Creation Review

Type
Creation
State
Successful
End Date of the Review Period

Reviews run for a minimum of one week. The outcome of the review is decided on this date. This is the last day to make comments or ask questions about this review.

Proposal

Eclipse Public Access Submission System (PASS)

Tuesday, February 9, 2021 - 15:42 by Sayeed Choudhury
This proposal is in the Project Proposal Phase (as defined in the Eclipse Development Process) and is written to declare its intent and scope. We solicit additional participation and input from the community. Please login and add your feedback in the comments section.
Parent Project
Proposal State
Created
Background

We developed PASS as part of the launch of Johns Hopkins University's (JHU) open access (OA) policy with funding from JHU's President and in partnership with Harvard, MIT, and Jeff Spies. JHU was eager to differentiate itself from other institutions that launched OA policies previously. PASS was designed specifically to align federal funding agency public access and institutional open access policy compliance while reducing burden on researchers. Through our partners, we connected to NIH (via the National Library of Medicine) which was interested in integrating such a platform with PubMedCentral.

Scope

The Eclipse Public Access Submission System (PASS) is a platform to assist researchers in complying with the access policies of their funders and institutions. Eclipse PASS is a web-based platform that allows US researchers to simultaneously submit their Open Access manuscripts into multiple repositories, while seamlessly complying with their various access policies. 

The initial scope of PASS included simultaneous deposit of articles into PubMedCentral and JHU's institutional repository ("JScholarship") based on DSpace and packaging of related information such as articles, metadata, grants, and identifiers (DOIs and ORCIDs). We integrated PASS with JHU's grants system, Coeus, which is used by other institutions and developed dashboard capabilities that allow researchers to view status of submissions. Even with this initial scope, we designed PASS to eventually integrate generally with other institutional and federal funder repository platforms, grants systems, and deposit or manage cited data from articles. Since the initial deployment, we have integrated PASS with the Open Access Button to make it easier for researchers to identify open access versions of their articles for submission.

PASS does so by making submission creation easier, connecting to relevant and available data from existing databases and systems (such as CrossRef, institutional grant databases, NIH’s data feed, Open Access Button API). PASS also provides and utilizes a common interface for services that would perform repository-specific deposit transactions.

 

Description

The Public Access Submission System (PASS) is a platform to assist researchers in complying with the access policies of their funders and institutions.

While many federal agencies have policies that require research results to be made publicly accessible, the requirements to comply with these vary greatly from one agency to another. These heterogeneous processes for compliance have become burdensome for researchers and their institutions. For research that is funded by multiple agencies, it can be especially complicated to ensure compliance with each funder’s access policy.

Meanwhile, many universities have implemented an open access policy, which also requires research outputs to be made publicly accessible. As a result, each publication can be subject to multiple policies each with its own workflow and requirements.

The PASS platform offers a unified approach for compliance, allowing the researcher to fulfil the access policies of their institutions and funders through a single website.

PASS aims to:

  • Provide seamless functionality for submitting/collecting accepted manuscripts and metadata relevant to one or more agencies’ sponsored research and/or relevant to the researcher’s institution.

  • Provide interfaces (e.g., APIs, SWORD, email-driven-workflows) for researchers, institutional administrators and agencies to access, collaborate on, submit, be notified about, and perhaps format accepted manuscripts and metadata.

  • Deposit manuscripts and metadata in the repositories specified by the policies to which the author is subject, and/or additional repositories requested by the author.

PASS was created by the Sheridan Libraries at Johns Hopkins University, in collaboration with the Harvard University Office for Scholarly Communication, the MIT Libraries, and with inspiration from Jeff Spies, formerly of the Center for Open Science.

https://pass.jhu.edu/

https://github.com/OA-PASS

http://bit.ly/OA-PASS

Why Here?

This project represents an singular opportunity to develop new markets for the Eclipse Foundation within the university and funding agency sectors. PASS has already engaged NIH, NSF and DOE and several universities (potential customers) in terms of its value. 

Researchers, administrators and funders have affirmed great value in aligning workflows related to compliance of institutional open access and federal funding agency public access policies thereby reducing burden on researchers. The American Association of Universities, Association of Public and Land-Grant Universities, and Council on Governmental Relations issued a memorandum to the White House Office of Science and Technology in response to an RFI about public access compliance. This memo is available at:

https://www.aau.edu/sites/default/files/AAU-Files/Key-Issues/Public%20Access/OSTP-RFI-Response-Public-Access.pdf

From the memo:

Ensuring public access to research requires significant coordination across a wide array of public and private entities and communities of practice, many of which have a variety of operating procedures (e.g. funders, researchers, institutions, scholarly disciplines, scientific societies, service providers, etc.). An added challenge to managing research sharing efforts is the decentralized nature of academic communities. This decentralization is also reflected in the variety of public access policies across the federal agencies. Policy harmonization across agencies is needed to incentivize researchers to engage in the open sharing of research outputs and assist institutions in compliance. A possible solution is the creation of more one-stop-shop access points for researchers that integrates grantee and funder operating procedures and requirements. One illustrative example is the PASS System developed by Johns Hopkins University, which is making great strides in simplifying the reporting, sharing, and compliance components of federally funded research.

