scholarly cyberinfrastructure - Geoff Bilder - #shakingitup14

Raw notes on presentation "Mind the Gap"  by Geoff Bilder (@gbilder) at Shaking It Up 2014.

Scholarly cyberinfrastructure

Bilder uses ORCid as an example of some success in this area.

Ideas developed with Cameron Neylon & Jennifer Lin from PLoS

wikipedia - cyberinfrastructure

need to transcend individual institutions

can't actually do simple things, let alone grand transformation

there is a reluctance to address infrastructure

1) infrastructure is boring
2) associated with huge bureaucracy

anti-pattern: when we talk about infrastructure
funder invites - future of scholarly communication
but researchers don't want to think about infrastructure

funder -> money into A (infrastructure)
researcher -> studying B (research)
researcher tries to overlap A and B

but research projects don't actually produce sustainable infrastructure

when the grant runs out, the server goes down, and the "infrastructure" they built disappears

cloud makes this even worse (not even a hard drive in an office somewhere you can recover)

"open", "distributed", "lightweight", "framework" makes things sound reasonable

Distributed begets centralized, because otherwise it doesn't work.

People trying to punt difficult decisions about infrastructure.

If you don't acknowledge from the start that distributed will be centralized...
you don't realize that you can be co-opted

minimal distributed - need many shims and adapters

trust issues

(there were) big problems with trust and ORCid (that had to be addressed)

ORCid initially CrossRef author DOIs

approach to address trust: principles

10 ORCid principles
http://orcid.org/about/what-is-orcid/our-principles

what would a scholarly infrastructure organisation be like?

coverage
transcending disciplines, geography, institutions, stakeholders

have to get beyond discipline silos

geography is an issue because funders are national

governance
should be stakeholder-governed
non-discriminatory membership

transparent operations (within constraints of privacy)

organisation cannot lobby
- because people fear organisations that just try to keep themselves alive

infrastructure organisation should have a living will

formal incentives to fulfil mission & wind-down

sustainability

time-limited funds are used only for time-limited activities

goal to generate surplus
- breakeven is an incredibly fragile position

goal to create contingency fund to support operations for 12 months

revenue based on services, not data
- need to ensure data remains open

revenue generation should be consistent with the mission

insurance
- open source: forkability

open source
open data (within constraints of privacy)
available data (within constraints of privacy)

insurance policy:
(with open data and open source) if you can bring the community with you (to a new location/organisation), you can win (control of your information and system)

patent non-assertion

Geoff says: if I were a funder, I would look for these principles

need to build foundation

stop asking researchers how to fund infrastructure

get people who are good at infrastructure to do infrastructure
get over the fear of directly funding infrastructure

thanks to
@cameronneylon
@jenniferlin15