facebook
twitter
vk
instagram
linkedin
google+
tumblr
akademia
youtube
skype
mendeley
Wiki

LINUX vs. WINDOWS: SOME THOUGHTS ON PATH DEPENDENCE

Автор Доклада: 
Protasova T. I.
Награда: 
LINUX vs. WINDOWS: SOME THOUGHTS ON PATH DEPENDENCE
LINUX vs. WINDOWS: SOME THOUGHTS ON PATH DEPENDENCE

УДК 339.923

LINUX vs. WINDOWS: SOME THOUGHTS ON PATH DEPENDENCE

Protasova Tetiana Igorivna, student
Central European University


Path dependence is frequently used as a motivating factor for antitrust regulations against Microsoft, aimed at correcting a market failure. However, the case of Windows does not constitute a market failure. A market failure implies inevitability, and in Linux’s case path dependence can be overcome. Currently Windows prevails in the desktop operating system (OS) sector. As the market drives standardization, sometimes it just so happens that it values factors other than work efficiency more. Computer games, overall image and popularity of a distributive, etc. all bring utility to consumers. These factors currently outweigh the satisfaction users can get from a larger speed of use or system safety. However, it will be shown that Linux has great potential for open source standardization and is a decent rival for Microsoft’s Windows.
Keywords: Windows vs. Linux, path dependence, technical efficiency, lock-in, network effect.

INTRODUCTION
According to recent studies, an average adult in the US spends 30% of his waking hours on a computer every day (BLS, 2009). Within the computer market, Microsoft’s share in desktop operating systems (OSs) is 87% (w3counter). This is why the possibility of Windows being inferior to any other desktop operating system triggers our concern and interest. The case of Windows and Linux was chosen for multiple reasons: not only do the two systems have core differences in software construction, but as an open source computer operating system, Linux’s market strategies are in principle of a different origin. While Microsoft’s Windows operates as a standard market production model, where a product is produced and sold to end users, Linux operates under open source model. As open source software, its source code is published and available to the general public, which implies multiple benefits over the standard model. These major differences support the concept that path dependence is the reason for Microsoft’s lock-in as a leader in the desktop operating systems sector.

According to S. J. Liebowitz and S. E. Margolis, path dependence means that where we go next depends not only on where we are now, but also upon where we have been in the past (1999). This dependence on historical events can have major implications for economic development. In 1989 Brian Arthur introduced the term ‘increasing returns to adoption’. He argued that specific practices become more valuable to each user as the total number of users rises. These increasing returns magnify small, random events in the market. Arthur argues that these principles are especially significant in technology-specific industries. These industries require standartization and grow immensely fast, thus widening the gap between rivals (Arthur, 1989). One small flaw which Liebovitz and Margolis notice in the theory of path dependence, however, is that bad economic outcomes are avoidable by small but prudent interventions (1999). This simply means, bad outcomes can be overcome with low costs, but in the case of Windows and Linux costs are not really low. Switching standards implies costs other than just the economic ones. In their Winners, Losers & Microsoft, Liebowitz and Margolis criticize the government's antitrust case against Microsoft. They argue that government lawyers are wrong to conclude that consumers are forced to accept inferior standards and high prices. People choose what they want, and what they want survives, at least for a while (1999). The current paper follows the existing line of thought, describing why switching all desktop computers to Linux may be costly and provides evidence against the notion of market failure in regards to our case. Unlike Liebowitz and Margolis’ aforementioned Windows case, it will be argued that Windows is not a better product in all aspects, especially in its core technical constituents. It is a market choice for reasons other than the technical benefits.

