First Monday

A case for open markets in education by Steve Midgley

This paper outlines a new approach, based on existing research, experience and practice, to improve public education and educational equity by making more effective technological systems for education available. It argues that effective administrative and educational data systems are essential to support widespread educational reform. These systems can best be sustained and developed within a competitive marketplace using open and closed source tools combined with open communities of practice and expertise. These innovations can help to ensure that such markets, tools and communities can evolve to support a very large number of districts undertaking comprehensive organizational reform.


Technology markets and education
Market impediments to education reform: The marketplace in educational technology
The alternative: Open markets
Openness for education
Open systems: Data integration
Open communities
Open philanthropy
Many markets




Public education in the United States is often maligned as broken and providing ineffective education for too many of its students. As the system prepares much of the student body for university and the country’s advanced workforce, its impact on the economy and social welfare cannot be overstated. Yet public school many students in the U.S. compete rather well with other countries. The problem is that the historical achievement of U.S. students is largely predictable according to disturbing variables such as race, class and first language. Delving into these statistics, white students consistently outperform their minority counterparts. For example, white students in fourth grade in 2003 were competitive with students in the Netherlands, while similar minority students are competitive with Armenia [1]. Achievement gaps between ethnicities and socioeconomic groups are the rule, not the exception in America, and they have persisted for generations [2].

Resolving these gross inequities remains critical to achieving economic well–being and equitable access to opportunities and resources. Social justice demands that all children receive a high quality education. Attempts to narrow this achievement gap through reform of various kinds — whether of schools or of school districts [3] — has, thus far, met with fairly low levels of overall success [4].

This paper outlines a new approach, based on existing research, experience and practice, to improve public education and educational equity by making more effective technological systems for education available. It argues that effective administrative and educational data systems are essential to support widespread educational reform. These systems can best be sustained and developed within a competitive marketplace using open and closed source tools combined with open communities of practice and expertise. These innovations can help to ensure that such markets, tools and communities can evolve to support a very large number of districts undertaking comprehensive organizational reform.

Archimedes said “Give me a lever long enough and a fulcrum on which to place it, and I shall move the world.” The educational community clearly needs such a lever that will translate and scale local success into success for many more educational organizations across the country. Administrative and educational systems and services in education span the full range of public education activities from student information, human resources, finances, educational content/strategies, learning management, assessment, professional development, strategic planning, accountability and program evaluation. Currently these systems are often “locked–in” to agencies: the cost and risk of replacing an entire system is so great, that it is very hard to justify. As a result systems remain in place even when they are not effectively meeting the needs of an organization. The prospect of improving the markets which create this and other problems and driving the creation of new systems and services with technology presents an important opportunity of making relatively small changes in strategy, which can lead to large changes in market behavior and outcomes for educational agencies. Improvements in these systems can help districts and other educational agencies improve their organizational and instructional capacities to help students learn at scale.

Recognizing the power and effectiveness of open and “peer produced” systems to create higher quality information in diverse arenas, this article suggests that similar approaches can have a positive impact on public education. It aims to explore where such open systems and open, peer production communities might be best applied to create maximum impact for education. To the end of laying out this argument and explaining the connection between open technologies and education reform, this paper begins with a discussion of the challenges created by technology in public education. It continues by addressing the impact of openness on markets for technology and then explores how to bring these same opportunities to bear for education. The article focuses, specifically, on the market for data integration and explains the way that open technologies can remedy gross difficulties caused by existing technology business models that impede reform. The article concludes with a discussion of open communities and open philanthropy to support this model for technology–based reform. The theory described in this article informs a practical project, known as the Open Educational Data Systems Initiative, that is designed to implement and test these ideas in practice.



Technology markets and education

One of the central problems for a public organization such as large school districts in achieving intended outcomes is the collective and disparate culture of which it is composed [5]. There is mounting evidence that when strategies to address the issue of changing an organization’s culture are brought to bear in an focused, effective manner within a school district, the results are remarkable and highly positive for students. Many organizations currently dedicated to assisting districts in attacking this problem focus on strengthening leadership and management skills, undertaking rigorous strategic planning and evaluation, rebuilding processes to be more efficient and effective, deploying appropriate professional development, developing coherent instructional systems and technology and engaging with the communities that compose the district in the reform work. A number of school districts have begun to see results using these practices, including Norfolk, Virginia [6], Sacramento, California [7], Aldeen, Texas [8], Clarksville, Tennessee [9, 10] and Gilroy, California [11, 12].

