Yesterday, I attended a symposium sponsored by Digital Science, Harvard, Microsoft, and MIT, called “Shaking It Up: How to thrive in – and change – the research ecosystem.” I made the trip to attend this event in person because I am focusing some attention on serving the sciences right now, and the sessions featured a remarkable array of mostly new initiatives in support of scientific research and scholarly communication.

The opening keynote featured an appropriately pointed but ultimately inspirational talk by CrossRef’s Geoffrey Bilder, who spoke about funding and sustaining cyberinfrastructure. Scientific research tends to be funded in a series of grants to individual researchers or research groups, and much infrastructure is purchased – or leased – through those grants. Imagine, Geoff urged, if physical infrastructure were funded in the same way – when your grant runs out, the lights go off. Geoff argued that a lot of technical jargon such as “open,” “distributed,” “framework,” and even “fabric,” are worrisomely misleading, used when a lack of trust prevents us from actually building the systems needed to accomplish the task at hand. When you “distribute” something, Geoff argued, someone else may centralize it, co-opting and enclosing the whole thing. And so the talk closed with a set of recommendations about how trusted entities can build and manage infrastructure, emphasizing transparency, a “living will” in case of systemic failure, a goal of generating a surplus allowing for reinvestment, and an incentive structure to wind down operations once the mission has been fulfilled.

Geoff’s argument might apply to national infrastructure like highways, airports, and the electric grid, but he is especially interested in the development of academic infrastructure like libraries, shared scientific instruments, ORCID, and HathiTrust. Some of the basic tenets of Geoff’s argument resonated for me with Ithaka S+R’s sustainability work, and in particular our definition of sustainability as “the ability to generate or gain access to the resources – financial or otherwise – needed to protect and increase the value of the content or service for those who use it.” It would be interesting to consider whether this definition could carry over to the types of infrastructure Geoff was speaking about, or if another type of definition would be needed.

Subsequent panels featured a variety of innovative services in support of scientific and scholarly research, many of which are in alpha or beta stages of development. These panels were held as lightning-rounds, so I’ll share only a few highlights.

Tiim Gardner spoke about Riffyn, a system designed to eliminate noise in data collection and thereby vastly improve the quality and reproducibility of R&D. Riffyn allows scientists to harness all their research equipment and tasks into a managed research process. It treats this research process as a designable object on which they can collaborate readily. With Riffyn, they can harness all their research equipment and stages into a managed process. The introduction of quality practices in biotech research processes is designed to increase productivity and reproducibility dramatically.

John Lees-Miller spoke about Overleaf, a well-used service for scientific writing in the cloud, somewhat akin to a Google Docs for scientific papers, but with special features to manage figures, equations, and article-like typesetting. Recently, they are integrating with innovative scientific publishers such as F1000, FigShare, and PeerJ, so that editorial and review processes can be cloud-managed in this native format.

Rob Siegel introduced Publiscize, a platform to help scientists broaden their audience and impact by creating jargon-free synopses of their works. While there are a growing number of services trying to help the underlying research articles maximize their impact, Publiscize appears to offer the ability to create complementary publications to reach alternative audiences.

Alex Hodgson presented ReadCube principally as an access service to address the increasingly well-documented shortcomings in the ability of academic libraries to provide seamless access to all the content that scientists require. (I appreciated his language in offering a critique of ILL’s “latency period.”) Integrating with publishers, ReadCube allows a researcher to purchase access to an article at rates such as $6 for 48 hour rental, “cloud access” (ie no downloading) for $15, or $35 to buy a PDF. He also showed an option for a library to pay some of these fees directly on behalf of its users as a sort of demand-driven access to materials.

William Gunn shared some of Mendeley’s new developments in the field of research discovery, arguing that pull methods (such as search) are reaching their limits while recommender-driven push methods (which I have elsewhere characterized as anticipatory) have great future potential.

Check out the hashtag #ShakingItUp14, and I understand recordings will be made available shortly.