Full text
Streamlining ACCESS-CI Resource Contributions: Human-Centered Integration of Science Gateways and Other Resources into the National Cyberinfrastructure 1st Rob Quick Research Technologies Indiana University Bloomington, IN, US [email protected] 2nd Esen Gokpinar-Shelton Pervasive Technology Institute Indiana University Bloomington, IN, US [email protected] 3rd Dinuka De Silva Pervasive Technology Institute Indiana University Bloomington, IN, US [email protected] 4th JP Navarro Argonne National Laboratory) Lemont, IL, US nav[email protected] 5th Winona Snapp-Childs Pervasive Technology Institute Indiana Unviersity Bloomington, IN, US [email protected] Abstract—In this project, we address a persistent and critical challenge in the field of cyberinfrastructure (CI): effectively integrating diverse resources, particularly those provided by science gateways (SGs), into the ACCESS-CI ecosystem. ACCESS-CI is a national collaboration that offers advanced CI services to the scientific community. Despite the promise of ACCESS-CI as a unified infrastructure, many SG administrators and CI resource providers encounter significant difficulties navigating registration and integration processes. The complexity of the system, coupled with inconsistent workflows and insufficient usercentered design, imposes a significant cognitive load on users, resulting in inefficiencies and frustration that can lead to the underutilization of available SG and other CI resources. To address these issues, our team employed human-centered design (HCD) methodologies to systematically understand user needs, pain points, and interaction patterns. Through iterative design cycles involving user feedback, prototyping, and usability testing, we developed a novel badge-based interface system that streamlines user navigation, clarifies the steps required to integrate and utilize resources, and significantly reduces cognitive burden. By structuring tasks into modular, visually intuitive badge workflows, users can more easily track progress, understand dependencies, and complete resource integration tasks with greater confidence and efficiency. The result is not only an improved experience within ACCESSCI, but also a flexible, extensible UI framework that can be applied to other complex CI ecosystems. This framework supports scalability across institutions and platforms. Ultimately, this work lowers technical barriers for a wide range of CI resource providers, including science gateway administrators. By making SG resources more accessible and easier to integrate into national CI ecosystems, this project enhances the scientific community’s capacity to conduct data-intensive research, collaborate across disciplines, and accelerate innovation. Index Terms—Cyberinfrastructure Usability, System Usability Improvement, Science Gateway Integration I. INTRODUCTION Cyberinfrastructure (CI) resources, which include SGs, are crucial for advancing large-scale simulations, data analytics, artificial intelligence, and a broad range of scientific research. Yet, their value lies not only in the additional cyberinfrastructure resources they provide, but also in how seamlessly they can be accessed, integrated, and utilized within the national cyberinfrastructure (CI) framework. These resources also require a method for potential users to identify the level of integration with the national CI quickly. The Advanced Cyberinfrastructure Coordination Ecosystem: Services and Support (ACCESS) plays a pivotal role in this landscape by coordinating a broad range of CI resources contributed by diverse Resource Providers (RPs) and making them available to the research community through a centralized platform. Through standardized APIs, shared services, and unified support, ACCESS helps democratize access to CI across institutions and disciplines, avoiding community resources from becoming isolated or siloed [1]. However, without a visual, user-focused way to identify resources with an appropriate level of ACCESS integration, discovery and utilization may still be difficult. Despite these strengths, the process for onboarding and integrating new resources into ACCESS remains a significant bottleneck. Resource Providers routinely encounter a fragmented and unintuitive integration workflow, relying on static documentation, opaque procedures, and limited support for tracking progress. These barriers result in a high cognitive load, inefficient task execution, and often, delays or underutilization of valuable computational resources. In response, this project addresses the integration challenge not simply as a technical problem, but as a human-computer