Yet for all the substantial accomplishments of individual districts and schools changing their local environment and culture to benefit students, many more educational agencies across the country have not generated the organizational “will” to undertake any major change initiative. They are unaware of the promising possibilities or lack the know–how to implement change. Even where reforming education in a district–by–district manner has worked, scaling it to the national level may be far too costly and cumbersome.

In order for this ultimate objective to be obtained the success of individual districts and the acquired knowledge, skills and capacity are not, alone, enough to make change across the country. There are forces around the districts that must be supportive and receptive to the change. Some of these forces and cultures may be addressed through community and political engagement and some reform support organizations are already embarking on achieving results in this area [13, 14, 15, 16].

The sector’s ability to translate these accomplishments and progress into widespread results remains a major barrier. Progress in related areas of other fields provides insight and hope that this barrier can be overcome. In particular, based on low transaction costs and low latency of the Internet, very rapid progress in “peer production” technologies or “user–generated content” has been made in diverse areas, from commons–based encyclopedias, shopping and real estate (e.g. Wikipedia, Amazon, eBay, and LoopNet). One of the promises of these solutions is that they use simple communication technologies to radically enhance the quality and quantity of information available to their users. And, critically, they allow their users to participate (sometimes wholly) in the process of continually improving this content [17, 18].

Recognizing the power of such “peer produced” systems to create a higher quality of information in diverse areas, perhaps such tools can have similar impact in public education? If so, where would they be best applied to create maximum impact? Recent Internet history suggests that areas related to technology are generally the most fertile ground for the development novel online solutions. These areas seem to grow more quickly and to support innovation earlier (e.g. AlterNet at, BITNET, LISTSERV, Usenet, Slashdot at

Several communities have begun to address learning and content management systems, for both higher education and K–12 systems (e.g. IMS at, Sakai at, uPortal at, Moodle at Less attention has been paid to technological systems for administration, which run many of the vital systems for these organizations (Kuali [] is notable) [19]. Yet administrative systems are in great need of improvement and information about these systems is not widely shared, even within technology communities inside the education sector.

Districts who use these technologies are often poorly equipped and skilled to make effective use of what technology they do acquire.

It is not only the school districts that are under–resourced, many of these education technology marketplaces are as weak as the districts they serve. While many vendors and products within these marketplaces are innovative and produce high–quality systems, the methods of marketing and sales as well as licensing, support and integration all display elements of dysfunction. Districts who use these technologies are often poorly equipped and skilled to make effective use of what technology they do acquire. Very often the acquisitions themselves are the result of flawed processes, which result in the implementation of an inappropriate solution entirely.

There has been some success at improving the situation but none of these existing approaches scales well or attacks the problem systemically. Providing skilled expertise to districts individually to make better selection and implementation decisions has a reliably desirable effect [20]. Issues with pricing models (not just the overall cost), purchase decisions, project management, systems integration, implementation failure, innovation rates and post–sales buyer leverage all create major problems in obtaining and make good use of educational technology, especially for administrative systems. Philanthropic investment in supporting more effective vendors to out–compete the current marketplace has had some effect as well, but the funds invested in a single company enhances the market by only a small amount — the market itself changes only very slowly in response to a single new product or service offering [21].

The problem lies in the fact that the markets themselves are not fluid and truly competitive: vertical market applications prevail with the associated high switching costs, bundled service and costly licenses. It has become very expensive and risky for districts to change to new systems and they often make do with what they have already implemented. In some economic situations, districts cannot afford to change their system and cases exist where districts have remained locked into a particular solution for 25 years or more [22]. Targeted philanthropic, governmental and academic investment and support for key areas of inefficient competition could radically change the terms under which companies compete to serve school districts and school districts could receive significantly greater value of systems and services for the same or lower costs.



Market impediments to education reform: The marketplace in educational technology

In the current market for educational technology, there is little innovation, not because vendors do not desire to improve their offering or competitive stance but because the market is largely closed across products and systems. Market conditions prevail that make it difficult for districts and vendors to innovate and this, in turn, impacts education.

For example, it is not possible for districts to easily switch between systems: poor data interoperability, long amortization of license costs and high entry costs for new systems all tend to cause districts to hold on to existing technology rather than explore alternative solutions. With limited choices about their future technology, districts have little ability to apply accountability for results and utility regularly upon a vendor. Most districts can only afford to change their systems (whether curricular or administrative) every 5–10 years. Unlike corporate markets, new district organizations come into the market only very rarely, and existing districts very rarely close or radically reorganize. As a result, the demand for systems comes almost exclusively from organizations already using an existing technological solution. In fact, we see rapid innovation and adoption primarily around new technologies, rather than existing solution categories [23]. There are simply not many opportunities for vendors to be held competitively accountable.

