The Limits of Open Source - Governments Should Adopt Open Source Now

An ideal candidate for open source software is government. Government, at least democratic governments, have no profit motive and should have no biases for artificially supporting one company at the expense of others. Governments have a moral obligation to their citizens to deliver the services as efficiently as possible, i.e., not to waste tax dollars. Except for some specific areas that obviously require high levels of confidentiality, e.g. law enforcement, tax collection, national defense, and security, most government functions should be conducted more or less publicly. There is no reason at all that the basic infrastructure tools, computer operating, and networking systems, used by government should be anything but the most cost effective, secure and reliable systems available.

When comparing existing open source to existing proprietary products, proprietary products tend to come out ahead in two areas: one is extensive feature lists and the other is normally referred to as "ease of use". Below I'll suggest how the feature lists are not generally appropriate to government and that the "ease of use" or not requiring training is largely a myth.

As for other factors that should be used by government in selecting software products, direct costs, reliability, and security, I'm working on a long review that deals extensively with these topics. Until that is online, I'll simply say the direct costs issue should be self evident. Open source products have simply do not have the licensing cost and licensing management costs that run to several hundred dollars per desktop and typically several thousand dollars or more per server.

It doesn't take much experience with a mixture of both open source and proprietary products, specifically Windows systems that dominate desktops and are the largest single server OS, to realize there is no comparison in the reliability area; open source products are at least one and possibly as much as three orders of magnitude (1000 times) more stable than competing Windows solutions. Once set up, Linux and *BSD family products simply don't / almost never crash. Other than in exceptional controlled environments, all Windows and Macintosh family products crash; the only question is how frequently. Tending to crashed servers and desktops is likely to be the single largest computer support cost most organizations face.

Any UNIX based system can be made secure. Windows systems cannot be made secure without removing the features that make them Windows systems. For years, Windows web serves representing a minority of the market have accounted for the majority of the break ins. 2001 has seen several Microsoft products, not limited to IIS, subject to fundamental security breaches actually affecting hundreds of thousands of computers. The UNIX community, including open source operating systems, hasn't seen anything comparable since the Internet Worm of 1989 and that was nothing more than a denial of service attack. Unlike recent Microsoft compromises that are resulting in remote administrative access and dissemination of confidential information, the Internet Worm did no actual harm except slow some systems down till they were not useable. Microsoft products experience slowdowns from worms and viruses multiple times per year, often with additional unwanted side effects.

As for training, Microsoft hides system details even from administrators and about every five years makes changes that undermine much of the what has already been learned about Microsoft products causing skilled and expensive staff to start learning the basics over again. UNIX and related open source products build on a heritage that is nearly thirty years old. New things come along but no skilled UNIX administrator ever has to discard a large part of their knowledge and start over because a company that dominates the market decided it's time for some new era in computing.

Governments need to manage their own operations. For the most part, governments have built or contracted the development of custom systems necessary to perform these functions and own the resulting systems.

The leading edge of commercial software development tends to be high end application development tools and services, that allow medium to large companies, to quickly deploy and adapt to changing needs, a variety of online projects. Recently this has been mostly e-business, and is increasingly becoming a variety of business to business processes, that are not necessarily conventional buying and selling. A hot area is the development of XML to allow businesses to exchange data without the need to custom program data exchange interfaces.

Compared to business to customer and business to business government relationships with individuals and companies are relatively simple. Today even retailing, a one to many relationship, requires business to customize the processes to each individual. All significant online retail business today incorporate considerable amounts of obvious customization from changing web pages based on purchase history, tailored e-mails also based on purchase history but also based on requests about specific products and product areas, to selectable checkout options that trade convenience for control. Physical stores still customize the old fashioned way through human interaction. The online world is adding human interaction as well via electronic "chat", voice/telephone and even sometimes video and other ways that customers can interact with staff. Customer relationships, are voluntary and the customer can end them at any time.

Business to business introduces many to many relationships that become even more complex. For example auto manufacturers deal with a variety of auto parts suppliers each of which typically deal with multiple auto manufacturers. Auto manufacturers also deal with raw materials suppliers such as steel, glass and paint each of which is likely to deal not only with other auto manufacturers but also other industries as well. Electronic data interchange was limited to a relatively small number of large businesses. In the new online world a growing number of companies will exchange growing amounts of information with increasingly diverse partners. Business exchanges will only occur when both sides see an advantage.

Governments operate in a fundamentally different environment than this fluid business world where there is always the potential for someone else to offer a better deal and thus force a change in relationships. Though there are multiple levels of government (national, regional, local) in its own area, each is a pure monopoly. Generally where you live, work or do business determines which governments you do business with. You don't get to choose governments except by relocating. In virtually all cases of interaction, the government sets the terms of the interactions. Some interactions are mandatory (taxes, licenses) and others are voluntary.

