From Us vs. Them to We: Collaborating for Change

Key Takeaways

  • It takes sustained focus, effort, and time to create the foundation of trust that underlies all viable collaborations.
  • While IT collaborations usually encounter challenges and entail some element of risk, the combined brainpower at the table can prove highly efficient and effective at finding solutions.
  • An enterprise-scale virtualization initiative at Indiana University pulled 15 members from central IT and more than 100 from edge IT staff in a large but effective committee that collaborated to realize multiple benefits.

After two decades of provisioning campus desktops via a high-touch, distributed model, Indiana University (IU) sought a more efficient mechanism that would scale to meet demand, provide agility and flexibility, and free IT staff from the repetitive maintenance and update cycle to focus on innovation. IUanyWare serves software to desktops, laptops, and mobile devices, at any time, to IU users wherever they are. IUanyWare is the first part of a virtualization solution on an enterprise scale. The second part will address special-use cases.

The virtualization initiative ignored most conventions in IT development at IU. A collaboration of some 118 IT staff comprised the project team. Of those 118, only 15 represented the central IT organization, and the project leads were "edge" staff from departments across the university. The gains go far beyond the solution they found, including:

  • Professional development opportunities for committee participants
  • Creation of strong support networks across departments, schools, and campuses
  • A more unified IT team, feeding into the One IU IT initiative
  • A conviction that virtualization technology can effectively serve the university's mission of teaching and research


The IT environment at IU was conceived, structured, developed, and delivered by University Information Technology Services (UITS), IU's central IT organization. UITS served the university top-down, as the central and sole provider of enterprise IT resources and solutions. In the 1990s UITS also seeded a program that encouraged schools and departments to hire their own IT staff to augment UITS resources and provide on-site, hands-on help to their own end users by loading software, handling security, and providing training. In this distributed computing model these "edge" IT staff were a culture separate from UITS, with minimal interaction. A couple of decades later, these duties, duplicated across departments and campuses, signaled a new role for technology and an area ripe for innovation. The time was right to investigate an enterprise system for provisioning desktops and applications across the university.

In 2009 IU's IT strategic plan, Empowering People, proposed collaboration as an efficient and economical way to provide IT resources. A culture of collaboration took shape in IT with the formation of the IT Managers Council, an advisory committee of IT leaders from all campuses, various schools and departments, and a few ex-officio members from UITS. The council was created to help operationalize the values of partnership and edge, leverage, and trust (ELT).1 (See the sidebar "Selected Readings on Edge, Leverage, and Trust.") Trust, the least tangible and most fragile element in ELT, is a prerequisite to introducing change in an organization or a culture. The concept of One IU IT, launched by the IT Managers Council, helped IT staff across IU view themselves as one community with shared goals, and over time, build relationships of trust.

Sue Workman, IU associate vice president, Support, discusses the importance of trust in collaboration:

Key Points from Sue Workman Video

In the concept of edge and leverage there was a third component called trust. To get leverage from the IT resources at IU, we had to build trust. That meant asking people to change, to move from silos to a strategic environment.

To build trust we created a relationship map to view all levels from executives to administrative staff. We invested time in lunches, meetings, discussions…

It took time for people to understand edge, leverage, and trust.

We worked on relationships for a couple of years to create the right cultural environment. We formed the Community Partnerships team as a liaison group to learn the needs of IU's campuses and to take services to them. We established an IT Managers Council of people from each campus and the major schools and leaders in the IT professional community at IU as an advisory body.

Initially there wasn't a lot of trust. It took time and working together. The edge has to trust central IT and central IT must trust the edge. Now we're all one university.

The need for a new desktop management solution at IU came up in meetings of the IT Managers Council. Some edge IT staff had created successful single-use virtualization solutions for their departments. A couple of edge staff on the council suggested evaluating virtualization technologies as a desktop management solution.

A group then formed the Client Endpoint Virtualization (CEV) committee to evaluate virtualization as a possible enterprise solution. A call for membership was issued, and 118 IT staff replied, all but 15 from the edge. With edge staff in the majority, and with their experience in virtualization, edge staff were the natural choice to lead the project.

Collaboration Is Risky

There is no playbook for collaboration. Each initiative comes with its own challenges and risks. The larger the collaboration and the more valuable the outcome, the greater the risk.2 Evaluating virtualization technologies for use across the enterprise was high stakes and high risk:

  • A project of this significance, led by edge staff, was unprecedented and overturned the typical UITS top-down development model.
  • The edge and UITS had never collaborated at this scale.
  • The committee combined two different cultures: the structured center and the independent edge.
  • Virtualization was uncharted, so the long-familiar development model with its clear-cut roles and responsibilities did not apply.
  • The committee had to create both itself and the solution it hoped to pioneer from the ground up.