There are simply not many opportunities for vendors to be held competitively accountable.

In technology markets more generally, competitive pressure is more fluid and regular and competition is often fierce. This dynamic can be observed in “back office” technology industries like database systems, Web systems or e–mail servers, where innovation is rapid and seismic. Vendors are able to build from the success of others because the switching costs for clients are often relatively low (as compared to “front–end” systems such as CRM, ERP, office productivity and other enterprise client tools). Training for back–end systems involves small numbers of staff, technology is largely hidden from core consumers and there are standards which usually force some compliance within each solution pathway making the cost of changeover itself much more reasonable than it once was.

For educational systems the opportunities are different because the systems of value generally require more end–user training and integration with other systems (they more resemble ERP and CRM type applications). The emerging trends from these core technology industries are very important and hold clues for how we can change the way in which the markets in the public education sector are meeting the needs of the buyers and clients.



The alternative: Open markets

Open solutions, such as Sakai, Moodle and uPortal, represent a shift in the approach to intellectual property management and therefore to competition for many marketplaces. Utilizing transparency in code, development and support as well as new copyrights, open source solutions have stormed many software markets including databases, Web systems, operating systems, home office solutions and e–mail productivity (e.g. MySQL/PostgreSQL, Apache, Linux, Firefox, OpenOffice at and Thunderbird at The resulting changes continue to be well documented in the industry. As a whole, open solutions consistently involve collaborative, decentralized development efforts. Once these systems reach a critical mass of market penetration, they frequently force traditional service and product firms to change the way business is done. In the midst of these changes, some analysts and skeptical corporations have asserted that open source technology reduces economic opportunities for the existing players and slows economic growth overall. As open source advocate, Bruce Perens, points out, this complaint mistakes the real issue:

[T]hose who base a business upon Open Source are asked: how are you going to be the next Microsoft with Free software? But this isn’t the right question if our goal is to achieve an improvement over the Microsoft model. It reflects the fact that most people have been thinking about software from an extremely vendor–centric viewpoint. [24]

Open markets offer alternative modes of production. Evaluating whether Red Hat will become “the next Microsoft” mistakes the issue completely. The consortium of companies, both big and small, creating business opportunities in the open source operating systems segment are already near to Microsoft in market capitalization (If you count IBM and Google alone, they exceed it). The point is that the mode of production is potentially different in the open model, rather than the vendor–centric or “vertical” model of technology production. As Linux is created in a “virtual corporation” of distributed contributors, the marketplace for building services and value on top of Linux is similarly distributed, with no monolithic figure having emerged as a direct competitor to Microsoft: but the market itself represents a very great threat to Microsoft [25].

In regard to the more general notion that open source has slowed economic opportunities overall, technology spending has consistently outpaced real GDP growth since at least 1997, except during the market correction 2001–2002. Open technologies are not, in any obvious way, taking money from the pockets of the technology industry (though it’s likely that they are redistributing the earnings to different companies than was true in the previous monolithic vendor market) [26].



How do open systems accomplish these market changes? In open economies it becomes possible to create more efficient networks of allied agents who benefit economically from each other’s activities. Openness can make it hard to guarantee that profit will fall within a particular corporation’s boundaries, but evidence continues to suggest that profit continues to be generated overall [27].

In the case of Linux, a popular open source enterprise operating system, giant companies such Oracle, IBM, Sun, SAP, PeopleSoft, Google and Novell have adopted the system in many ways to complement their existing package of products, technologies and services. This has resulted in a great competitive advantage for them (even many of IBM’s current mainframe solutions incorporate Linux, extending this technology far beyond its original design). By adopting such an open system, these companies have been better able to innovate and develop new solutions.

Newer companies such as RedHat have also been extremely successful using almost entirely open tools. Their core products are based largely on open source licensed software. The apparent contradiction in giving away their intellectual property is explained by their business model of selling service, support, maintenance, reliability, and ease of use. RedHat can provide guarantees of quality to clients who need these assurances. Because entry–level, free solutions are largely compatible with RedHat’s enterprise products, the upgrade pathway is well established for companies who find their initial experiments with Linux so successful that they need to guarantee higher levels of enterprise service than they can provide internally.