Businesses, especially larger ones, tend to need leading edge software that helps the adapt to changing customer and partner requirements. Those that don't go out of business. Governments simply don't have these needs. If a government makes a fundamental mistake, it doesn't go out of business; the citizens pay for the mistake through higher taxes. In democracies there is indirect accountability via elections but this a slow process that does not have quick or direct consequences for the government the way that a customer switching suppliers does.

The government needs to make massive amounts of information available, most of which is official. Generally each government agency is the authoritative source for information which it must distribute. After an agency has collected or created the information, regardless of its sources, when it is ready to disseminate this information there is no need to correlate their information with other agency's information. After the government has performed what it deems appropriate analysis, any needs to correlate information or enhance the value of government information, by aggregating information from different agencies should be the job of the businesses that expect to benefit from the enhanced data. Government information is purely take it or leave it; if you don't like the way the government presents its information you're free to find a commercial information provider that does a better job of packaging what you need.

For information dissemination, open source products already provide all the tools necessary for most governments' information dissemination needs. Nearly all government information that needs to be distributed can be distributed at almost no cost via web pages or HTTP or FTP and a variety of standard file formats. For textual information, these formats might include text, HTML, and XML and possibly PostScript given the widespread availability of PostScript converters. Given the widespread availability of free viewers for .PDF this could be a valid format but if Adobe were to try to gain financially by this (start charging for viewers), then the government would have an obligation to not use the .PDF format.

For years the FEC has distributed huge amounts of relational data in simple text delimited files via FTP. This is the appropriate approach. It's up to anyone who wants to use this, to read the FEC documentation on format and keys, and develop the appropriate procedures to load the data into a useable database. It would be improper to distribute this data as pre loaded Oracle or SQL Server databases. Complete databases would result in enormous, rapidly changing databases and force users of other products to buy the supported products. Partial databases would limit uses. Providing the complete raw data allows any user to pick and choose the data relevant to their purposes.

The government also does not face one of the major issues that tie private companies to Microsoft desktop products. This is the document exchange issue. As long as the government is standardized internally it won't have internal exchange issues. Governments simply need to decree the acceptable document formats for any electronic documents submitted to them and anyone who wants to do business with them has to comply. Typically, where governments accept electronic information, they already specify the format. The government won't have desktop exchange problems as long as proprietary document formats such as .doc, wpd., .xls, .ppt are not among the acceptable formats.

Like much of the consumer world, government office workers often believe they need the ever increasing feature sets of desktop office products. Unlike commercial environments, that may need slick marketing presentations, most of the governments' desktop information exchange needs, both internally and externally can be met by any professional quality word processor and e-mail system. Adequate quality open source tools have been available for some time.

The only impediment to wholesale government adoption of open source desktop products are relatively minor worker training issues. I say relatively minor, not because there are not significant differences between the de facto standard, MS Office and any possible replacement, but that staying with Microsoft products absolutely guarantees ongoing and significant training costs.

Microsoft claims that their products are a standard and reduce training costs because they are so widely used. They never mention that every new release introduces changes which typically require training, and usually contain incompatibilities with pervious products causing common document conversion and corrupt document issues. About every five years, Microsoft introduces a major user interface change, while Macintosh and UNIX have delivered evolutionary changes. Microsoft has indicated another major interface change after XP. In the Sept. 4, 2001, PC Magazine, in an interview, Bill Gates said "I honestly believe that within the next three or four years we will propose a radical new user interface change . . ." So much for Microsoft consistency. In the long run the government will save massive training costs by changing to open source desktops.

Some government agencies are already moving towards open source products. Governments at all levels from the federal government to city and county governments should begin systematically moving both servers and desktops to open source solutions. Where clear needs are not already met by existing products, the government should hire (or divert) the necessary programmers to add the needed functionality to the open source products. This will likely cost a small fraction of proprietary licenses on the scale used by governments. It will also benefit the public and most businesses by enhancing products available for all to use. The government should not generally get into the software business. Only where it really needs a capability, not presently available in open source, it makes better sense to put it there, than to pay ongoing license fees to a vendor with no motive except profit.

Microsoft fears a systematic government move to open source. First, governments as a group represent a significant revenue stream so losing them to open source would have an important negative impact on revenues. Further if the largest organization in the world, the U.S. government, ran on open source, every business would at least ask if a similar move represents a viable option.

Top of Page - Site Map

Copyright © 2000 - 2014 by George Shaffer. This material may be distributed only subject to the terms and conditions set forth in https://geodsoft.com/terms.htm (or https://geodsoft.com/cgi-bin/terms.pl). These terms are subject to change. Distribution is subject to the current terms, or at the choice of the distributor, those in an earlier, digitally signed electronic copy of https://geodsoft.com/terms.htm (or cgi-bin/terms.pl) from the time of the distribution. Distribution of substantively modified versions of GeodSoft content is prohibited without the explicit written permission of George Shaffer. Distribution of the work or derivatives of the work, in whole or in part, for commercial purposes is prohibited unless prior written permission is obtained from George Shaffer. Distribution in accordance with these terms, for unrestricted and uncompensated public access, non profit, or internal company use is allowed.