interaction issue that demands user-centered solutions. Applying human-centered design principles [2], our team developed a badge-based interface system for the ACCESS Operations site. This redesigned UI reimagines the integration workflow as a guided, stage-based process that aligns with the mental models of RPs. The interface offers straightforward navigation, visible progress indicators, and milestone-driven task organization, reducing complexity and enhancing task efficiency. This redesign not only improves the usability and effectiveness of the ACCESS integration process, but also introduces a scalable and adaptable framework for improving similar complex platforms. By prioritizing accessibility, cognitive ease, and streamlined interaction, our work supports more inclusive and efficient participation in national CI, contributing to broader goals of collaboration and scientific innovation. Specific to the SG community, this work provides a welldocumented and defined integration path with the national CI community. Thus, ACCESS badges have the potential to increase the discoverability, impact, and collaboration opportunities for science gateway projects. II. PROBLEM SPACE Currently, SG administrators and other CI resource providers integrate their systems into ACCESS using static README-style documentation spread across multiple locations. This resource-intensive approach provides minimal contextual support and requires users to piece together the necessary steps without guided interaction. Through interviews (N=5), site-wide survey data, workflow analyses, and observations of onboarding sessions, we identified two key usability gaps: 1) The system provides uniform workflows regardless of an RP’s experience level. 2) The UI offers limited affordances for tracking actions, receiving confirmation, or understanding task dependencies. New RPs often misinterpret prerequisites and overlook necessary steps due to the disjointed presentation of information and the absence of clear guidance. Experienced RPs, by contrast, expressed difficulty in locating specific workflows, verifying the status of submitted tasks, or accessing completed configurations for updates. The following table lists the initial set of badges, along with their applicability to Compute Resoruces (C), Storage Resources (S), Cloud Resoruces (CL), or Science Gateways (SG). Science gateways differ from other ACCESS-CI resource providers, which provide computing, storage, and cloud resources. Science gateways serve as a service, leveraging web interfaces, high-speed networks, HPC computing, highcapacity storage, and other CI resources. And while they differ slightly in the roadmap process, many integration processes are similar. The modular design of the badging allowed these differences to be easily implemented. TABLE I BADGE APPLICABILITY TO RESOURCE TYPE Badge ACCESS Resource Type SG C S CL Resource Description ✓ ✓ ✓ ✓ Allocated ✓ ✓ ✓ Ticket Support ✓ ✓ ✓ Enhanced Networking o o o Cybersecurity Coordination ✓ ✓ ✓ System Status o o o Globus Transfer o o Software Publishing o o Documentation ✓ ✓ ✓ Open OnDemand o o Pegasus o o Training o o o o Affinity Group o o o o Usage Metrics ✓o o o Performance Metrics o o o a’o’ denotes optional III. METHODOLOGY To guide our research and design activities, we adopted the Double Diamond Design Process [3]. This human-centered design framework structures innovation across four key phases: Discover, Define, Develop, and Deliver. This approach allowed us to ground our interface redesign in the lived experiences of SG and RPs, ensuring that our solutions addressed real-world pain points and were shaped by iterative feedback. During the Discovery phase, we conducted five semistructured interviews and multiple observational sessions with members of the ACCESS RP Forum, an actively engaged and representative subset of the broader RP community. Participants were recruited through convenience sampling, based on availability and interest, with a focus on capturing perspectives from both new and experienced RPs. In these sessions, participants were asked to complete onboarding tasks using the existing ACCESS integration system while thinking aloud. We recorded their interactions, verbal commentary, and nonverbal cues to identify moments of confusion, inefficiency, and misaligned expectations. Fig. 1. Double-Diamond Design [6] Our analysis revealed key friction points in the onboarding experience, including unclear task progression, a lack of contextual feedback, and the absence of role-specific guidance. These insights directly informed the design of our badge-based
interface system, which was developed to provide structured, progressive guidance and reduce the cognitive load associated with navigating a complex integration process. In the Develop and Deliver phases, we created iterative prototypes and tested them through usability walkthroughs and informal feedback sessions with the same participant pool. Users responded positively to the redesigned interface, particularly appreciating the integration roadmap and the centralized landing page, which offered a clear and holistic view of the process. Compared to the original system, participants made fewer errors, relied less on external documentation, and demonstrated higher confidence in completing onboarding tasks. One RP summarized the impact succinctly: “I finally understand where I am in the process—it’s not a guessing game anymore.” These findings validate the effectiveness of our humancentered approach and reinforce the importance of aligning system workflows with users’ mental models [4]. The success of our prototype not only improves integration within ACCESS, but also offers a replicable methodology for improving usability across other cyberinfrastructure platforms. IV. DESIGN PROTOTYPE The current state of the redesigned UI reflects four core design improvements, each shaped by user feedback and grounded in principles of human-centered design. These enhancements were consistently reported by participants as more intuitive, efficient, and aligned with their real-world workflows. 