In 1996 Mark Roe offered a categorization of path dependence. He labels different forms of path dependence as weak, semi strong and strong. Weak form path dependence is when the chosen path is not inferior to other alternatives, but the alternative may be preferable to the general public. Semi strong path dependence is when the chosen path is not the best but not worth fixing. And only the strong form is when the chosen path is highly inefficient, but we are unable to correct it (Roe, 1996). In key characteristics, e.g. technical efficiency and consumer feedback, Windows proves to be inferior to other alternatives. However, the market is able to correct the inefficiency. Thus, the case of Windows can be categorized as semi strong form path dependent. Although inferior in terms of efficiency, Windows proves to be a consumer choice and a prevailing standard for other reasons than hardware conversion costs. Here not technological efficiency, but the psychological factor seems to be the deciding factor. Brian Arthur found that path inefficiency is possible where there are increasing returns. He argued that strong form path dependence can exist only when the information regarding the returns to each choice is available to relevant decision makers (1989). In the case of Linux and Windows desktop OSs however, this information is not truly available to the end user. Thus our case cannot be defined as market failure.

I. MARKET MODEL
The market model that is chosen by an OS, has major implications for the consumers’ surpluss. While Microsoft’s Windows operates as a standard market production and development model, where a product is produced and sold to end-users, Linux operates under the open source model. As an open source software, its source code is published and available to the general public. Thus anybody is allowed to copy, modify and redistribute the source code free of loyalties or fees. It is generally arged that open source operating systems:„increase social welfare by increasing the viability of such development efforts and thereby lowering prices” (Casadesus-Masanell and Ghemawat, 2006). The lowering prices are insured by the open source legal structure.

So, as an open source product Linux-based OSs are totally free to download and use, update or add any programs one may need. Any person who has internet access can download Linux on his computer and easily get acquainted with it. Further, one can easily download any program for Linux, or even download a few similar programs and choose the best one - a program that can provide maximum efficiency to a specific user.

Many then pose a simple question: how does it function if it is free? This concept does not seem to go in line with the capitalist market model we are used to. However, open source generates good revenues, and its main source of profit lies in service provision, e.g. customer support, consulting firms and corporations, tutoring and on-line tech support. As to individual programmers who are willing to contribute, open source is a way to make a portfolio for the employer and become accredited in the market. Further, there are plenty of Linux-based distributives that are sold, just like Windows’ products, only generally cheaper. Like all distributives using Linux kernel, these systems possess another substantial advantage over Windows - this advantage is efficiency.

II. TECHNICAL EFFICIENCY
Historically it so happened, that Linux kernel was used at major research centers, and thus its operating efficiency was the core motivator for improvement. Interface friendliness and availability of computer games were put aside for better days, as Linux was initially designed for work. Historical dependence has insured that Linux, that took the non-commercial path and operated in research institutes, was locked in that particular user group. Meanwhile, Microsoft put aside the technical imperfections and aimed at giving the market what it needed - and it needed entertainment by large. However, Linux’s specialization has led it to outrun Windows in terms of efficiency. In the coming days, operating efficiency may prove to be more important than the interface improvements, as the kernel constitutes the base for an OS’s later success in any specialization path it desires to take.

Now let us consider the technical differences that ensure Linux’s superiority in terms of efficiency. First of all, Linux uses RAM memory more effectively than Windows. Linux’s "swap partition", a special feature aimed exclusively at paging operations, reduces slowdown due to disk fragmentation at the time of general use (Ubuntu Documentation, 2010). Disks are much slower than RAM, so the fact that Linux can use RAM memory for much longer before swapping to disk, is an important comparative advantage. Windows however does not support such features. Its processes are more frequently moved out of RAM to swap space on disk; hence, users generally experience slower response time from the computer compared to Linux. Moreover, the function of swap partition is missing in Windows completely and is replaced by a swap file; the input-output speed of the later is much lower.

Apart from a more effective RAM usage, Linux’s core uses less memory and thus can run on any simple device, as simple as a microwave oven. Further, effective usage of Shared Libraries helps economize on space and allows downloading updates that weigh much less than Windows’ updates. This saves time, as well as Internet traffic. Windows does not use Shared Libraries effectively. Even if two programs share one library, the library is usually double-saved in two different folders of each program. Windows duplicates the libraries; this is why Microsoft Office can take up a gigabyte of memory, while Open Office, which has almost identical functional characteristics, needs just 190 mega bites. Moreover, versions of the libraries may differ slightly, but this ends up in complicating the updating process for Windows.