In fact many open source software vendors such as MySQL, SugarCRM and Jaspar see open source simply as an alternative to marketing and sales expenses [28]. They avoid heavy expenses on high cost, slow sales cycles with low sales conversion ratios. For them avoiding these costs largely makes up for the loss of revenue on any software licenses they might have sold. These companies also find that their open license strategy avoids many political and strategic barriers to entry that plague traditional vertical businesses. For many organizations, technology acquisition is managed through the purchase order system: when a purchase order has not been authorized by management, no technology licenses can be acquired. Open systems companies find that since their product has no license fee, it is often implemented by technological staff inside companies long before management is aware of its existence. After staff demonstrate operational functionality, management is often eager to find support contracts and other guarantees for production use: which these open companies can offer for a reasonable profit, having avoided the traditional expenses of a protracted and unreliable sales cycle.

There is another, more recently formalized aspect of openness, which doesn’t have a consistent name yet. Referred to as open communities, peer production and social networks, these open systems offer a powerful way to incorporate human contributions, in ways not widely explored in “traditional” open source software endeavors. Combined with the opportunities created by open source software methods, these open communities might make it possible to design whole industries, drive innovation into stagnant markets, and create economic opportunities and growth.

The power of these social production systems has been explored academically in some detail [29, 30]. In business, their application has exploded in this decade (Wikipedia, Craigslist, Friendster, Amazon product reviews/lists/relations, eBay, MySpace, Flckr,, Google co–op and many others). These systems all rely on their users to create the core information that makes the site valuable. The greater the contribution of users, the greater the overall value the site has for its users.

As opposed to “wicked problems” in technology ... these systems represent “wicked solutions.”

As opposed to “wicked problems” in technology (which describe issues which become geometrically more difficult to solve as scale rises in a linear manner and with extreme effort), these systems represent “wicked solutions.” By aggregating users together, these sites are able to solve problems which are either insoluble or much more expensive to solve with traditional economic methods. The possibility of creating new open communities such as these has captured the imagination of many large corporations as well as a great number of committed individuals. The question remains as to which problems in the overall human economy can be approached with these methods, and how best to do so.

New forms of community production should be utilized wherever applicable. However, issues of what we will refer to as “criticality” should be considered when designing community or peer production opportunities. Criticality is a key component of these solutions, insofar as they can relate to education and instruction, or other technical fields. “High criticality” applications involve fragile solutions, where small errors in production can manifest as massive failures in outcome. “Low criticality” applications are more robust to production errors (where erroneous production results in failure proportional to the error). Examples of high criticality applications include: a manual for a complex brain surgery procedure, software code and instructions for operating city water purification plants. Low criticality applications include an encyclopedia article on Grauman’s Chinese Theater, pictures of my dog, advice for fixing a home office problem in WindowsXP and a consumer review for a brand of garbage disposal.

To date, open communities appear to be most powerful in low criticality applications, where they create remarkably high quality solutions in extremely short periods of time. While open source software, for example, makes use of open communities for support, it does not typically do so for the coding process, because small errors in software code create major failures (such as failure to compile or much worse). The core business of production for open source software looks largely hierarchical and structured — in many ways it is similar to the way software is produced in a corporation (though open source programmers are often unpaid and therefore choose what to work on, rather than being assigned a project) [31, 32] — because of the high criticality of code writing.

In the next section, we will explore the application of these lessons about open technology markets and open communities for education.



Openness for education

The opportunities for and created by openness — both openness of code and of communities — demonstrated in the business world raises the possibility of bringing these approaches to public education markets. This is a broad claim and to support it, we must narrow our focus on one area of technology. Educational data systems such as those used for tracking student, financial, learning management and human resource data represent, in particular, an attractive area for initial focus for several reasons:

  1. The existing market for educational data systems is a largely closed and proprietary market with innovation being limited to entrepreneurs, rather than community–based “peer production.”
  2. Districts typically have to cobble together disparate, expensive and poorly integrated systems or they have to build the core systems themselves. Districts also hold on to existing technology longer than they might otherwise, because integration issues and high licensing fees often make new solutions impractical or unaffordable.
  3. Educational data systems (EDS) hold data for the heart of a district’s systemic reform data: students, teachers, classes, programs, curricula, assessments, discipline, attendance and grades. Improving these systems and providing new opportunities for vendors will greatly enhance a district’s reform efforts. Currently data silos exist everywhere in public education, and existing marketplaces create little incentive or pressure on vendors to solve certain problems on their own (in fact the markets place pressure on the vendors to keep their data systems closed).
  4. There already exists a deep body of knowledge about systemic reform, which can be effectively translated into the development and deployment of open data systems and associated communities.
  5. Educational data systems represent the nexus for many types of reform in public education. Opening these systems will encourage all data systems and vendors (open or closed) to be developed based on fundamental open standards which will benefit districts widely. Such openness will also assist in “vertical” data collection efforts by individual states, improving the quality of information available for strategic and political decision–making.

