Free World Notes
This file contains supplemental notes to the manifesto “Only the Free World Can Stand Up to Microsoft”, currently published at http://www.gnu.org/philosophy/free-world.html.
You may write the author, Tom Hull <email@example.com>. Hull is also the author of the Ftwalk programming language, a script programming language which is free software available for Unix systems.
In general, this critique reflects a more general line of thought, which is based on the recognition that the inefficiencies and ulterior motives in our current modes of production require much unnecessary work to produce products and services of often dubious merit for grossly inflated prices, effects which diminish the quality of our lives and the worth of our work. Nonetheless, my proposal here is not especially radical: it does not challenge the precepts of intellectual property; it requires no political action (not even the application of antitrust law); it can be initiated by a small group of people, and to some extent simply builds on work already done by various individuals and groups.
Some paragraph notes:
- Commercial software companies typically divide their costs into several sectors: development; manufacturing; marketing/sales; service; general and administrative. Development costs are usually less than 20% of revenues. By far the largest cost is marketing/sales, so most of what the customer is actually paying for is the persuasion to convince the customer to pay so much for something that costs so little to develop, and practically nothing to reproduce and deliver.
- More expensive software often includes after-the-sale service, which should be considered a marketing/sales cost, since it props up an extravagant price structure. Service should be considered a separate cost, independent of development. Free software is always delivered with no service, and customers who need service can obtain help independently, since the inner workings of the software are public knowledge.
- Media companies have comparable cash flows, but necessarily work within the technical standards of their media. Consuming their products does not in any way prevent or even disincline one from consuming competitive products.
- Microsoft likes to expand its operating system to eliminate the market for add-on software, such as for disk compression and networking. Microsoft's claim that IE is part of the operating system is spectacularly spurious.
- Microsoft's dominance is at least partly due to the lack of any significant challengers. Apple and IBM used their operating systems to lock customers into their hardware, and would at any rate have been rejected by the rest of the PC industry, which at least with Microsoft got access to the same product. Unix vendors have stuck steadfastly to higher priced markets, avoiding direct competition, even though NT is aimed directly at destroying Unix. The longer Microsoft goes without serious competition, the harder it gets to mount any such competition.
- The last sentence is a slight exaggeration. Many capitalists do in fact realize that they will never be in the position to wield the sort of power that Microsoft commands, and as such have no use for the megalomania that goes with such power.
- The main point, however, is that under current circumstances no sane investor will directly challenge Microsoft. The cases in other industries where challenges are made to dominant companies depend on the discovery of some significant cost advantage (e.g., MCI's challenge to AT&T), but cost advantages are essentially impossible in software, unless you're willing to forego all your margin, a position no investor will take.
- Antitrust laws work more for the protection of other businesses than to protect consumer interests, although consumers generally do benefit from increased, more even handed competition, at least in the long run. In the short run consumers may benefit more from crippling price competition. Netscape, for example, having gained a dominant market share in its niche, still cannot raise its prices because of Microsoft's competition, which is a windfall of sorts for customers.
- We talk much about the advantages of “letting the market decide,” but most business activity is oriented toward rigging the market. Look at any business plan and the key section will be something like “Barriers to Competition,” because competition kills profits, and successful companies are the ones that avoid competition, or at least are able to dictate its terms.
The key thing here is that the free software must have at least the same level of quality and utility as the commercial software that it challenges, which means that it must be professionally designed and developed, tested and supported. Which means that free software must move well beyond its current niche as an academic hobby, to a point where it is supported by well-financed organizations that can attract and support quality workers.
Of course, Microsoft (and all other commercial software companies so threatened) will do their best to compete with free software, and can be expected to do so as desperately as they compete with everything else. There will be many arguments floated as to why commercial software is better than free software. Many of these arguments are variations on the master salesman's boast that he can sell more $10 bills for $20 than a less convincing huckster can give away. Such arguments can be defeated by establishing that free software is quality software and makes sound economic sense. Some arguments are more substantial: commercial software companies have a huge head start; some such companies have convinced many users to trust their brands; the true costs of software include the time that it takes to learn and use, so no software is really cost-free; the investment that users and companies have in commercial software can make switching painful; many people still regard commercial software as something of a bargain.
One issue that needs to be recognized and understood is the notion that free software, openly published in source form and freely inspected by anyone who has an interest or desire to do so, is worthy of far greater trust than closed, proprietary, secretive software. I for one found the installation of Microsoft's Internet Explorer to be a very scary experience: the computer running totally out of my control, recongifuring itself, plugging into Microsoft's own web sites, setting up preferences and defaults according to Microsoft's business machinations.