Additionally, through its packaging and linked data capabilities, PASS also offers the potential to create information graphs that connect authors, grants, papers, data, metadata, and identifiers. These types of information graphs could lead to improved research services for discovery, use, search, etc. For-profit publishers have expressed interest in this capability given the considerable effort they currently exert trying to infer such connections.

The value of hosting this project in a community manner relates to hosted services for PASS. Our initial partners of Harvard and MIT both indicated that they would prefer having a hosted service rather than install local versions of PASS. Through a current National Science Foundation grant, we are exploring the possibility of a hosted service that would be used by universities widely to support their researchers for compliance across a range of funding agencies.

Future Work

The functionality that has been identified for the next 12 to 18 months includes generalized integration or interfacing with multiple institutional and federal funding agency repository platforms, and submission or management of cited data from papers. Before focusing on such functionality, we need to address these foundational issues: 

  • Scalability – PASS has been used on an individual user basis at JHU. With the Gates requirement, we imagine that simultaneous use of PASS will increase and use across institutions will also increase. An evaluation of PASS’ ability to handle simultaneous users would be important.

  • Security – While we have done our best to ensure security, we could definitely benefit from an external evaluation of possible security implications or issues. Somewhat related to this theme is an evaluation of our current mechanisms for authentication and authorization (i.e., are they sufficiently secure and scalable for other institutions).

  • Deployment – With our current PASS Docker environment, what would be the best deployment option for other institutions that accounts for their grants databases, IT environment, etc. It’s worth noting that the most likely option for many institutions would be hosted solution so this might be more of a “How would Eclipse deploy PASS?” question. Possible future work could include making the deployment process less vendor specific such as using Kubernetes Manifests or a Custom Resource which would make hosting easier.

  • Review, stabilization and documentation of PASS APIs - Assemble, review, and specify all public HTTP and programming APIs required to interface with PASS.  Some APIs used in PASS are not completely specified and/or are functionally obsolete (e.g. Fedora 4.x API, which pre-dates the Fedora API specification effort)   

  • Operationalization - Provide the necessary logging, monitoring, and metrics-gathering capabilities to PASS in order to assess its health, stability, and resource usage of a running instance of PASS.  Assess and fix any known major bugs or threats to application stability in a production deployment. 

  • Generalization guide - Audit the PASS implementation for JHU-specific assumptions, and define and document integration,extension points and branding guide from the perspective of an institution looking to incorporate PASS into their own infrastructure and brand it accordingly.

Project Scheduling
  • Initial contribution: 1 week after starting the effort (1 full time backend dev)

  • First successful build: 3 weeks from the start of effort (assuming we have 1 Fulltime UI dev and 1 Fulltime backend dev)

Project Leads
Interested Parties

National Institutes of Health (National Library of Medicine)

National Science Foundation

Department of Energy

Harvard University (made progress toward an initial hosted service)

MIT

Arizona State University

California Digital Library

Duke 

Iowa State

Notre Dame

Stanford

University of Michigan

Open Access Button

 

Initial Contribution

The initial contribution of PASS reflects the codebase as deployed at JHU.  It consists of:

  • Individual microservices, written in a mix of java and go, intended to be deployed in a containerized environment by an external orchestration mechanism.  Some of these microservices operate in an asynchronous fashion in response to internal events on a message bus, while others expose an HTTP API for clients on the web.

  • A Single Page Application (SPA) that serves the user interface to PASS for patrons wishing to deposit manuscripts.  This is written in Ember JS

  • A docker-compose based development environment, capable of running the entire PASS stack on a developer's local workstation

 

This initial PASS stack relies on the presence of external services or infrastructure in a production deployment:

  • SAML/Shibboleth identity provider (PASS functions as a Service Provider)

  • Container runtime and orchestration engine (e.g. AWS Fargate, or Kubernetes)

  • Relational database

  • S3-compatible object storage

  • Messaging bus (e.g. ActiveMQ)

 

The codebase is developed in a mixture of javascript, java, and go, and copyright is held by Johns Hopkins University, licensed to the community via an Apache 2.0 license.  Current and past contributors include JHU employees and contractors.

Source Repository Type