There is an interesting opportunity in the education sector because unlike many fields, there is great social value to obtained from improvements (and therefore a strong philanthropic interest in changing the overall outcomes). It may be possible to seed open changes here using private, philanthropic funding, combining this with more traditional governmental and other grants and expenditures to build these solutions into robust and widespread innovations. In business systems, generally open innovation follows direct economic opportunities or consumer demand and participation. In education these systems might follow similar growth paths or ones more aligned to the specific creation of value for consumers rather than economic interests or existing consumer interest. The question here is: can philanthropically supported systems help to create demand and interest for applications and solutions not previously available to public educators? [33]

In all cases, these solutions have potential market–changing impact but they are not market replacements. Unless economic value is created by the deployment of these tools, they will likely not be adopted and supported over time, regardless of the level of philanthropic support. The existing marketplace can quite easily choose to avoid these innovations if it is not clear they are designed to enhance revenue and sales opportunities. Therefore vendors need to be involved in the early stages of development, along with state and local education agencies and other interested parties.

By working with vendors early in the process, as the products are readied for widespread release, there will be an availability of economic agents who (for their own profit–oriented self–interests) can deploy the solutions with a high level of quality. These vendors can compete against each other on the terms of their own service of the open system, much as IBM, Red Hat, Novell and other integration and hardware vendors compete using the Linux system. If a client finds a product adequate but the service and support lacking, they can replace the integrator and find another with more suitable skills, creating direct market competition and “post–sales leverage.” Alternatively a client can hire multiple vendors to deliver various aspects of service and support. In traditional models, where the whole system must be replaced when the vendor is removed, competition is highly reduced.

Even where an existing data system itself can suffice, the service, maintenance and feature development offerings are still inadequate for many reform–oriented purposes. Furthermore, it is often very difficult to buy one of these innovative offerings from a vendor other than the one who sells the core system (a classic example of a vertical market monopoly). By “decoupling” these aspects of systems from each other, much higher levels of competition can be created in each area, driving vendors to provide quality in all areas and allowing new vendors to enter into only certain areas of the vertical market (such as technical support or feature development).

Development of innovative new features is another significant area of weakness in the current marketplace. There is a large potential to stimulate innovation here by providing a common platform for various districts and other entities to contribute their individual innovations. Districts often have funds for specific modifications or innovations for an existing system to better suit their needs. These modifications are currently locked up behind a proprietary system. By providing an open system solution, any district can seek to develop an enhancement and propose to the community at large that it is of general purpose and benefit. The community can use the innovation if there is need.



Open systems: Data integration

The educational data systems market is well suited for a targeted effort to enhance competition and increase value for many educational agencies. This can be accomplished through two primary strategies:

  1. Open communities: Support open communities to share information about systems, their use and effects.
  2. Open systems: Develop open systems which will encourage vendors and districts to buy and sell data systems more efficiently.

In this section, we focus first on open systems and the example of the positive impact that openness can have on data integration.

While creating complete open source applications is still a developing field, we have more experience with creating back–end systems for communication in general (e.g. Apache, Sendmail, BIND). In the education sector, application interoperability among business systems is a clear existing weakness. It prevents enterprise system integration, while also inhibiting new innovations, by making it particularly difficult to connect new, small and innovative systems with existing, core enterprise applications. While open standards exist to support data integration, market uptake has been tepid at best. There are no effective open implementations of these standards that permit data exchange between applications (whether open or closed) [34]. Creating such an open implementation can create a broader support of the standards themselves (For example, Apache greatly enhanced the standardization of HTTP).

An open source data integration toolkit to permit extensive application integration would introduce highly desirable market changes. First vendors need to have a market incentive to support the tool. Data integration standards are already supported in concept — requirements to support them appear often in RFPs [35] for new systems. These standards are not actually widely deployed in many implementations, however. A likely reason for this is that both vendors and educational agencies may be reluctant to deploy commercial data interchange applications. If a vendor installs closed integration tools across enterprise data systems, the client doesn’t own the infrastructure. A new vulnerability exists where the effects of vendor instability or poor function are greatly increased. Even more than in application systems themselves, the failure of application integration is often far worse than having no integration at all. Similarly, “renting” data integration solutions from a vendor (via license and maintenance fees), makes it very difficult to replace the vendor. An open tool does not have these vulnerabilities. A vendor might still charge maintenance fees for the solution, but there can exist other vendors who also offer maintenance services against the same system. Unlike closed systems, open systems and the vendors who deploy them are firmly coupled.

