Copyright © 1998 The Trustees of Indiana University

Institutional/Enterprise Licensing
Achieving Standards by Providing SoftwareTo The Institution

Brian D. Voss
& Jill Marx Schunk
Indiana University

Acknowledgements:

We would like to especially thank the following individuals for their contributions to this paper -- either directly or vicariously -- and for their efforts in support of the students, faculty, and staff of Indiana University through their work and leadership in the area of institutional software licensing.

• Richard White, IUPUI Purchasing, Indiana University
• Mark Lynch, Manager Computing Advising Team, University Information Technology Services, Indiana University
• Vice President for Information Technology Michael A. McRobbie, Indiana University
• Guy DeStefano, University Director of Purchasing, Indiana University
• Laurie Antolovic’, Office of the Vice President for Information Technology, Indiana University
• Prof. Michael Dunn, Chair, University Information Technology Committee, Indiana University
• Assoc. Vice President Garland C. Elmore, University Information Technology Services, Indiana University
• Assoc. Vice President Christopher S. Peebles, University Information Technology Services, Indiana University
• Indiana University President Myles Brand

There are many others who every day play a part in the delivery of excellent information technology services in support of teaching, learning, research and service at Indiana University. We wish to express our appreciation to them for their efforts, which allow us to have such excellent things to write about!

For years, university purchasing and information technology organizations have wrestled with the concept of defining and enforcing standards for pervasive desktop and LAN software. The result of which is often a contentious process of selection, followed by what some would call a dictatorial period of enforcement. This rationing of support, and the process to get there, exhausts IT organizations, purchasing organizations, and users alike. But this need not happen, if the institution can leverage the demand of the preponderance of its users, to achieve broad-based access to pervasive software by all.Indiana University has been a pioneer in working with vendors (Novell, Corel, Sybase, Microsoft, and many others) to develop an approach to licensing that allows the university to leverage its investment to put the maximum amount of software in the hands of all affiliate students, faculty, and staff. By responding to user demand trends, and then acquiring highly demanded software for all affiliates, IU has actually managed to accomplish two goals: 1) An abundance of software available for use by all; and 2) Demand-based standards for the most widely-used desktop and groupware software needed by all.These two objectives, while appearing to be at cross-purposes, are not. The key is for IT organizations to set standards by providing standard products -- while continuing to encourage and support diversity by monitoring user demands and being responsive to changes. Another key is a proactive approach on the part of Purchasing organizations to seek out opportunities for more global licensing, and in partnership with the IT organization to negotiate and implement these licensing arrangements.

As Indiana University (or any other institution, for that matter) began to examine the potential for institutional licensing of software, a number of key conditions of the environment had to be in place. For IU, the process of setting the stage was a combination of planned efforts, natural evolution of our environment, and we suppose a bit of good fortune. However, six facets of the IU environment made it possible for more broadly based licensing approaches to become an option.

1. Centralized Purchasing

Although each campus of Indiana University has an organizationally separate Purchasing department, the institution has a University Director of Purchasing, responsible for coordination of purchasing activities across the 8-campus IU system. Perhaps more important than organization structure, the institution approaches acquisition with a singular philosophy. While the campus-based Purchasing Departments are focused on the specific and diverse needs of their campus, they all function together within the strategy of institutional buying. Contracts with vendors are nearly always negotiated with a system-wide theme, so that the power of the entire university’s purchasing can be leveraged for the good of all. Suppliers are selected for their pricing and service, but often a key element is their ability to deliver to all eight campuses.

Critical Mass of Software Purchases

Simply put, the University buys a lot of software. As one would expect, perhaps as much as 75% of the institutional software outlay (not counting central systems products) is done to provide basic, perhaps somewhat pedestrian software. In this area, very few products (and companies) compete for this fairly large market.

Ability to Track Institutional Expenditures

Key to consolidating the power of the millions spent on software (and other) products is IU’s ability to understand how institutional funds are being spent. Two key systems in place combine to effectively give University administration a view of how money is being spent, and where it is being spent. The ‘How’ function comes from IU’s Financial Information System (FIS), which allows for decision support reporting on both local and global levels. Finely delineating a series of expense classes helps IU understand the general purpose of every expenditure made within its responsibility centered budgets. The "Where" function is provided by The Online Purchasing System (TOPS). TOPS tracks money spent in acquisitions by vendor and item description and this allows for decision support information on many levels.

Centralized Information Technology Leadership

Over the past ten years, IU has been steadily gathering together its diverse information technology organizations and building a consolidated structure. This process culminated in the establishment of the position of Vice President for Information Technology (VPIT) in 1997, and the charge to that office to direct all IT strategies and activities across the IU system. The central information technology organizations (CITO) on the two core campuses (IUB and IUPUI) are part of the VPIT’s direct operational division structure. The IT organizations on IU’s smaller six regional campuses share joint reporting relationships between VPIT and their campus chancellor. The VPIT, as a member of the President’s core leadership team, can bring focus on IT needs -- in the case of this paper, software licensing -- to the highest levels of decision making in the institution.

Strong Partnership Between IT and Purchasing

There is great potential for competition and disharmony when both IT and Purchasing functions are so strongly centralized.. However, at IU, this pitfall has been avoided and the two work in close cooperation in terms of strategy and day-to-day functioning. This is due to a number of factors. First, leadership is cooperative and encourages an environment of true partnership. Second, the implementation of the TOPS system actually brought the two organizations together, as both worked side-by-side to implement IU’s first truly distributed information system. And third, the people in the two organizations have built solid interpersonal relationships over the period of years. The people are not just colleagues -- oftentimes, they are friends. Perhaps this last element is the most important, and of course, the one most likely to fall into the ‘good fortune’ category of events which have led IU to where it is today.