Then there was the problem of basic terminology. What exactly does "virtualization" mean? The concept has different meanings to different people, even among techies. What can be virtualized? Why doesn't the current reality match our expectations? What can we promise various audiences given different security requirements like HIPAA and FERPA?

Even the activities typical in most collaborations involved obstacles. The first and biggest was that committee members were spread across campuses. Given logistics and workload, rarely could everyone make meetings and discussions. It took uncommon focus and dedication for edge staff to meet their commitments to the schools and departments who paid them and still have time to evaluate new enterprise tools.

What Made It Work?

How did the committee evolve from a relatively unstructured, primarily volunteer group into a productive collaboration?

The committee's loose internal structure looked nothing like the traditional bureaucratic development process, where each role is defined and assigned. The project lead was a persuasive virtualization advocate who recruited many of the 118 committee members and built support beyond the committee. Three subcommittees — software, product evaluation, and licensing — were each led by an edge staff member with expertise in the area. The project lead and two other edge staff, also strong personalities and known technical experts who believed virtualization could effectively serve the university's mission, led the subcommittees.

The fact that only a few staff were assigned full-time to the project could have hindered progress, but the committee's size became an advantage. As some staff dropped back, others came forward. When new skills or roles were needed, motivated staff eager to learn or assume more responsibility found opportunities for professional development.

Mark Niswander, enterprise operations planner, UITS, talks about his opportunities for professional advancement:

Key Points from Mark Niswander Video

I was initially administrative assistant to the director implementing virtualization technologies for Indiana University. In this project I provided 100 different IT leaders with tools for quick decision-making — project reports, executive summaries, timelines, Gantt charts. I flourished because of the collaborative nature of the project. The IT leaders mentored me on what it takes to have a successful IT enterprise project.

I was fortunate to be in the right place at the right time. Central IT saw the need for operations planning between the edge and UITS and in central enterprise projects. I was able to change my position into a more structured role around operations planning.

As often happens, there emerged that special class of IT staff always willing to make time for the next challenge. A dedicated 14-person core of leaders from the edge and UITS made decisions, wrote summaries, and gave progress reports, while the three subcommittee leaders provided momentum and inspiration.

To compensate for distance, the group leveraged multiple technologies. Teleconferencing supported meetings, and a video bridge remained open on engineers' meeting days. Documentation and discussions were supported by Oncourse and SharePoint, and a collaborative website served as a communications hub for news, directions, and decisions. The website also collected input, which was welcomed at every stage of the project.

To manage decision making, the committee adapted a successful model from the collaborative process used in developing Empowering People:

  • Parameters were set for discussion.
  • Open channels were established for input and feedback from the 118 committee members and other interested IT staff.
  • Decisions were in the hands of the few — the project lead and the chairs of the other two subcommittees.

Making Progress Toward a Solution

One of the first decisions the committee faced was determining whether a solution based on virtualization was feasible. Here again the committee's size proved beneficial. The range and depth of technical knowledge and variety of perspectives at the table helped the committee declare that the time was right to investigate a solution involving virtualization:

  • Virtualization technology was now viable.
  • IU's IT infrastructure of networks, supercomputers, storage, and support was aligned.
  • Edge staff had been successful in single-use instances of virtualization.
  • Early adopters of virtualization were enthusiastic, and other departments were waiting in line.
  • The idea addressed three strategic action items in Empowering People, so would likely gain the backing of university and IT leadership.
  • IT center and edge staff had built sufficient trust that IU could now consider a unified technology for managing personal computing.
  • Economic realities supported a leveraged environment across IU.

To move from theory to practice, the committee needed to test virtualization at an enterprise scale. The single-use cases the edge had created didn't prepare IT to architect, install, and configure an enterprise prototype. The CEV turned to consulting source Gartner, Inc. for help defining the basic criteria for an RFP seeking vendors who would build prototypes on-site. Committee members tested and compared three vendor prototypes in actual use cases to find the best fit for IU's needs, fiscal reality, and IT infrastructure.

The chosen vendor, Citrix Systems, Inc., then spent an initial 12 weeks on-site developing the prototype we planned to install. Engineers from IU sat alongside vendor engineers as they designed and rolled out the prototype, observing and recording the steps and the decisions behind them. Today IU engineers can handle much of the engineering and architecting themselves. Learning from the experts, hands-on, in IU's own environment presented a rare professional development opportunity. For example, Stephen Jeffers, originally UITS senior support advisor, developed sufficient skills to serve as the project's principal enterprise systems engineer and is now considered IU's expert.

Scott Hemmerlein, senior systems engineer, IU School of Medicine, and Stephen Jeffers, principal enterprise systems engineer, UITS, discuss their experience and gains from collaboration:

Key Points from Scott Hemmerlein and Stephen Jeffers Video

Jeffers: As a former help desk advisor, the virtualization project has provided me with the opportunity for professional development.