With a functional, deployed integration toolkit, enterprise application vendors could also compete with each other on a finer grain of application function than they do now. Currently vendors can only sell complete applications, such as student information systems. Vendors cannot sell a piece of their application to supplement or replace another vendor’s application, which is already deployed for an education agency. A widely accept data integration tool would change this, allowing vendors to sell pieces of their systems, creating a more fluid market for competition.

An example of this would be where a complete student information system (SIS manage grades, attendance, discipline, course rosters, class schedules, teaching assignments and student demographics) exists in a district. While school districts cannot afford the cost of replacing their entire system, they might realize that the attendance component of the system is inadequate. With open data integration tools, it would be possible for another vendor to offer to replace only the non–functioning attendance component. In the current marketplace, vendors generally have to offer a complete systems replacement, and cannot sell services and systems on a finer grain of functionality. Providing such fine–grained competition would certainly increase competition and economic opportunity for many vendors.



Open communities

In addition to the functional problems with data integration, the educational data systems market also suffers from extremely poor information. Many educational agencies find data system vendors through staff inquiries into personal networks or through vendor trade shows. They often cannot find qualitative information about vendor performance, other than through vendor provided references. Peer support communities, where they exist at all, are almost always sponsored directly by vendors (and are often restricted to existing clients only). Since vendors may charge for support on a usage basis, open peer support is not always even in their economic interest.

Legal contracts for these applications also display poor information qualities. Contracts are often signed based on vendor–provided documents and are often not negotiated in any business sense. These contracts frequently omit conditions regarding specification for features, quality of service or service level agreements, cancellation clauses, or performance–based payment schedules.

It may be possible for open communities to help with these issues. While open source methods can provide a market catalyst for major changes in the way software is built and sold, open communities can help create better information around that technology and support how it is used in educational communities. Specifically, the initial low–criticality, high–value applications where open communities have displayed “wicked solutions” in other areas, relate to access to information and support. Specifically, building open information communities around the following content areas could provide major benefits:

1. Vendor product and service listings
a. “Amazon” or “CNet” for education. Provide peer product reviews and community–generated vendor reputations.
b. Open customer support forums based on product lines as well as on general solution areas (such as data warehouses, SIS, financial systems, etc.)

2. Contract Commons (at
a. Sample contracts and clauses for various applications (ASP, vertical, open purchases). (Because this is a “high criticality” application, open source software development methods are more appropriate than peer production techniques).
b. RFP network: RFP distribution to subscribing vendors.
c. RFP builder: online guides and community support for developing effective RFPs. Also RFP libraries of existing documents.
d. Support network and market exchange between experts and those needing legal assistance.



Open philanthropy

In order to build enterprise level educational data systems for the public school sector (open or closed), substantial investment and effort are required. One of the primary reasons to invest in building an improved closed educational data system would be to recoup the initial investment, and garner ever–increasing profits as success of a superior product drives more sales. This naturally implies that the initial investors want to recoup the initial investment (a very common implication in capital economies). Quite notably, private philanthropy can distinguish itself from traditional capital investment approaches in this regard. While many philanthropic investments are currently modeled on private industry’s investment strategies (often simply because they are familiar), the rise of open source solutions offers a powerful opportunity to change the face of philanthropic investment for creating social returns and benefits.

A philanthropic institution’s return on investment is primarily focused on social outcomes rather than profit and capital returns (this is required by law in many cases). An open systems approach does not always promise large profits to for–profit investors. It does often promise the delivery of very rapid and extensive market penetration, while driving new levels of competition, service and usability into the marketplace. These changes can draw new private sector innovation as self–interested businesses seek to profit from new market opportunities. It promises to create tremendous social returns for relatively modest initial investments (Wikipedia’s social value far exceeds the monetary investments required to create it). The open solutions strategy creates collaboration among clients and supporters, while bringing more intense, innovative and competitive market forces to vendors and suppliers [36].

Open systems development itself does not always guarantee great opportunities for profit. Economic opportunities must be created for the vendors who support and develop an open system for clients. These vendors will certainly compete on a much more even playing field because their services are “decoupled” from any underlying, open systems. A district can change a vendor without changing the underlying system. Because districts can more easily substitute one vendor for another, vendors experience much greater incentive to deliver timely, measurable and substantial results that fit the client’s actual need.

Innovative, reform–oriented philanthropic foundations are well suited to facilitate the development of these educational data systems because of the extensive, positive and strong partnership that they have with districts around the United States. The information that these districts can provide about how a new system will be designed is an asset almost impossible to obtain by any other agency or company.