Institutional Commitment to Information Technology

Evolutionary Development of Software Acquisition

It has been a long road from where we were at IU, to where we are today with regard to software acquisition. Along the way, as the six elements described above were put into place they acted as triggers for the evolution of software acquisition. As well, changes outside of IU, with the vendors and within the software industry played an important role in getting us to where we are today. Let’s take a moment to review where we were, and better understand the forces and events that gave rise to IU’s approach to institutional software licensing.

The Bad Ol’ Days

As desktop computers emerged from the primordial ooze of the mainframe computing industry, the crawled upon the shores of a university purchasing landscape more used to buying books, furniture, supplies, and the like. Computing itself started out very slowly, with only a few PCs representing islands of isolated computing. We all are aware of how rapidly the distributed computing environment developed, and the various forces that played a part in the rise of highly distributed desktop computing environments. That discussion is the fodder for a thousand papers, so we will not review it herein. Let’s just say that those in the IT side may not have noticed what was happening to our brethren in Purchasing! Let’s focus on that.

In the days before the six elements developed, ad hoc purchasing of software products was the rule of the day through the early days of personal computing. The central IT organization was fixated on its continuing (and growing) role in providing central resources (systems, networks, support, etc.). Users themselves made the decisions -- sometimes with support from the IT organization, but oftentimes on their own -- about which desktop software to buy and where to buy it from. Computing was viewed as a new ‘toy’ in many ways, and price was not the determining factor in decisions regarding software. Purchasing had little expertise in the IT field, and simply continued to pay their role as order takers and processors.

However, the first three elements began to take shape and thus acted as ‘evolutionary triggers’ on the way the institution approached software acquisition. The centralization of Purchasing strategy came into being first, as a precursor to the development of the third element -- the ability to track institutional expenditures. With the arrival of TOPS on the scene, and the growth in the size of institutional purchases of software (approaching critical mass), software purchases came under closer scrutiny. A result of the evolutionary development in these first three areas was that Purchasing adopted a desktop-computing focus, establishing positions to coordinate desktop computing acquisition. This included both hardware and software. Perhaps important to note, as the IT environment itself included both central and distributed components, Purchasing mirrored that environment by focusing resources in both areas. They were closely linked, but Purchasing understood the different philosophies, attributes, and directions in these two areas needed separate focus.

Evolution -- Volume Purchasing and Educational Pricing

In the software area, these evolutionary events and events in the industry led to volume purchasing agreements and educational discounts. Within the institution, the presence of better information and focused analysis led to better buying practices. The ability to track how much was being spent and where it was being spent led to knowledge in the minds of buyers in Purchasing. That knowledge was leveraged into negotiations for better pricing, based upon the age-old practice of "buying in quantity."

Vendors, too, played their role. As vendors and product manufacturers began to understand the role of Higher Education in their own marketing philosophy, they established ‘Educational Discounts.’ Often portrayed as contributions in the public interest, based upon their understanding that Universities had funding constraints, in reality, these programs were most founded in the vendor’s own best interests. We’ll discuss that further shortly.

The results of these two elements -- better purchasing and better vendor programs for acquisition -- led to software becoming more affordable for both institutions and their affiliates (students and faculty, primarily). And as software became more affordable, more people acquired it. The synergy of these two factors contributed to the explosive growth of desktop computing.

Evolution -- Site Licensing

The timing was becoming right for the advent of site licensing. The remaining three elements described above -- these focused on changes in the institutional role of information technology and especially the development of strong bonds between IT and Purchasing -- set the stage for new approaches. Both sides were ready.

The institution now had a better grasp on its software purchases. Critical mass, in terms of money being spent on software was realized, as millions of dollars were invested in software purchases. The growing bond between IT and Purchasing was, itself, reaching a level of maturity. Through collaborative partnership, Purchasing gained both an understanding and a higher level of expertise in information technology. And the CITO broadened their understanding of the importance of the acquisition process, and not only acknowledged the perspective, but began to develop expertise in many acquisition skill areas (negotiation, contracts, etc.). While each retained its separate focus and expertise in discipline, both sides knew more about the other and were able to develop a shared perspective of the entire process.

Software manufacturers and distributing vendors also became better at understanding the role of Higher Education and how to deal with it in terms of more global licensing agreements. They had a desire to solidify business relationships with universities -- most likely out of a desire to reduce their day-to-day marketing efforts and the costs of piecemeal acquisition negotiation and processing of orders. They developed a keen desire to gain market share, and to protect their gains. Essentially, they wanted long-term contracts in order to lock-in their customers, and make it harder for competition to get in to the institution. And finally, they truly began to understand the role of Higher Education in training the workforce (and decision makers and product users) of the future. They now understood that if they could ‘hook’ students on their software, they’d be able to keep them later when they moved out into the ‘real world’ and had to make decisions on which software to use; one uses what one feels most comfortable with.

The stage is now completely set for a new approach to licensing, as both the institution and the vendors are set to take a major leap.

3. First Contract: Novell in 1995

IU’s contract with Novell was the outgrowth of the institution’s ‘standardization’ on Netware. This standardization was quite natural -- at the time, Netware was truly the only game in town when it came to LAN operating systems. IU was spending a significant amount of money to fuel its growth in the area of distributed environments, and was getting fairly decent pricing under Novell’s educational offerings. This spending was at critical mass, and both Novell and IU were ready for something more strategic in nature.

A key factor was that IU’s 1995 ‘CIO’ (though not yet a full VP, as is the case today) came from Novell’s Higher Education Market Division. This ‘link of fortune’ was probably the key catalyst to both sides taking a pioneering step toward a site license. When both parties sat down together, the current institutional expenditures were great enough, in light of Novell’s desire to solidify a business relationship and protect market share, to facilitate a site license. IU collected together all it’s Novell expenditures (primarily for Netware) and leverage that money to get a global-use site license.

At the time, Novell also owned the WordPerfect Suite of desktop applications. In what might best be described as a tangential ‘throw-in,’ Novell included that suite into what was mostly a Netware licensing agreement. This after-thought -- and we truly believe it was just that -- was to have much larger ramifications for the future. Desktop computing was still growing explosively and the ability to take what was, essentially, a contract for infrastructure services and make it more pervasive in nature would play a big role in the evolution of both licensing and software use at Indiana University.