Hemmerlein: My role as sole virtualization engineer at the School of Medicine and juggling other work was challenging.

Jeffers: Now, with a virtualization team of engineers and collaborators across the university, things are going much better.

We were skeptical about a collaboration of this size, but the additional strengths, points of view, and perspectives sped the project up.

Hemmerlein: Bringing people from across campus is a great way to run an enterprise project. Collaborators feel like they have a say in what's going on and are part of the overall, large project team for the whole university.

Jeffers: Agreed.

Meeting Varied Users' Needs

IU's virtualization solution needed to accommodate two distinct types of uses:

  • The largest population, with the most pressing need, encompassed faculty, staff, and students, whose desktops, laptops, and mobile devices required software fundamental to their university life and work.
  • The second use group comprised the known special-use cases, plus unknown others to come.

The edge staff knew that some larger units had already begun investing in their own virtualization solutions. If this trend continued, IU would have separate vertical stacks of technology unable to integrate and too costly to sustain over time. Such information cued the committee to create a unified solution based on an architecture generic enough to meet the needs of 80 percent of the population and sufficiently agile to be adaptable for each special-use case (the other 20 percent).

Rollout of the enterprise solution is planned over two phases, beginning in fall 2011. The first phase, IUanyWare, will allow the IU community to access from their desktops, laptops, tablets, or smartphones, at any time, from any location, more than 200 software titles, including some formerly available only in certain computing labs.

The second phase focuses on creating special-use solutions for such entities as the Schools of Medicine and Dentistry and the School of Public and Environmental Affairs, each differently configured to accommodate individual privacy and security requirements including HIPAA and FERPA regulations. Other use cases tagged for deployment include the IU Libraries, IUPUI administration, and regional campuses.

Duane Schau, director, Leveraged Support, UITS; Marc Wilson, IU School of Medicine; and Pete Bucklin, director of Teaching and Learning Technologies, IU School of Informatics and Computing, explain aspects of the technical solution chosen:

Key Points from Duane Schau, Marc Wilson, and Pete Bucklin Video

Schau: We consulted the Burton Group [Gartner] to define core criteria for the project.

The RFP process for the prototype included:

  • Enterprise virtual desktop.
  • Enterprise virtual application.
  • Enterprise store allowing students, staff, and faculty on-demand access to 200 to 240 applications for use anytime, anywhere, on just about any device.
  • The core components had to meet the needs of schools and departments across seven IU campuses.
  • Inside this, each department and school will have special use cases tailored around virtualization that need to tie into the enterprise product for client virtualization.

Wilson: For the School of Medicine, we needed a standardized product flexible enough to meet the needs of the edge. We followed the 80/20 rule, standardizing most, with enough flexibility to modify it to our needs:

  • In medicine, clinical information with lots of sensitive data must remain safe.
  • Downtime must be kept to a minimum.
  • Mobility: Doctors moving between clinics and hospitals need ubiquitous access.
  • The vertical software for medical clinics often lags behind technology.
  • We have to support software that may be two or three versions ahead of our vertical software.

Bucklin: For Informatics, we do a lot of research and teaching in security. For our security graduate program we need the flexibility to drop virtual operating systems into virtual networks, to conduct research and execute on labs.

We need access across multiple campuses, and many locations

Schau: The enterprise providing core components for client virtualization will reduce overall cost and allow schools and departments to tailor use to their missions.

Anticipated Gains

Even at this stage of the initiative, the university expects many benefits. For the user, IUanyWare:

  • Provides on-demand, ubiquitous access to the most-used applications, including Microsoft Office and Adobe, and hundreds of other software packages including statistical and mathematical software
  • Equalizes opportunity for all IU populations: overseas study programs, nontraditional students, mobile users, and those on sabbatical
  • Solves the problem of crowded technology labs and of students having to go to certain labs to use specialized software
  • Delivers freedom to adjust study schedules around individual preferences instead of open lab times
  • Makes the cycles of upgrades, patches, and security updates transparent, freeing users from having to keep current and from the dislocation caused by upgrades
  • Democratizes software regardless of budget, since all schools, departments, campuses, and their populations have the same access to the same resources
  • Lightens the cost burden on students by giving them access to optimum IT resources, even using older, less-powerful systems with smaller hard drives

Financially, IUanyWare creates efficiencies by:

  • Providing a timely, sustainable, and efficient solution for IU's strategic goal of supplying software in abundance
  • Allowing for concatenating, tailoring, and managing spending on software licenses
  • Requiring less-powerful desktop devices, thus extending IU's lifecycle dollars
  • Allowing staff to focus on innovation — key in sustaining a competitive, forward-looking IT environment that attracts partnerships and supports grant seekers

Collaboration Succeeds

It took the large-systems expertise of central IT combined with the local knowledge and perspective of the edge to create a solution agile enough to serve both the broad needs and the multiple, specific needs across IU. It took the many trust relationships of both groups to successfully work through the countless challenges on the way to that solution.