Furthermore, these foundations have significant ties into the philanthropic and private investment communities, which will make complete funding for the project less difficult. Such a foundation’s educational partners also stand to greatly benefit from the development of the initial systems (as well as future market changing systems throughout the sector). This dovetails with the efforts to extend school reform models into the larger community public districts. All key aspects of its research–based reform model can be tightly integrated into any system whose creation is fostered by educational community: this is knowledge management for long–term, systemic results.

While there are many successful and productive educational data systems vendors, the educational data systems market itself is broken, and both providers and consumers are locked in cycles of poor decision–making and action that must be changed. Providing a new system will only be beneficial if new terms of competition among vendors change their abilities to help districts to succeed. As in many areas where foundations operate the districts do not have all the capacity that they need in order to succeed alone: the goal is to help vendors continue to build their businesses working with districts.



Many markets

By innovating first in the educational data systems market, a developmental strategy of opening new markets through open source systems can be undertaken by a consortium of philanthropic and governmental investors. The educational data systems market is the most approachable, but from that base, additional educational systems can be incorporated. These include: curriculum materials, assessment instruments, professional development services and leadership development services. Some, and often the most ambitious, will be created directly by consortium–based investment of organizations, which can deliver the required enterprise software development. Many new systems, however, will be built directly by the districts themselves. This is a critical premise of the entire market strategy. Development may in cases be in piecemeal fashion and alternatively, especially as a result of new grant procurements, in coordinated detail and significant scope. Because the core data support systems and communities will already exist, all further investments will leverage this the core development environment, software frameworks, open data models and existing vendors communities.

The creation of open data models via an open integration toolkit is therefore very important because the absence of freely flowing information between educational systems today is an extremely pernicious problem that seriously hampers all types of systemic educational reform efforts. Existing private vendors within the educational markets have virtually no incentive (and often strong disincentives) to solve this issue on their own — the first vendor to open their system becomes highly vulnerable to having clients switch to other systems, with no corresponding possibility for the clients of other vendors to move in the opposite direction [37].

If a new open markets are undertaken in other educational service categories, the open data systems and communities will have already been developed to support the free exchange between the respective systems. By providing such an enterprise framework for new development work, the new markets will be based within the existing models, and linking their data elements becomes greatly simplified.

This property of common data models across open markets is extremely important for another reason as well. As the open educational systems gain market share, financial and other system vendors (building open or closed solutions) will have very strong incentives to ensure that their systems coordinate with existing open standards. If educational agencies come to expect and require interoperability, and have the tools to implement it, vendors will have incentive to sell into this market, as well as pressure to ensure their product lines support such interoperability [38].

This is a very important point that will serve to greatly benefit public education and reform: given sufficient market share for the open systems and sufficient (self–interested) engagement by various other vendors to create data pipelines to the open system, the open system becomes a de facto data exchange platform for all types of data, even between closed, proprietary vendor systems in different solution domains (e.g. curricula, assessments, finance, HR and student data). Opening the market to a sufficient degree thus serves to open data exchange among vendors of all kinds, across many markets. This increase in value provides a correspondingly strong feedback loop towards additional new (and profit–oriented) participation. Vendors and districts will make self–interested business decisions, but because of the new market system, they will be induced to create better products and higher value.

Finally, to return to the questions of equity in education and the economically and racially disparate achievement of students in the U.S., this project [39] will make it much easier for educational institutions to change the way they currently operate, to better serve students who are currently not being effectively educated. Educational agencies currently operate without sufficient information, and therefore often make inaccurate or inappropriate decisions. This project will make the markets more efficient, and provide ways to bring educators in contact with those who have the experience or expertise to help solve problems or provide information. By developing integration tools, educational agencies will be able to deliver services to their teachers and students more effectively, as well as provide them with better information. All these changes will assist in helping public education deliver on the promise of providing high–quality education for all students. End of article


About the author

Steve Midgley is the Program Manager at the Stupski Foundation in Mill Valley, California. He has been researching issues of openness, technology and education for the last 12 months, and has worked in his current position at the Foundation since 2001. Prior to that he co–founded, a successful commercial real estate listings firm, which provides services to buyers and sellers, using some principles of open interchange, for profitable ends.
E–mail: steve [at] stupski [dot] org



1. TIMMS study at and

2. Of course, these gaps in student achievement are not inherent to the students themselves, or even to their economic or social situation. There are examples of achieving success and parity that prove that strong academic achievement for all students is possible. This paper will cite several such examples.