The contract would be for products only. In this first licensing agreement, we had not yet learned the lesson of covering the entire spectrum of product use -- support, training, and other services were not a part of our thinking at this stage of our evolution. But lessons learned by this ‘mistake’ were corrected in subsequent endeavors.

The Contract:

The agreement between IU and Novell was truly a pioneering effort. The result was a three-year contract that provided unlimited distribution of Netware for institutional servers, and the granting of a perpetual license for those installations (whatever quantity might be installed) at the end of the contract. Unlimited distribution of the WordPerfect Suite (academic version) was also included, and similarly, all institutional installations would become the property of Indiana University at the end of the contract. Distribution to affiliates for at-home use was also a part of this contract, making it possible for IU to freely distribute the products (on CD and via online servers) to students, faculty, and staff for their personal use. A non-perpetual license was granted for this type of use -- meaning that so long as the contract ran, any affiliate could use the products, but at termination, those copies were no longer legal.

As a pioneer, IU got some special ‘first-in’ pricing. As Novell later developed a larger Campus Licensing Agreement, they would reassess the value they felt they were imparting, and raise the price. As an interesting aside, the product groups within Novell -- and really, in all vendors we’ve subsequently dealt with -- were to criticize the agreement as having ‘cost the company money.’ This is a very facile view. What these people tend to do is look at the size of the institution and dream ‘green dreams’ of how much money could have been made, had per-piece pricing been employed. What they overlook is the fact that institutions would not ever make that level of investment, if older pricing models were the rule. These gratuitous opinions are really dinosaurs in today’s environment. One can only hope that they go the way of the dinosaur, and do not influence corporate opinion. The truth is this: Institutional Licensing is good for vendors for the reasons specified above. Lusting after how much money could have been had with per-piece pricing is akin to dreaming about what might have been, had one only picked the correct numbers in last Saturday’s Lotto drawing! {Here endeth the editorial comment on this subject.}

FLASH: Novell Sells WordPerfect!

In early 1996, Novell sold off the WordPerfect Suite to desktop software specialist Corel Corporation. Novell really had always viewed desktop products as a tangential business, and in corporate re-focusing activities decided to get back to their basics. The implication of this sale for IU was that we now had a new vendor to deal with in an existing (and transferred) obligation under contract. Corel had to honor the agreement, but had no understanding of the concept of institutional site licensing -- they were still at a previous evolutionary stage. So IU had to begin the process all over again, though this time we had a contract in place to help motivate the vendor to pay attention to us!

The process of ‘splaining things to Corel took about six months. Aside from contractual motivation, they truly were open to considering a new licensing approach (they were evolving nicely!) and the process was approached in a spirit of cooperation. Details were eventually ironed out and a second site-license resulted. The original three-year license was extended, to encompass the period of negotiations. So in effect, IU received an extra six months of coverage for Corel products at no additional charge. In the background, this was truly a three-party contract negotiation. Novell and Corel needed to discuss, in light of the IU-Corel negotiation, how they would split the income from the now-separate contracts. This element probably led to the process with Corel taking the time it did.

Everything was ‘AOK’ again. Though we did start to understand the limits of our licensing. When we negotiated the Novell agreement, remember that the primary focus was institutional use of Netware. The concept of individual licensing was tangential to the process, and the contract language really didn’t take into account the aspect of off-campus use. (Off-campus use of Netware was not a part of the license, nor was it necessary because individual users don’t really run their own at-home servers.) But this concept does apply to more pervasive and pedestrian products, such as those now under contract with Corel. In short, non-perpetual licensing for individuals was the way the Corel contract would read. Our attempts to acquire perpetual licenses for affiliate use were not successful; and one really can not blame Corel for not wanting to grant that ability, given the ‘pioneer pricing’ IU had in place. This event would later help shape subsequent negotiations … but that’s discussed later.

IU pioneered the Site Licensing Concept with Novell and Corel. Subsequent to their initial developments with IU, both vendors launched larger programs for higher education. Over 100 other institutions eventually took advantage of these new programs to provide software on their campuses, and to their students, faculty, and staff for at-home use.

Let’s Do It Again -- Sybase in 1995

Again, events developed to set the stage for another pioneering effort in institutional licensing. This time, with perhaps a more infrastructure component -- Relational Database Management Systems (RDBMS).

IU had acquired, in 1993, a limited-quantity simultaneous user license for Sybase’s RDBMS products in support of our first client/server university information application (the Financial Information System -- FIS). But the FIS was growing, and with that growth came the demand for more RDBMS licensing -- we needed more simultaneous users to be covered. In addition to the expansion of the FIS, the University was embarking on other client/server application projects in Library Automation. And on the horizon was the plan to develop client/server applications for Student, Human Resources, and other university information systems. The net of this is that IU needed institutional licensing for RDBMS to make it part of the infrastructure and remove it as a ‘line item’ in each project’s development.

Sybase as a company was also ready for a new approach. The company had begun to realize (along the lines discussed previously) the need to focus on the higher education market. They had put into place a high-level position to focus on the HE customer set and provide a strategy for marketing Sybase to higher education. As well, in the Spring of 1995, Sybase was in need of improvement in their sales. So the idea of a 7-figure contract (low 7-figure!) was appealing, as was the opportunity to leverage a new program to provide better results from the HE market.

The agreement between IU and Sybase was another pioneering effort. The result was a three-year contract that provided unlimited distribution of Sybase SQL Server and related products for institutional servers, and the granting of a perpetual license for those installations (whatever quantity might be installed) at the end of the contract. Sybase had just acquired PowerSoft, and thus a quantity of PowerBuilder desktop development software was also included. While the contract did not really include any end-user tools (for pedestrian use by all affiliates), it would have a broader effect beyond university information systems.

