Linux in Government: The Government Open Code Collaborative
As we celebrate the holiday season and prepare for the next round of legislation, a group of state and local governments has banded together to collect and distribute freely the costly software that normally runs taxpayers $100 billion annually. Called the Government Open Code Collaborative or GOCC.gov, this organization states that its members work together voluntarily to encourage "the sharing, at no cost, of computer code developed for and by government entities where the redistribution of this code is allowed".
In addition to state and local governments, the organization also encourages collaboration between public sector entities and non-profit academic institutions. With Web facilities hosted by the University of Rhode Island, GOCC.gov has a repository dedicated to hosting open-source software for download by any state or local government.
As so many people have said, "Nothing is as powerful as an idea whose time has come." And GOCC.gov is an idea whose time is long overdue. Open-source advocates attempting to initiate legislation and fight the battles on the floors of the various Houses, only to discover the political might and opposition of Microsoft, now have an alternative. State agencies now can download software for free and use it to create a cohesive and standard government infrastructure.
Instead of every county in the country buying the same Commercial Off-the-Shelf Software (COTS) 3,750 times, they simply can find what they need, download it and install it--a design similar to the distribution of Linux. Think of the cost savings and standardization this offers Homeland Security, law enforcement, the judicial system, deed databases, eGovernment applications and financial applications, to mention only a few areas. Additionally, the concerns of connecting various disparate databases across the country, a topic we heard about daily during the last campaign season, can be put to rest.
An example of the kind of software you can find on the GOCC.gov site is Election Tally, contributed by the city of Newport News, Virginia. Election Tally is a parameter-driven Web-enabled application written in Python and utilizing ModPython and MySQL. It generates an election tally report by extracting files for the state Board of Elections and produces a video simulcast.
That's pretty heady technology available to everyone in the country. In my voting precinct, our team had to generate the results and post them on the door of the polling place. If we could interest the Dallas County Election Board in adopting Election Tally, it would allow us to interface our polling machines directly with headquarters. The commission immediately could begin its audit, save time and eliminate voter fraud.
Of the many types and kinds of participants, eight states now participate to some extent in GOCC.gov: Rhode Island, Massachusetts, Pennsylvania, Utah, West Virginia, Virginia, Texas and New York. In the majority of cases, individual agencies have joined. Of course, GOCC.gov hopes to attract every state in the country. The more governments that participate, the faster the adoption rate can grow, along with the cost savings.
According to the GOCC.gov site:
The organizing meeting of the GOCC was sponsored in December 2003, by the Commonwealth of Massachusetts, in conjunction with Harvard University and MIT. The morning session at Harvard University's Kennedy School of Government was conducted by Professor L. Jean Camp, who presented an excellent tutorial on the various license options available to code sharers. Sparing every expense, the attendees took advantage of the Commonwealth's excellent public transportation system and used the MBTA's Red Line to make the transition to MIT for the afternoon program.
The afternoon session at MIT included an audio bridge for those folks that could not attend in person. The genesis of the collaborative was vetted and launched through a discussion facilitated by Dan Greenwood of MIT's E-Commerce Architecture Program. Dan has been a significant contributor to this initiative from inception. Through a series of subsequent audio conferences, the group agreed to the operating rules for the collaborative and the repository, the governance and officer structure and the actual announcement process. Highlights include the following:
The GOCC will be entirely independent and not affiliated with any professional or private sector entity.
The GOCC will accept no financial or in-kind assistance from any private sector company. All initial members will be either municipalities, legal entities of state government, or academic non-profit institutions.
Four officer positions were established to serve for one year:
Chairperson: Peter Quinn, CIO, Commonwealth of Massachusetts
Municipality Representative: Mike Wells, CIO, Gloucester, Massachusetts
Technical Lead: Jim Willis, CIO, Secretary of State, RI
Policy Lead: Patrick McCormick, Harvard University, Kennedy School of Government
On June 30, 2004, GOCC.gov made an official announcement that it was in business. Again, according to the Web site, "The Commonwealth of Massachusetts Information Technology Division; the Rhode Island Office of the Secretary of State; the Pennsylvania Office of Information Technology; the Utah Governor's Office, CIO Section; the Kansas Secretary of State Office; the Kansas Treasurer's Office; the Missouri Secretary of State Office; the West Virginia Auditor's Office; the City of Gloucester, MA; the City of Worcester, MA; and the City of Newport News, VA, [announced] the formation of the Government Open Code Collaborative (GOCC)."
The GOCC wants to offer only code licensed under an "OSI Approved License" or any other open-source license deemed acceptable under the GOCC's operating rules. And according to the GOCC, this code routinely is referred to as either "Open Source Software", "Free Software" and, less frequently, by other similar names.
GOCC also states that:
Government entities, defined as a federal, state or local government, an authority or other sub-national public sector entity of the United States, can join the GOCC as Members by signing the GOCC Operating Agreement through an authorized representative. The signatory or their designee can then appoint additional members within their entities.
Private non-profit U.S. academic institutions can also become members by signing the GOCC Operating Agreement through an authorized representative. The signatory or their designee can then appoint additional members within their institutions.
