IASSIST 2017: Reflections from Kansas

IASSIST 2017 took place at the University of Kansas in Lawrence from 23 to 26 May 2017, under the title “Data in the Middle: The Common Language of Research.” Its archived program offers more than a record of dates and speakers. It captures a moment when researchers from information science, agriculture, geospatial analysis, health, computing and policy were testing a shared vocabulary for data-intensive work.

For an Australian audience, that Kansas setting is both familiar and instructive. Australian institutions face many of the same pressures: growing datasets, tighter expectations for reproducibility, fragmented systems and a need to turn research into public value. The difference lies in scale and context, from the compact campus city of Lawrence to national networks spread across Canberra, Melbourne, Sydney, Queensland and regional Australia.

The conference as a shared lens

The phrase “data in the middle” described the practical space between collection and use. That space included standards, metadata, cleaning, analysis, visualisation, governance and communication. The strongest reflections from the program suggest that progress began when participants stopped treating those as separate technical chores and started viewing them as one chain of decisions. A dataset had value only when another team could find it, interpret it and apply it responsibly.

Lawrence made that exchange unusually accessible. Around the University of Kansas, people could walk between session rooms, hospitality spaces and businesses near Massachusetts Street, with Clinton Lake and the wider Kansas landscape extending beyond the campus. This compact geography encouraged chance conversations among people who might otherwise have remained within disciplinary silos. It also gave practical meaning to a conference organised around common ground rather than a single discipline.

The plenary sessions and parallel presentations brought big data, deep learning, digital agriculture and global food security into the same frame. Their proximity revealed where the themes touched: satellite data feeding crop models, machine-learning outputs needing transparent documentation, and food-security decisions depending on evidence assembled across institutions. The archive records those links, while the programme design itself demonstrates how a mixed audience can discover them.

What attendees carried home

For attendees, the value of IASSIST 2017 lay in moving between examples of work rather than hearing one grand solution. Researchers returned to their organisations with sharper questions about interoperability and evidence. The recurring lessons can be summarised without reducing the event to a slogan: data work succeeds when technical choices align with the decisions that data will support.

Several themes stood out across the sessions:

• Data sharing is workflow design, not simply the transfer of files between teams.
• Metadata, provenance and clear documentation are essential parts of analytical quality.
• Deep learning depends on data preparation, suitable validation and an honest account of uncertainty.
• Digital agriculture shows how domain knowledge can be as important as model sophistication.

These points matter because they shift attention from access alone to fitness for purpose. A dataset can be open and still difficult to use; a sophisticated model can be accurate and still difficult to justify. Attendees also gained a richer appreciation of global food security as a shared challenge requiring cooperation across climate science, soil data, supply chains and local decision-making. The strongest takeaway was not a particular tool, but a habit of examining the complete evidence pathway.

What organizers learned

The organizer perspective turns attention to programme design, communication and operational detail. A conference about common language must practise that language. The archived schedule, plenary information and presentation listings show how carefully the event balanced shared moments with specialised discussion. That balance allowed newcomers to build context while specialists still had room for technical depth.

The operational pages for check-in, transportation and local recommendations reveal another form of organizer reflection. These details were not incidental to the academic programme. They shaped who could arrive on time, where informal exchange happened and whether visitors felt welcome in Lawrence. For an Australian audience, that is a useful reminder that the “user experience” of a conference includes travel, orientation and local information, not just the quality of the slides.

Several lessons from the event remain directly relevant:

• Build a common vocabulary before asking participants to tackle complex, cross-disciplinary problems.
• Use plenaries to connect major themes, then preserve small sessions for detailed, discipline-specific work.
• Treat schedules, maps, check-in guidance and local advice as part of the scholarly experience.
• Leave room for unplanned conversations, since collaboration often begins outside the formal session.

The final reflection is about follow-through. An archive keeps the programme accessible, but responsible stewardship requires that presenters release materials where possible, preserve licensing information and make citations usable. Organizers who understand documentation as shared infrastructure help participants extend the value of a meeting. This is particularly valuable for global networks, where a participant in one country may later work with a team that encountered the research only through the archived record.

An Australian reading of the Kansas exchange

Australia’s geographic scale changes the meaning of a compact conference environment. May is autumn in Sydney, Melbourne, Adelaide and Canberra, with cool mornings and changeable conditions, while northern Australia is moving through the dry season. Lawrence in late May sits in a different seasonal rhythm, with spring growth across Kansas and longer daylight. For Australian delegates accustomed to flights, time-zone changes and vast distances between collaborators, Lawrence’s walkable campus setting can make the density of the programme feel unusually immediate.

The local comparison also offers constructive contrasts. Canberra connects national agencies, grant administrators and policy institutes; Melbourne has strong health, artificial-intelligence and university networks; Sydney combines finance, health and commercial analytics; and universities across Queensland bring deep expertise in tropical agriculture, water and remote sensing. These are quite different from KU’s single-campus hub, yet each supports conversations like those held at IASSIST. The shared need is to connect research environments with the organisations and communities that will use the results.

Australian workplace conversation often favours directness and informality, with phrases such as “fair enough,” “have a go” and “no worries” softening disagreement without weakening it. Researchers can therefore borrow the conference habit of translating specialist language into plain explanations while retaining rigour. The useful question is not whether a method is fashionable, but whether a policy analyst, farmer, clinician or public servant can understand what the method supports and where its limits lie.

The Australian market for research data services combines competitive grants, public investment, university consultancies, start-up partnerships and government procurement. Contracts and tender rules reward clear deliverables, security and demonstrated value, while collaboration grants often require organisations to share both costs and credit. International networks remain important in that setting, and an international data exchange can offer a way to extend common standards and specialist relationships across borders. The Kansas lesson is to design projects around those relationships from the beginning, rather than trying to create a common data language after the datasets have been assembled.

From event memories to enduring practice

Looking back on IASSIST 2017, its central achievement was the creation of a temporary common language across research cultures. The event did not resolve interoperability, transparency or food-security problems. It did make those problems easier to discuss with precision, and it connected machine-learning and big-data methods to the human decisions that give data significance.

The programme also illustrates why reflection should cover both intellectual content and event design. Attendees gained from cross-disciplinary examples, while organizers coordinated plenaries, specialist sessions, practical guidance and an environment conducive to conversation. Each perspective depended on the other. Good sessions need logistical care, and effective logistics gain meaning when the programme gives people a reason to keep talking.

Dimension	Emphasis at IASSIST 2017	Enduring practice
Research problem	Big data and deep learning across disciplinary boundaries	Begin with the decision or public need that data must support
Communication	A common language for researchers and practitioners	Explain methods in plain English while stating assumptions and limits
Infrastructure	Presentations, schedules and plenaries brought communities together	Preserve searchable records, citations, licences and reusable materials
Agriculture and food	Digital agriculture connected with global food security	Combine local expertise, environmental evidence and transparent modelling
Conference design	Formal sessions and Lawrence’s compact geography encouraged exchange	Design for participation, orientation, informal contact and careful follow-up

That combination of intellectual exchange and thoughtful delivery is why the archive still speaks to organisers planning hybrid, regional or international meetings. Its strongest message is also a modest one: shared data work depends on shared habits of listening, documenting and explaining. Those habits remain useful long after the final session in Lawrence, and they travel well into Australian research communities and the organisations they serve.