Through the institution’s establishment of a standard RDBMS for university information systems -- one that was used in the private sector -- pedagogical use developed quickly, as did the use of the a large RDBMS for departmental applications. Essentially, IU leveraged it’s university IS decision to provide needed database tools for use by faculty in instructing their students in the use of a ‘real world’ tools. As well, a number of large research databases were converted to Sybase, as support in the environment grew rapidly. And finally, the availability of a large-scale tool at no additional cost allowed an explosive growth of departmental information data management systems.

Key to making this happen were two new attributes that had not been included in the Novell/Corel agreements -- Training and Support. The Sybase contract included a substantial component of training credits, which were initially targeted toward training CITO database administrators (DBAs) so that they could continue roll-out of the university information systems. But IU leveraged these credits to establish a program of education for non-CITO staff in departments. These staff were charged with maintaining departmental information systems for departmental IS or research purposes. This ‘fertilization’ of the environment, coupled with freely available products led to a huge growth of distributed data management at IU. IU and Sybase partnered closely in the delivery of training and the development of this rich model through excellent support services to these newly placed DBAs. Aside from product, perhaps the greatest success of the Sybase Licensing Agreement was this enrichment of IU’s data management environment.

In fact, the planned advancement of the larger institutional information systems was delayed. Library, Student, and HR systems were not deployed during the life of the Sybase agreement. However, the expanded use of product in pedagogical and distributed information systems was so great as to have made the investment in Sybase pay-off in spite of the lack of broader institutional use. And though these large systems did not take-off, many smaller more focused university information services and decision support structures benefited from the presence of Sybase (and PowerBuilder) products.

IU pioneered the site licensing concept with Sybase for RDBMS products. Sybase would eventually offer similar programs (though at an increased price) to several other institutions. While the number of these was not as great as what resulted from the Novell and Corel agreements, the reason is probably more based in Sybase’s overall position in the RDBMS market, rather than a lack of perceived value in institutional licensing in the industry. Case in point: Oracle subsequently developed their own higher education marketing program and their own institutional licensing agreement. Perhaps we, at IU, may be so bold as to think our work with Sybase influenced industry directions a second time.

Regardless of that, IU did learn more from this second foray into institutional licensing. We had learned the value of support and education to the ultimate success of any site license. Simply put, in order for the institution to get the most out of its site license, people have to be able to use and value the products. Education and the quality of support are key to that process. This lesson shapes IU’s position in licensing negotiations today. And would have a key effect on the next major effort IU would take.

Again, events developed to set the stage for another pioneering effort in institutional licensing. This time, with perhaps the most broad implication to-date, for products demanded most heavily by our institution.

IU had acquired a good set of experience in both how to acquire institutional licensing and in the benefits to the institution of such licensing. We looked next to where huge shares of our software expenditures were going, and found ‘critical mass’ existed potentially for Microsoft products. Microsoft desktop tools (essentially, MS Office) were the tools of choice for a huge share of our user base. In spite of the Corel agreement that made the WP Suite available at no additional charge, users were selecting Microsoft products with a 5-to-1 preference -- even though they had to pay for them! In the server OS area, LAN administrators -- especially on the IUB campus -- were selecting Microsoft NT-Server over Netware. Over 40% of the total LAN OS market was, in 1997, using NT-Server (nearly 80% of the IUB campus LANs were running NT). And that number was steadily growing each month, as LAN administrators evaluated the benefits of NT over Netware and migrated their servers to the new OS.

IU was spending an amount approaching $1-million dollars per year (projected for FY 1998-99) to acquire MS products under the Select program. This investment was growing steadily by about 20%-25% each year. Not included in this figure was application software bundled with hardware purchases (a common practice with desktop computer vendors).

But most of this investment was for new licenses; we weren’t seeing much investment in upgrades. The result of this was that over time, IU had developed a ‘horribly diverse’ environment in terms of the many versions of Microsoft OS and applications to be supported. This was creating a tremendous support burden. Each new release required an investment in support, but the CITO and Local Support Providers (LSPs) could not abandon their investment in previous versions, because so many existed. We needed to standardize within the Microsoft product set on the more recent versions -- but could not do so with ad hoc purchases.

Another aspect was individual use. IU was seeing -- despite the lack of a formal requirement for computer ownership -- that our student population was equipping itself with at-home computers. In 1997, about two-thirds of the student body had a personal computer at home. And the incoming crop of freshman each year featured a growing percentage of ownership -- up to 80% projected for the incoming Class of 2002. This, coupled with most all faculty and many staff owning personal computers, presented a huge demand for software. And of course, no individual purchases for at-home use were included in our figures

; students, faculty, and staff with personally-owned computers were buying the software out of their own pockets.

At this same time, Microsoft itself as a corporation was reviewing both its Higher Education Market strategy, and its offerings to that market. In early 1997, Microsoft established a marketing group, at a high level within their corporate structure, to focus on the HE market. Microsoft realized the facets discussed earlier -- the need to solidify business relationships with HE, the need to gain and keep market share in HE, and especially the desire to equip tomorrow’s workforce with both the training in MS technology and a preference for it. Much is made of Microsoft’s approach to marketing and the IT business, and IU is not sitting in judgement upon that nor taking a side in the national debate. However, it does seem to us that Microsoft is evolving in the way that other software vendors have evolved, and as highlighted by this paper.

Principles of market approach aside, Microsoft was also seeking to re-evaluate their concurrent licensing offerings to higher education. In 1997, Microsoft decided to end their concurrency programs. As well, Microsoft appeared to be re-assessing the role of their current HE offerings -- Select and Open. -- perhaps viewing them as growing ‘stale.’ Whatever the reasons or motivations, the stage was set for Microsoft to explore new approaches to licensing to the higher education market.