3. “School districts” is a generic phrase in the U.S. referring to an organizational unit chartered by a State government to manage a group of schools for a community. The districts almost always have local community leadership, either elected or appointed by local officials, such as a mayor (or some combination). As having substantial organizational autonomy, school districts have been targeted by foundations and state governments as a promising avenue for systems–based reform.

4. “Data for Closing Gaps in Public Education,” Ross Wiener EdTrust, March 2006, at

5. Douglas B. Reeves, 2004. Accountability for learning: How teachers and school leaders can take charge. Alexandria, Va.: Association for Supervision and Curriculum Development.

6. Carolyn Kleiner Butler, 2005. “Success in the city: A once troubled urban school system is lauded for blazing a new path to academic progress,” (3 October), at

7. “Sacramento’s Legacy,” Broad Foundation, at

8. “Aldine Independent School District,” National School Boards Association (25 September 2002), at

9. “Leading for Learning,” Supplement to the 14 September 2005 issue of Education Week, at

10. “Theory of Action,” Education Week (14 September 2005), at

11. “Forward Motion,” Education Week (14 September 2005), at

12. Heather Bremner, 2006. “Accountability Task Force Releases Report,” Gilroy Dispatch (16 February), at

13. TURN Exchange, at

14. The Education Trust, at

15. The New Teacher Center at the University of California, Santa Cruz, at

16. Center for Performance Assessment, at

17. Yochai Benker, 2003. “The Political Economy of Commons,” UPGRADE, volume 4, number 3 (June), pp. 6–9, and at

18. Steve Weber, 2004. The success of open source. Cambridge, Mass.: Harvard University Press.

19. Administrative systems refer to systems which administer student, human resource or financial related data for business operations, reporting or analysis, as opposed to learning management and content systems which hold data related to the instructional process.

20. Steve Midgley, Stupski Foundation: internal research and investment 2001–2005 with 11 districts across the U.S.


22. Steve Midgley: private conversations with several medium and large districts nationwide, 2001–2006.

23. Stupski Foundation 2001–2006, internal research: Product adoption for partner districts is usually around new technologies such as plain paper assessment management. Existing systems such as SIS, Finance and HR are rarely examined for replacement or update.

24. Bruce Parens, 2005. “The Emerging Economic Paradigm of Open Source,” at

25. See the “Halloween Memos” published at one time by Eric Raymond and now still available on

26. Phillip Neuhart, 2006. “Tech: Monitoring Its Leading Economic Indicators,” Wachovia (6 March), at

27. Clayton M. Christensen, 2000. The innovator’s dilemma: The revolutionary national bestseller that changed the way we do business. New York: HarperBusiness.

28. MySQL, Jaspar and SugarCRM representatives, UC Berkeley Entrepreneurs Forum, Haas School of Business, 29 September 2005.

29. Yochai Benkler, multiple articles dealing with this point, at

30. Beth Noveck, 2005. “A democracy of groups,“ First Monday, volume 10, number 11 (November) at

31. Weber (2004).

32. Eric Raymond, “The Cathedral and The Bazaar,” at

33. Initial investments by Ira Fuchs at the Mellon Foundation in Sakai, Kuali and others suggests a positive answer.

34. The main data standards in the education sector are SIF (at and IMS (at

35. RFPs or Request for Proposals — documents created to describe a need to which vendors submit bid proposals.

36. “Archimedes’ Lever, Interview with Ira Fuchs,” at

37. The concept of a single, monolithic datum is expressed in the SIF specification. This is unrealistic as well. As Apache/HTTP has demonstrated profoundly, data interchange tools must specify completely the methods of communication but must remain very open as to the content of communication. The content of communication can still be driven by standards but by allowing the standards of content (HTML) to evolve over time while not remaining tied to the communication standards (HTTP), systems can self–assemble in a highly efficient manner among other nodes in the network with similar uses and business pressures.

38. Currently, vendors of disparate product lines have tremendous difficulty in coordinating their exchange of data because each vendor has a strong incentive to keep their data models private. Any free movement of data offered by only one vendor, is highly anti–competitive for that vendor. The open system model has no such disincentive, and therefore almost all vendors have an interest in providing exchange services with the open systems.

39. The project is currently partnership with several foundations, school districts, county offices of education and a large state department of education. For more information about this project or to inquire about participation, please contact the author at steve [at] stupski [dot] org.

Editorial history

Paper received 15 May 2006; accepted 19 May 2006.

Contents Index

Copyright ©2006, First Monday.

Copyright ©2006, Steve Midgley.

A case for open markets in education by Steve Midgley
First Monday, volume 11, number 6 (June 2006),