One of the most difficult parts of IT change is selling it to users. Rolling out a new technology can entail hosting focus groups and town halls and developing communications plans, announcements, and other publicity, all in service of preparing the audience. Thanks to the edge, the solution is selling itself via the peer network of early adopters who vouch for virtualization. In the words of one edge staff member, "Instead of asking ourselves how we can build interest, we're now wondering how we can slow demand."

From this experience CEV members have learned the advantages of collaboration. Between edge and center lies a deeper understanding of each other's priorities, audiences, and challenges. In tackling technical challenges, people made connections and formed networks of expertise to leverage in the future. Many gained a new appreciation for the caliber of professionals across IT and pride in the IT community. This enlarged collective perspective blurs the boundaries between center and edge.

Stephen Jeffers, virtualization engineer, UITS; Charles McClary, director of IT, IU Division of Campus Recreational Sports; Mark Lynch, manager, Support Systems and Licensing, UITS; and Tom Zeller, senior technical analyst, UITS, talk about the benefits of collaborating on the virtualization project:

Key Points from Stephen Jeffers, Charles McClary, Mark Lynch, and Tom Zeller Video

Zeller: In the past central IT designed the service and announced it. Here we widened the group involved in thinking through the design.

McClary: This gave departmental support a great opportunity to be involved. We worked with virtualization, collaborated on solutions, got feedback, and created relationships I leverage to this day.

Jeffers: It was a great opportunity to get exposure to big projects like this, get out in front of the technology, and get feedback from collaborators. We considered a lot of opinions and technical perspectives. The solution benefits everyone as whole — not just central IT.

Lynch: This expanded my view of software licenses. So many were involved from different areas, I got a new perspective on priorities. It's no longer about the number of licenses and dollar amounts. Now we see the benefits of a centralized service in a virtualized environment.

Zeller: The project evolved over multiple years. It started as thin clients with a limited niche application and questionable return on investment. As we widened the scope of thinkers, we saw new ways to deploy the technologies, virtualized applications, and number of use cases. The ROI improved.

McClary: We view resources as a commodity tool, meeting needs more quickly than before. UITS staff and edge units now see the caliber of IT staff across IU. We are blurring lines among IT groups.

Lynch: A project with this many collaborators may be a first for IU. Usually you're simply busy with your own job. The old model of central IT simply building something no longer works.

Zeller: Up to 130 people were involved while a core group handled the design. That balanced well.

McClary: We've set a new expectation for the future. People outside central IT see opportunities to be motivated and lead.

Zeller: Many younger technologists had an opportunity to grow their skills. In the past more senior IT leaders would have architected this solution. This collaboration helped nurture our younger talent.

Lynch: This is a paradigm shift. It's a new beginning, technically and as a partnership. The benefits of teamwork show.

Zeller: This is an opportunity for departments to value their support providers more highly, and for those IT staff to learn new skills and advance to higher-value-added functions.

With this collaboration, One IU IT has reached a milestone. Staff have a new expectation for the work of the future: not as silos, but as one community. The gains from this experience extend beyond the virtualization solution to include abundant opportunities for professional development, the creation of strong support networks, and a more unified IT community.

Sue Workman, associate vice president, Support, IU Office of the Vice President for IT; and Duane Schau, director, Leveraged Support, UITS, talk about collaboration in IT and a new era for One IU IT:

Key Points from Sue Workman and Duane Schau Video

Workman: Initially some people thought we were crazy [when we launched this project].

Technology and innovation are where they were 20 years ago, moving quickly. 20 years ago, we distributed computing at the university. Now 20 years later, we're making ELT [edge, leverage, and trust] work.

Schau: Working with IT staff across the university we can provide skills and abilities we couldn't provide individually. This is a great exercise in co-laboring and takes it to the people with great results.

Workman: One IU IT sells itself.

Buckminster Fuller said, "To change something, build a new model that makes the existing model obsolete."3 The new model at IU is collaboration.


Jan Holloway directed and produced the videos included in this article, working with videographer Yvonne L. McCann.

  1. Brad Wheeler, "Accelerating the Journey into the Third Era," Keynote, Indiana University 13th Annual Statewide IT Conference 2008, Bloomington, September 17, 2008.
  2. Brad Wheeler video, "Thriving in the Era of Rabid Collaboration."
  3. R. Buckminster Fuller, quoted on Creating, Creative Quotes and Quotations, On Change.