Persistence paid off for IU in this regard. For almost a decade, IU had been trying to interest Microsoft in a site licensing program. And this effort from IU increased once we pioneered the concept with Novell, Corel, and Sybase. Every few months, IU would "knock on the door" at Microsoft to try to get their interest in an institutional licensing program. That knock was not answered for years -- until events within Microsoft readied them to answer. And in the summer of 1997, IU knocked again -- and this time, Microsoft came to the door! Discussions commenced in July of 1997, and negotiations began during the Fall. By January 1998, the concept was set for a new licensing program, and IU and Microsoft signed the first Enterprise Licensing Agreement in March of 1998.

The Enterprise Licensing Agreement (ELA) between IU and Microsoft was a huge pioneering effort. The result was a four-year contract that provided unlimited distribution of Microsoft products for use within the institution and use by affiliates in their homes. Unlike previous agreements however, a key component was the granting of perpetual licenses for student affiliates. In this way, once they receive the software under the ELA, it is theirs to keep and use even after they leave IU.

The contract covers a variety of desktop products:

• Desktop Suite: Office for Mac and Intel
• Upgrades for Operating Systems to any 32-bit (and beyond) OS Microsoft releases
• FrontPage (web page development tool)
• Visual Studio (desktop data application development tool set)
• Outlook Email Client

A web browser is also included. The debate rages on as to whether this is part of the OS, or a separate product. IU’s contract -- while not stating in any terms that would embroil it in the ongoing debate -- covers either situation.

The ELA also provides for additional products, though not in unlimited quantity. These include 1,000 licenses for a combination of NT Server, SQL Server, and Exchange Server.

Other significant facets of the IU/MS ELA:

• Premier Support Program enrollment
• Extensive support for the analysis of Exchange as IU’s messaging infrastructure
• Training Credits for MCSE-program courses (provided by the local MS reseller)
• Certification Test vouchers -- several hundreds of them (provided by the local MS reseller)
• On-site Consultation (provided by the local MS reseller)
• ‘Grandfathering’ of institutional purchases made in the six-month negotiation period
• Three months (April ’98 through June ’98) free distribution -- contract started 7/1/98

IU and Microsoft pioneered the ELA, and it was the foundation of Microsoft’s recently announced Campus Licensing Agreement (CLA) program. The IU ELA is different than the CLA in many ways, including perpetual licensing for students and a number of other aspects. But it also paved the way for other ‘special’ licensing agreements modeled closely on the ELA, including the one signed by the University of Texas system in the Fall of 1998. While Microsoft’s focus in higher education seems now tightly aligned with the CLA format, they have stated a desire to continue to explore ELA-like opportunities.

Since March 31 of 1998, when distribution under the ELA first began, IU has distributed well over 90,000 products to its constituents for use on campus and at home. The most popular package is, as one would expect, the Office suites for both Mac and Intel platforms. But the release of Win98 this past summer was also popular. Upgrades for both Win98 and WinNT have been significant, and it appears that our environment is making progress in upgrading to the most recent releases of OS, based upon their ‘free’ availability. Today, the preference for Microsoft desktop (pedestrian) applications has increased to about 7-to-1. There is conversion from Corel, but most of this is growth of new users.

NT Server continues to grow in popularity, over Netware. We believe that this trend was well underway prior to the presence of the ELA (as stated previously). But again, the availability of a ‘no additional charge’ conversion path has spurred migration. NT is now the strategic LAN OS for Indiana University, and support for it in terms of LSP skill and CITO services grows every day.

Since the ELA was signed, over 100 topic certifications have been achieved by LSP staff on all campuses. Training courses -- both those provided under the ELA and those provided internally by the CITO -- have seen continued growth in demand. Our LSP community is quickly getting better skilled in supporting ELA products, and this certainly enriches the environment and makes the ELA products more popular with our users.

There has been a wealth of sharing of information about this contract since Day One. All aspects of the contract, its impact on IU, and the CITO’s support are not going to be covered in this paper. But an elaborate web site was put into place and has been updated continuously. It can be found on the web at http://www.indiana.edu/~msela. We refer all detailed questions about the IU/MS ELA to this source of information. However, we do point out that communication about the ELA was an important consideration as we embarked upon its implementation. We suggest that other institutions follow suit when large-scale enterprise licensing agreements are instituted.

Many questions arise as to what impact the ELA has had upon IU, in terms of diversity or product offerings. Later in this paper, we will address and hopefully dispel many of the ‘urban legends’ surrounding the ELA at IU. But points of fact concerning previous contracts and product deployments are:

IU did not renew our Novell agreement. This was not so much due to the presence of the ELA (though we do not deny it had an effect). Rather, demand for Netware was and is shrinking as LSPs make decisions to migrate to NT. The ‘critical mass’ in terms of expenditures did not warrant signing a new, unlimited license. This was particularly true in that the cost of a renewal was significantly hire than the ‘pioneering pricing’ IU had in place under the first agreement.

IU did not renew our Corel agreement, in terms of a broad site license for unlimited distribution. Again, this was due to shrinking demand for WordPerfect that was the case long before the ELA was signed. Again, ‘critical mass’ in terms of budget devoted to meeting demand for Corel products does not exist. However, IU has worked closely with Corel since the ELA was signed, and just recently (November 1998) put into place an alternate program for providing Corel products to IU for institutional licensing and affiliate at-home use. This is not a site license, but rather, a volume purchasing agreement that makes Corel products available for a very low price. The purpose of this new agreement is to recognize that a portion of our population does, in fact, prefer Corel products and does not desire to migrate. However, there is no institutional commitment to Corel, so the costs are borne by the individual user. Corel has wisely priced the product under this new agreement at a very low fee, and the CITO is subsidizing production of CDs to keep the costs in line with what students can get MS products for under the ELA. Details of this program can be found on the web at http://www.indiana.edu/~corel.