People belonging to a government entity or private non-profit academic institution that has not signed the Operating Agreement can participate with an Observer status. Representatives of non-profit associations of public entities can also participate as Observers. Observers have to be sponsored by a Member. Observers are able to join the GOCC list server to receive announcements and participate in discussions and are encouraged to participate in the GOCC bi-weekly conference calls.
Have you ever heard the cliche about prisoners running the asylum? Well, this gated and restrictive organization fits. Get a group of academics together with management and grow the group in the infertile soil of bureaucracy, and you will spend almost all of your time waiting. Watching the group from a distance over the past year reminds one of inexperienced farmers trying to plant a field of corn by reading books.
GOCC.gov is a Cathedral trying to say it's a Bazaar. You might as well call the Java programming language open-source software. GOCC.gov goes through the motion of calling itself an open-source collaboration, yet it excludes the people that can bring the vision to fruition.
GOCC.gov ignores the existing base of software by excluding vendors from donating their solutions. It excludes contributors from the Linux and Open Source community not affiliated with a government or academic entity. So where will it find the people with the skills to develop the repository? Within its own infrastructure? Perhaps you now see why I used the cliche about asylums.
Can anyone see a business model here? Read the GOCC.gov charter and you discover that it has built one more bureaucracy to oversee its existing bureaucracy, with oversight over the new bureaucracy. What else would one expect?
In an article titled "IBM: 'Inertia' holding back government desktop Linux adoption", the executives of Big Blue identify inertia as the primary reason governments haven't adopted open-source solutions in England. IBM's public sector business development executive Jeremy Wray said, "[the] single biggest factor holding back government departments from migrating to the Linux desktop is inertia. At the moment public sector departments lack a compelling reason to act." Unwittingly, he has described a property of all bureaucratic organizations, one that IBM itself has helped foster. If bureaucracies don't have a problem to manage, they have no reason to exist.
The idea of inertia comes from Newton's first law. One definition identifies inertia as "the property of an object describing its tendency to stay at the same velocity (or at rest) unless a force acts on it". So inertia is a property, not a cause. Inertia is a property of bureaucracies, and it doesn't change unless acted on.
If you want to see how vigorous GOCC.gov has been over the past year, look at its list of software. That's right, you're looking at five pieces of software, one of which is an application. That's what GOCC.gov has accomplished in one year. If you think that's strange, consider also that it took this "Collaborative" six months to announce it existed.
Now, drop down to the membership list and look at the contributions from Texas. Click on that link, and you get "There are currently no items in this folder".
The CIO of my great state has taken some pride in letting people know that the "Texas Open Source Bill" hasn't passed and won't pass. As she has said in public, "it's dead". Yet, within Texas, the Department of Information Resources touts its open-source sharing plan, as seen here.
One of the touted programs in Texas is the Governor's Office database of solutions for free source code. When you visit the Web site, you find the same solutions that have existed for two years. You also can find the same testimonial that has existed for the same time period. This is an attempt to say they have all these pieces together so the legislature won't force the issue as they tried in 2003. Unfortunately, the Senator sponsoring SB 1579 understands the issues and plans to act in 2005.
In its present condition, GOCC.gov cannot work. In a closed community, a member must receive some benefit to join. If I join and contribute software, what do I receive in return? If the operation is gated and closed and I must provide software support, I at least need to be able to swap for something in return. If I can download any software for free without becoming a member, why would I want to expose my organization to legal liability? When you look at the GOCC Operating Agreement and at the organization, someone outside of the Collaborative--"the Member contributing code"--assumes liability for the code working. Why would I want to do that?
Read the GOCC Operating Agreement to get an idea of the restrictions placed on those who can contribute code and their responsibilities. The restrictions and the lack of incentive provide no cost-benefit ratio. Governments have accountability to their constituents, and GOCC.gov needs to ask two simple questions: Why do constituents have to be the first ones to pay for the application, spend the money supporting it and risk liability? And what do they get in return?
The Collaborative should accept contributions of open-source software from vendors who have implemented existing solutions. In other words, open up the repository to people who have products that governments need and cannot afford. Secondly, let the vendors build a business plan around the GOCC repository.
For example, if I have a project such as the LibraryofTexas.org, let the vendor who already has open-sourced the project contribute the software. If a small town in Alabama wants to deploy it and has the manpower, they would do it themselves.
However, if a city such as Cleveland wants to deploy the software and would rather hire the originator of the product to install it and train people, then the originator could charge for their services. The contributor gets his software exposed on what could become the largest government software repository in the world. That exposure would provide vendors with an incentive to contribute, because they have a chance to increase their install base.
The alternative to having vendors contribute and provide services is to wait. GOCC.gov can wait for everybody in government to learn Linux and then do everything themselves. By the time that happens, though, we won't need computers any more. We'll have evolved into cyber-beings and telepaths.
Tom Adelstein lives in Dallas, Texas, with his wife, Yvonne, and works as a Linux and open-source software consultant locally and nationally. He's the co-author of the book Exploring the JDS Linux Desktop, published by O'Reilly and Associates. Tom has written numerous articles on Linux technical and marketing issues as a guest editor for a variety of publications. His latest venture has him working as the webmaster of JDSHelp.org.
