. I imagine this would be an agile project managed on github, though I
was particularly intrigued by the degree to which it sounds like your shop
has focussed on github as a management tool.

Whew! That's a mouthful.

I am wondering whether you have any advice given this summary. In
particular, I am wondering if you have any thoughts about who might be a
suited to being a project manager for this kind of effort and whether you
have any wisdom to impart about the choice of embedding staff in a host
institution vs. trying to manage a distributed team? At CNI you seemed
interested in the potential of the project itself, I also wonder if this
still sounds worthwhile to you or what warnings of advice about the scope
of the task itself?

Finally, you had mentioned both a staff member at GWU and a few other names
when we were sitting at CNI, but if I wrote that on a piece of paper there,
I am having a devil of a time finding it (partly because I am on the road
this week). If you have any thoughts about people who would be good for
this development team, let me know. The SHARE steering group imagines this
could grow into something on the scale of I2 or Sparc, so it could have
longer legs than just this notification system too. Of course, that will
depend partly on how well we pull off this first step.

Yikes! Another mouthful.

OK, I'll stop there.  :)  Maybe I should just call. If so, give me a time
and number and I'll try to ring you up.

Thanks!
...Eric


Eric Celeste / [log in to unmask] / http://eric.clst.org / 651-323-2009