Netscape was not licensed, as the company gives away its browser product to users. It is interesting to point out that IU users preferred Netscape to Internet Explorer by an almost 9-to-1 margin prior to the ELA. Nearly nine months later, after thousands of copies of IE have been provided to our users, the preference is still 9-to-1 … if not slightly beyond that. IU has made the commitment to provide both browsers, in prominent location (i.e., on the desktop) so that users can effectively make their own choice.

In summary, we believe the ELA has been very effective for IU, and are very pleased with all aspects of the success of the program.

Perhaps many of us remember the line from the movie Caddyshack: "Be the ball, Danny … Be the ball." What that means to us, in terms of standardization, is to not try to force standards upon our users. Rather, it means to let them lead us to standards. We make our decisions about standards based upon user demand, and then try to put into place an environment that both supports that standardized use and makes it easier for people to make standard selections when they’re decide what software (or hardware) to select.

Using the tools described in the first section of this paper, the role of the CITO and Purchasing should be to assess what users want -- what they are selecting, and perhaps why. When one finds a plurality or majority of users are selecting a given product, then through institutional licensing and the development of a rich support environment, we can declare that product a standard.

Once that decision is made, and licensing and support are in place, then another key factor is getting the ‘standard’ product into the most users hands that you can. The mistake many institutions make in trying to implement standards is to centrally and perhaps in isolation select a product and call it ‘supported.’ While at the same time, saying alternatives are ‘not standard’ and are thus, not supported. This, as most support-related publications and experiences suggest, is not a good or happy approach! But if you select what people are using and make it easy for them to have it, then you encourage a standard behavior, rather than discouraging non-standard behavior. If there is a key point to hit-home -- it is this: Support the standard well and make it enticing.

At IU, one thing we’ve done well is that we get standard products into people’s hands. We deploy those products in our on-campus facilities. And we make acquisition of software very easy by first making it inexpensive or free (licensing) and then putting it on a convenient format for distribution (CDs and Online). The success of the ELA is largely demonstrated by the number of products people have received -- most of them on CDs. We sell the CDs, not in an effort to recoup licensing costs, but instead to simply cover the cost of production and distribution of the CD. At this time, we sell any CD for $5 each. At that price, we’ve found people sometimes are buying software they don’t immediately need, because what the hell -- it’s there for such a low price!

We’ve also given the CDs away for free at special events. When we signed the ELA, we had a one-time kick-off effort to give away some CDs. We planned to give away 6,000 CDs on our campuses, over the course of a four-hour program. Well, in less than an hour, those were gone! And we ended up giving away nearly 7,000 vouchers for free CDs. We subsequently gave CDs (Microsoft Office and our own IUware CD filled with other useful software) away for free during Summer Orientation to incoming freshman. Thus ensuring that not only did they not spend their money to buy products that we had, but also to ensure that all of them had the standard. Finally, we give away CDs to our Local Support Providers (LSPs), for use in departmental installations.

Our Online software distributor is also getting a new look. Long a Netware or NT-based server, we’re now moving to Web-based distribution -- like everyone else in the industry. We think this is an excellent way to distribute small products, like updates and drivers and utilities. And for on-campus distribution, where high-speed networking is in place, it’s also a good tool for LSPs. But this service is simply an augmentation to our CD distribution, which is our main focus. Right now, about 85% of our software distribution comes from the CD program. We anticipate that it will become stable in the 85%-90% range in the next year. The online method is simply our way of appealing to a portion of our user base. Which segues nicely into a discussion of avoiding exclusivity.

One thing that popped up when we signed the ELA was a concern that by doing so, we’d exclude other products from our environment. Perhaps part of this had to do with our announcement, wherein we said "IU signs exclusive agreement with Microsoft." Well, bad choice of words! We meant ‘exclusive’ in terms of first of its kind. Not in terms of excluding other options!

As an institution embarks on an effort to standardize, that effort is driving primarily by a desire to contain the costs of supporting the IT environment. And diversity in the extreme certainly has very high costs (hidden to many) in terms of support. But standardization can not be totalitarian in nature. Majority rules … but the minority has the right to exist, especially in academia!

Institutions can maintain a rich environment by ensuring that diversity in some measure can continue to exist. This can be accomplished by developing complimentary licensing programs for competing products. The new Corel program at IU is a great example of this. As well, support for competing products must continue to exist, in some form, even in the standard environment. Again, user demand should be your guide. At IU, even with the ELA, Netscape continues to be the web browser of choice -- by a 9-to-1 margin. This is less an example of ‘supporting the minority’ and more an example of how in spite of an institutional license, other ‘standards’ can exist. But in the case of minority-use products, support for many must continue to be provided.

Perhaps the CITO itself can support these products, through use of targeted application areas where diversity is the rule. Perhaps user groups can emerge, where collectives of users of less-demanded products can support each other. And perhaps, one of the costs of not using the standard is that those who want the freedom to use them must seek out their own support. The key here, again, is that the CITO (and Purchasing) should not try to dictate what software people buy and use. Rather, they should help people to make good choices based upon the University’s resources, and to understand the total cost of ownership of any software purchase.

Institutional licensing does carry with it some perception problems. Sometimes perception is truly reality, when an institution adopts a totalitarian philosophy toward standards implementation and enforcement. But usually, as was the case with IU’s ELA with Microsoft, there is a great deal of myth and urban legend involved. Let’s use that as a case study on the issue of Exclusivity.

Again, in our exuberance in actually bagging ‘The BIG One’ after so many years, we allowed ourselves to say we were a "Microsoft University.’ As a result of that statement, and the current state of widely disparate views about Microsoft Corporation, a number of popular misconceptions about our ELA arose. Let’s look at each, and the facts that don’t support them being true.

First off, we never said that Windows was our standard for desktop computing. What we did say was that it was our standard for Intel desktop computing. The decision about Mac versus Intel -- or better put, the war over the issue -- has been raging on our campus, in higher education, and in the industry in general for well over a decade. The ELA was barely a pop-gun flash in that war. But many believed, including some of our closest internal advisors, that we were setting the stage for the abandonment of Macintosh at IU.