Linux is also much safer. Viruses designed for Windows do not function with Linux. Furthermore, users’ safety and privacy is ensured through constant updates that are entirely free. These updates function in such a way that the system is changing its characteristics, mutating so to say, every time it is updated, hence it simply cannot “catch” a virus. A virus requires stability, a stability that only Windows can provide.

III. REASONS FOR PATH DEPENDENCE
Given all its imperfections in comparison to Linux, why does Microsoft prevail in the desktop sector? Microsoft’s share is 87% compared to Linux’s 1.6% (registered distributives only, data from w3counter). Thus Windows’ economy of scale proves to be a great moving force. As shown by Scott E. Page (2006), positive externalities associated with increasing returns to scale exaggerate the degree of path dependence. For Microsoft it means lower unit costs, enormous contributions to marketing, lobbying to hardware producers and forming a positive public opinion with the users.

So what are the costs concerned with changing the standard to Linux? First, there are the hardware conversion costs. Some of the devices may need drivers which work with Windows only and it may take a while to find a driver for a specific device to work with Linux. Right now, however, Linux developers strive to make universal drivers to fit multiple devices. Secondly, some argue that, speaking globally, there is a shortage of specialists in day-to-day software support (Computer Economics Report, 2006). However, specialists are usually replaced by forums and communities, which are eager to help out. Journals and magazines provide tutoring and advice too.

However, the main reason for semi-strong form past dependence in the case of Windows, is that the market suffers information asymmetry and a fear of change. Only a limited number of people are aware of Linux’s benefits over Windows, and the majority of others consider it a system for programmers and computer fanatics. Poor marketing and historical specialization have created an erroneous image in the minds of consumers. Another, seemingly irrational, reason for the lock-in is that, for many, the availability of computer games is a great deal breaker. Although Linux provides its users with some games, the majority of new complex games are designed for Windows. An average user values his recreational time at home and thus buys a system that can satisfy his wants and needs the best. By choosing Windows at home, he gets used to its interface and later desires to have the same operating system at work, forcing corporations to buy expensive licensed versions for their employees. These corporations constitute the main source of income for Microsoft. This same market strategy aimed at domestic users, applies to illegal software usage. The fact that Microsoft does not take strong measures against illegal domestic use of its OSs is, same as in the case of computer games, in its contribution to the process of standardization.

To conclude, it should be noted that although the case of Windows and Linux can be considered semi strong form path dependent, there is clearly specialization involved. This year Linux constitutes almost 100% in the Top 500 supercomputers, over 90% in the server sector and is one of the leaders in the mobile phone firmware sector, with its Android (Linux and Java based mobile phone firmware). Thus Linux’s higher functional efficiency proves an effective moving force for its domination in the sector of computers designed for work. Here no degree of market aversion to Linux can set it back.

Sources used:
1. U.S. Bureau of Labor Statistics, American Time Use Survey, 2009. Accessed 4 April 2011. 
2. W3Counter. Accessed 4 April 2011.
3. Stephen E. Margolis. "Path Dependence. Evidence for Third- Degree Path Dependence", 1999. Accessed 9 April 2011. 
4. Mark Roe. "Chaos and Evolution in Law and Economics," 109 Harvard Law Review 641 (1996)
5. "Competing Technologies, Increasing Returns and Lock-in by Historical Events." Economic Journal, 99. 106-131, 1989.
6. Stan J. Leibowitz. “Winners, Losers, and Microsoft: Competition and Anti-Trust in High Technology.” Independent Institute; Revised edition (March 1, 2001)
7. Scott E. Page. “Path Dependence,” Quarterly Journal of Political Science, 2006, 1: 87–115
8. Community Ubuntu Documentation. "Swap Faq". Accessed 4 April 2011.
10. Ramon Casadesus-Masanell, Pankaj Ghemawat. “Dynamic Mixed Duopoly: A Model Motivated by Linux vs. Windows” Management Science, Vol. 52, No. 7, July 2006, pp. 1072–1084

10
Your rating: None Average: 10 (9 votes)
PARTNERS
 
 
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
image
Would you like to know all the news about GISAP project and be up to date of all news from GISAP? Register for free news right now and you will be receiving them on your e-mail right away as soon as they are published on GISAP portal.