Sometimes I wonder whether Microsoft's underlying goal isn't simply to make the world safe for computer viruses. I'm not an especially paranoid person, but how can you ever know?
Consumers nowadays are so often (and so effectively) fleeced that there is much resistance to paying for something you can get away with not paying for, so this will be an uphill educational battle. There is a game theory problem here: Who should I commit to paying for a development which I can get for nothing if only I wait for someone else to pay for it? But if everyone waits, no one benefits.
There are other ways to handle this level of funding, such as imposing taxes on computer hardware (sort of like the gas tax is used to build roads) or even on commercial software (sort of like using cigarette taxes for public health). Developing countries, in particular, should support free software development, since the notion of intellectual property must appear to them as one more form of tribute to the rich. These approaches require political efforts that are sure to be contested and hamstrung. I'm inclined to start small, start voluntarily, and see how far reason and civility takes us.
It should also be emphasized that there is at present a substantial amount of free software already written and available, and that there are many organizations and individuals that have contributed to the development and dissemination and support of free software. What is missing is a systematic approach to funding development, and a strong and consistent system for user feedback and direction.
- I would estimate that free software can be developed to quality standards that meet/exceed commercial software for less than 25% of the price of equivalent commercial software. This estimate is based on common R&D expenditure levels plus a generous amount for those organizations which coordinate development and promote use. Given that free software is not compelled to become obsolescent (it can continue to be used as long as it is useful, whereas commercial software must obsolete old product to promote the sales of new), the costs for free software will decline over time, sharply except for the cases where new needs arise.
Much of this work is already being done. What's missing is not so much the people or even the organization as a coherent sense of the economic imperatives. To date, free software has largely been driven by political sensibilities and the traditions of academic freedom, which have led it into a hodge podge of areas, many of which have very little impact on common needs and usages. (Some, such as the Web, have had major impact, and as such have attracted enormous commercial attention.) However, the driving force behind free software must be economics: why do we spend so much money propping up empires when all we really want are clean, simple programs that do our work? And why do software professions have to work for commercial companies when their skills and work are more immediately needed by users?
The argument that large companies (government, any organization that spends serious money on software) should routinely support free software development is strong and well focused. Even if such an organization never directly used free software, its existence would provide a damper on prices and a strong bargaining point with commercial software vendors. It is a win/win bet: free software, cheaper software, more options, more competition.
It is completely obvious that free software organizations must be international in scope. It seems likely that most of the support for free software will come from outside the US, perhaps by an overwhelming margin.
This proposal does not dispute the rights of intellectual property owners. Under this proposal it should be possible to buy or license technology where appropriate, and inventors should consider the possibility of selling their inventions to the free world. Whether intellectual property rights in fact encourage innovation in any useful way can be debated separately.
Another aspect of this proposal is that it does not try to kill off the profit motive in software development. As I envision it, most of the free software work would be done by small companies bidding on contract proposals, presumably with the intent of making a profit. (The companies are likely to be small because they won't need to float a large marketing/sales organization, which is the main advantage big software companies have over small ones. Also because the free software networking organizations should work for providing sharable resources, such as capital and services, saving small companies from having to overextend themselves.)
My proposal is that free software will start out aiming to produce the most basic and most broadly used software: it will in effect harvest the “cash cows” of the commercial software industry, rather than attempt to innovate at the fringes of development. (Of course, innovators are more than welcome to contribute.) Beyond free software there will still be shareware and commercial products, which will to some extent compete with free software and to a larger extent open up new niches where free software is not yet available. The free software industry will provide a damper on the sort of prices that can be charged. It will also help lower the costs of all software development, and may eventually provide a salvage market for discontinued commercial software. Shareware may be a fruitful ground for speculative software development, with the goal being to develop and popularize a new product that can be sold off to the free market.
Finally, I believe that no restrictions should be placed on the use of free software: that it can be repackaged, sold, incorporated into commercial products. Free software will reduce the development costs of commercial software, which will help make commercial software cheaper, better, more competitive: all good things. The goal after all is better, cheaper, more usable and useful software: victory is not measured in bankruptcies. The impulse to segregate free software from commercial software is doomed, as is the impulse to isolate free software from commerce. We live in a jungle of commerce, which no one can truly flee from, regardless of how offensive it may seem. The proposal here is to start to take short, deliberate, sensible steps toward reclaiming parts of that jungle for everyone's use and betterment.
This implies, of course, that (following the Reagan demonology) Microsoft et al. are “The Evil Empire.” That's a joke, of course, but if it didn't harbor a shred of truth it wouldn't be funny.