Prior to the ELA, Macintosh installations had been on a slow decline toward settling in at about 10%-12% of our installed base. There were a lot of reasons for that, explored ad infinitum in all sorts of literature and conjecture on the subject. That aside, at IU the basic reason for the decline had to do with price and the fact that most users made use of pedestrian applications. For these applications -- word processing, spreadsheets, web browsers, telnet clients, etc. -- there is really no difference between the Wintel and Mac interfaces. So people chose what was cheapest, and often what their neighbor had. Mac owns certain areas still -- graphics especially. And IU -- especially the CITO -- is very interested in maintaining the diversity for the value it provides.

Since the ELA, the decline of Macs has not changed at all. It’s still slow and heading toward a 10%-12% base (which we believe is true in the industry market). But there have been some notable conversions -- but for reasons not having to do with the ELA. Our Student Technology Centers (STCs) at IUB have, historically featured about a 60/40 split, Intel to Mac. Never did IU’s campus population hit that level, though in the early 90’s, Macs did occupy about one third of faculty desktops. And since faculty desktops translate into teaching preferences -- and Mac faculty taught more classes in those days -- we established our STC deployment in support of that fact.

However, in the last few years, faculty desktops are migrating (for their own reasons -- god love faculty!) to Intel. At this time, we believe that about 80% of the faculty have an Intel computer on their desk, running a Windows OS. Demand for our STCs for instruction followed that trend, as more and more faculty integrated technology into their instruction. As a result, in 1997, we saw tremendous compression (i.e., unmet demand) for Wintel STCs, and lots of empty seats in our Mac STCs. So in December of 1997 (before our STC staff knew about the ELA effort) they devised a plan to convert two labs on campus from Mac to Intel, shifting the ratio to about 64%/36%. Not an earth-shaking change, mind you … but it was announced a month after the ELA was announced. Hence, the rise of the legend that the ELA was going to cause Macs to disappear.

We anticipate, based upon the same demand principles, that we’ll convert more Mac sites to Intel. But it will be slow -- a lab or two (out of 40+ total) each year. Until we match our supply of technology with demand by faculty and students.

Also, when Apple released the IMac this past August, it did generate some interest as an at-home device. It sold fairly well (though our data shows that students or their parents are still choosing Intel at a 10-to-1 rate). But our users were most happy to couple their new IMac machines with the Office’98 CDs in our bookstores!

Again, this was urban legend. IU never had any intent to push standardization to totalitarian levels. As stated, we adopt standards based upon user demand, and work to establish complimentary licensing for alternatives. The new Corel program is proof of this. And for the third time, Netscape continues to hold a huge preference advantage in our user community.

Ignored in this myth are two factors -- the state of ‘pedestrian’ desktop applications, and the overall place they play in the IU environment. Concerning the first, we’ve discussed previously the fact that for the largest share of users, if you’ve seen one word processor (or spreadsheet, or presentation package, or web browser) you’ve seen them all. The vast majority of our users don’t care about product differences -- because in reality, there really aren’t any. Market competition is fierce, and an outgrowth is that most products look, feel, and perform the same. In pedestrian applications.

Concerning the second, one needs only look at IU’s STC software list to see that the products covered under the ELA comprise about 5% of the total number of packages supported! While perhaps as much as 80% of the desktop use is for these pedestrian applications, the pareto principle holds nicely. 20% of the use is with 95% of the products. And most are not in areas where Microsoft even offers products! For verification, point your web browser to http://uhura.uits/indiana.edu/apps/public/apps.cfm and see the whole, diverse, list of products supported in our STCs (and hence, on campus).

Two things wrong with this.

First, IU has long understood how support for IT in higher education works, and has been a leader in developing a support model and applying that model in practice. We understand that as use of IT progresses to the point of increasing investment in it, investment in support is a fundamental part of the deal. We have no plans to cut support resources. Instead, our IT Strategic Plan calls for increasing that support (refer to http://www.indiana.edu/~ovpit/strategic/).

Second, there is some truth to the fact that we wanted to contain software support costs. But the ELA was an effort to curb support costs of Microsoft products in specific! As we discussed earlier, we were getting eaten alive with the diversity in the MS-OS and MS applications deployed and used. We wanted the ELA so that we could standardize within the Microsoft product set -- not to standardize (necessarily) upon Microsoft products.

There are many more myths surrounding IU’s ELA with Microsoft. But most all are simply perception problems resulting from lack of factual information and often made by people who don’t even know where The State of Indiana is on the map! We have probably exacerbated these perceptions through our choice of words to announce the ELA, and that we lacked understanding truly how ‘shark-filled’ the waters were surrounding Microsoft in higher education. C’est la vie. That is another paper altogether.

Let’s take a quick aside to talk about how the lessons learned from our software strategies might be leveraged to address more broadly other aspects of the IT environment. One particular area is desktop hardware. We’ve made some progress here, but will be making more as a result of action items in the Strategic Plan.

The university has worked to establish volume-based purchasing agreements and educational pricing programs with hardware vendors. Hardware today is very much a commodity, but due to its expense, it is still bought in piecemeal fashion. But we’ve found that standardization on commodity desktop hardware can be influenced -- and the overall effectiveness of our hardware expenditures increased -- if we can leverage university purchases better. IU has a very distributed financial responsibility system in place. Departments make their own selections of hardware and buy whatever they can buy within their budgets. But despite the presence of institutional the aforementioned pricing agreements and discounts, we still aren’t leveraging our money to its fullest extent.

The very nature of the distributed buying, and the more ‘autonomous’ approach to hardware funding has not allowed us to achieve the ‘critical mass’ mentioned as a key element. However, we did have one experience with this where we were able to achieve critical mass, and as a result got a very good price for desktop hardware (in volume) and saw some movement toward standardization on more recent desktop technology. A key thing to keep in mind, due to the commodity nature, is that perhaps the name on the case is less important than ensuring that what’s inside is standard and current.