1) Centralized Landing Page with Tailored Entry Points: To streamline orientation and reduce initial cognitive load, we developed a centralized landing page featuring three clear calls-to-action: Learn More (for new Resource Providers seeking introductory guidance), Start New Integration (for those ready to onboard a resource), and Integration Dashboard (for experienced users managing ongoing integrations). This role-sensitive structure supports different levels of familiarity and aligns with users’ mental models, enabling faster and more confident task initiation. 2) Five-Stage Integration Roadmap with Progress Tracking: To improve RP’s sense of direction and task continuity, we introduced a five-stage roadmap that outlines the complete integration journey: Register Resource, Pick a Roadmap, Select Resource-Specific Badges, Complete Badge Tasks, and Submit for Verification. This visual framework serves as a cognitive scaffold [5], helping users understand their current position, anticipate next steps, and reduce uncertainty throughout the process. This process is depicted in Figure 2. 3) Resource Management Dashboard: To support efficient management of multiple integrations, we implemented a dashboard that consolidates key functions: monitoring active and completed resources, editing resource details, and viewing real-time integration status. This centralized hub reduces navigation overhead, allowing RPs to manage tasks with greater clarity and fewer interruptions, thereby improving overall workflow efficiency. 4) Badge Completion Process: Each onboarding requirement is organized into a task-specific badge containing clear instructions, progress tracking, and submission status. Once completed, badges are reviewed by ACCESS staff through a lightweight manual verification process that ensures technical accuracy and policy compliance. This human-in-the-loop review maintains quality standards while minimizing delay. Approved badges mark the resource as integration-ready; those requiring revisions are returned with specific, actionable feedback. This structured yet flexible approach supports both clarity and speed in the onboarding process. Fig. 2. Double-Diamond Design [6] V. NEXT STEPS AND BROADER IMPACTS Currently, the alpha-phase implementation prototype is in place for Resource Providers, and testing is in progress. Insights from these ongoing evaluations are informing the next phase of design refinements, with a near-term focus on completing the researcher and concierge interfaces. These additions aim to provide a cohesive, role-sensitive experience that supports the full range of users interacting with the ACCESS platform. Significantly, the badge-based system was developed with scalability in mind. As technologies evolve and Resource Provider use cases diversify, new workflows and badge sets can be introduced with minimal disruption to the existing structure. This modularity allows the platform to adapt to increasing complexity and domain-specific requirements without sacrificing usability or clarity. Additionally, the current scope of the project focuses on SG administrators and resource providers, not on the researcher’s discovery of resources. While the work thus far provides a streamlined way for resource integration, testing from the user side still needs attention. How does the researcher benefit from the badging system? And does this streamline the discovery of resources? The ACCESS-CI Web and Branding Team will conduct this researcher-facing design work and will leverage the effort of ACCESS-CI Operations to streamline resource provider integration.
Overall, this work demonstrates the transformative potential of applying human-centered design to national cyberinfrastructure systems. By reducing onboarding friction, clarifying workflows, and aligning system interactions with user mental models, our approach significantly improves the accessibility and efficiency of resource integration. We believe this design framework provides a replicable model for other complex ecosystems and can contribute to broader efforts to enhance usability and impact across the research infrastructure landscape. ACKNOWLEDGMENT This material is based upon work supported by the National Science Foundation under Grant No. 2138307. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. REFERENCES REFERENCES [1] ACCESS. (n.d.). Retrieved from https://access-ci.org/ [2] Preece, J., Rogers, Y., and Sharp, H. (2002). Interaction design: Beyond human-computer interaction. New York: John Wiley and Sons. [3] UK Design Council’s Double Diamond Model: What is the framework for innovation? Design Council’s evolved Double Diamond. [4] Nielson, J. (2010). Mental models. NN/g Nielson Norman Group. https://www.nngroup.com/articles/mental-models/ [5] Norman, D. A. (1988). The psychology of everyday things. Basic Books. [6] Rosala, M. (2020, March 15). Discovery: Definition. (Updated August 2, 2024). Retrieved from https://www.nngroup.com/articles/discoveryphase/