In FY 96-97, we did have a positive experience that we hope to repeat and perhaps establish a philosophy of institutional hardware purchases. In that experience, the CITO funded the purchase of about 400 machines for the Bloomington campus. With this large volume, we got an excellent price (about $500 less than current piecemeal pricing, even under discount). We offered that pricing to departments who wanted to invest their own funds, as well as individuals. The net result was that the 400 machines sponsored by the CITO ultimately grew to over 1,000 machines acquired in total (about 900 on campus). Suddenly, there was a ‘standard’ for hardware!

We hope through action items of our IT Strategic Plan to get the institution into a better program of life-cycle replacement of hardware. As the need (and budget to support the program) comes into being, we hope to be able to again "pool" purchases across the campus(es) and get into a stronger negotiating position with hardware vendors. This effort may, again, be best spurred by a one-time investment from the CITO or central university or campus administration. It is our hope that this time, if we buy 1,000 machines with special funds, another 2,000 machines might be purchased with in-place departmental funds. That would, indeed, result in more standardization in desktop hardware.

Back to institutional licensing -- and perhaps specifically the ELA -- let’s look at what it has meant to Indiana University to be a pioneer in this area. Like all things in life, there are some benefits to being first, and some hardships. Briefly, let’s cover those now as a conclusion to this discussion.

Being first will always carry with it some positive perks. As we saw in all cases of our site licensing ‘scores’ we received preferential pricing. Usually, this was the result of excellent negotiations on our part based on a firm understanding of how much money we had to spend. Sometimes, it was the result of IU having a better grasp of the situation than the vendors, and thus better knowing the ground. Once vendors saw how it worked (and how widely accepted it was at IU, and thus how valuable to IU) they often thought they could get more money for subsequent purchases. It could have worked the other way in any of these cases. It could have been that the price on the market of the software would drop, and we’d be locked into a contract that cost us more than piecemeal purchases would have. Maybe we’ve been good. Maybe we’ve been lucky. But our experience is that it pays off financially to be first.

This is also true in terms of …well … terms. We’ve seen that we get a better set of services and attributes of licensing by being first. In the case of the ELA, our pioneering efforts include perpetual licensing and CD distribution. These are not features of Microsoft’s broader CLA program. But again, Microsoft is doing ELA-like deals (Texas), so it’s not as if we got the only good deal on terms to be had.

We have often gotten ‘extras’ thrown into the bargain. Training, support, other products, etc. This may be less a benefit of being first and more due to the fact that these things are important to us, and hence we insist on them.

We’ve gotten some good headlines too … lots of publicity. In the case of the ELA, IU was featured in several videos that Microsoft made. And of course, we managed to have Bill Gates come as a speaker to IU; a session that later showed upon CSPAN nationally. The ELA helped increase IU’s visibility as a leader in IT in higher education, and in higher education in general.

We’ve also been able to help some our colleagues get a better deal on their software purchases. Usually as a result of programs that follow our pioneering efforts, but in some cases due to helping them directly understand the issues as we’ve learned them. And we’ve received some praise in our community for having made it easier to own a computer and use it, as a result of the pervasive distribution of software to IU affiliates.

But every silver lining has a cloud …

Being first means you have no model to follow, and you make mistakes. You later see that you wish you’d done something differently, that perhaps did get included in the broader offering that followed your pioneering efforts. For us, these lessons were more internal and applied to the next deal, rather than wishing we’d waited on the current one. The best example is the inclusion of training, support, and other services when we negotiate site licenses. As we’ve stated many times, these factors are as important to us as the products themselves. Because we’ve learned that without them, you don’t get the value out of the products.

When it comes to headlines, there are negative ones too. We learned a lot about how national media (mostly the online variety) has their own agendas. And making out a deal with large corporations as a good thing usually isn’t one of their agenda items! We had reporters take what we said out of context, or spin our comments in support of myths and contrary views. One reporter even told us: "Hey, my editor isn’t looking for happy news. He’s looking for controversy because that sells." He added: "Don’t take it personally." Well, easier said than done!

We learned how the media, in search of a story will blow things out of proportion. The best example was a comment that appeared on one of our newsgroups -- iu.gripe. This statement was made, concerning a discussion about the forthcoming distribution event to kick off the ELA: "Gosh, maybe someone ought to go down there and protest." Now for perspective, this newsgroup has an active participation of about a dozen ‘regulars’ who have many gripes about everything. It’s recreation for them. There was no follow-up comment to that posting. However, in the online story that appeared the next day -- and was taken up by every other reporter in the first days following the ELA announcement -- protests abounded! The first story said: "Protests are planned." Subsequent stories converted that to: "There were some protests." And we even saw one that said: "There were widespread protests."

Sort of makes you think what one reads in the media needs more careful examination?

And finally, not all of our colleagues or constituents are pleased by our pioneering efforts. Some debate the appropriateness of university site licensing and standardization. Some carry standardization, in their minds to totalitarian levels. And some simply wish they’d been there first. But riding the bow wave can be rather brutal at times. One must set their keels deep if they seek to be pioneers.

In the case of the IU experience, we believe being a pioneer has worked to our benefit, because it has been of benefit to our students, faculty, and staff.

We believe IU has achieved both goals we’ve had in place: Abundance of software available for use by our affiliates, and also building an environment that provides excellent support. We believe universities can implement standards, based upon demand and then the effective delivery of demand-based-standard products that entice users to use the standard, rather than punishing them for not using it. We believe that diversity -- in terms of competitive products -- can still be achieved even in an environment that seeks to promote and encourage standards. And that the support savings from standardizing on the few products that make up the preponderance of use can actually be reallocated to the support of other applications. This rich and diverse environment is a goal of Indiana University. And we believe as a result of our work in strategic planning, software licensing, and support model, we’ve achieved our goals and positioned ourselves to continue to meet and exceed expectations at IU.