There is a terrific war going on in the computer industry today, which has long-term implications for giants in the industry, especially IBM and Microsoft. It is instructive to those in India who blindly worship Microsoft, whereas the Linux direction is especially appealing for a developing country, in particular one that has the technical wherewithal to use open source software.
Fortunately, there is the Free Software Foundation of India http://fsf.org.in: appropriately enough for a radical egalitarian idea, it is located in Kerala!
It is clear that there is tremendous momentum behind the open source software community, as exemplified by Linux, GNU, Apache, Mozilla and other popular applications, for which the original program, or source code, is available free of charge to all users. This is totally contrary to the proprietary software approach of many vendors, notably Microsoft, which guard their source code most zealously.
The success of open source has been due primarily to the fact that a large number of expert programmers all over the world contribute their work for free, in the spirit of co-operation and community. Their work is shared by the 'copyleft' concept, whereby the improvements made by anybody can be used by anybody else, but in turn, their innovations have to be made available to everyone else as well.
Astonishingly, this chaos works. A perceptive essay by Eric Raymond called The Cathedral and the Bazaar is a good introduction to the rationale and the practices of the open source movement. The end result is that Linux, the best-known example, has now become a legitimate threat to other operating systems, especially Microsoft's Windows.
An operating system is the master controller program that runs on a computer: it is in effect the guts of the system, and is therefore a critical piece of software.
To talk about Linux, we need to look at the AT&T Bell Labs operating system Unix, which was invented in 1969 by Dennis Ritchie and Ken Thompson and described in a landmark paper in the Bell System Technical Journal in 1974 or so in an article entitled 'The Unix Time Sharing System.'
It was a revolutionary operating system, one that was created essentially by just two people in one year, Ritchie and Thompson, whereas conventional OS's, such as IBM's operating systems for mainframes, were written by teams of hundreds of people toiling for years. And Unix worked better than anything else in the market, and had the sterling distinction of being written in C, an easily understood language, as compared to the arcane assembly language (much easier for machines to understand than for humans) which other operating systems used.
As Unix was easily understood, and because AT&T made its source-code available free of charge to universities (not because of any altruism, but because as a regulated monopoly at the time, they were not allowed to profit commercially from the inventions of Bell Labs), Unix became the basis of a most of the interesting work in operating systems around the world. The work done at the University of California, Berkeley, was especially important, resulting in the BSD stream of Unix versions.
Linus Torvalds, a Finnish student, set out to rewrite the 'kernel' or the core of the Unix system, himself, so that it would be free of any licenses to AT&T. Richard Stallman and associates at the Free Software Foundation (creator of the concept of 'copyleft') and the GNU Project ('Gnu's Not Unix') had already written most of the rest of the operating system in AT&T-free code. Together, their efforts have resulted in what is currently referred to as Linux, which should properly be called GNU-Linux.
An army of volunteer programmers around the world contributes to the Linux effort by picking up and improving various components and then offering their versions back to a central repository, controlled loosely by a few 'wizards.' The only remuneration the programmers get is through the recognition of their peers that they are 'alpha programmers:' all they get is bragging rights if their code is accepted.
Now fast forward a few years. Microsoft is feeling the heat from Linux, as the free operating system is improving by leaps and bounds. Unix itself, which could, at one time, have stopped the Microsoft juggernaut in its tracks, was doomed by a schism in the ranks, which pitted Sun Microsystems and AT&T against IBM, DEC, HP, etc. I was in the thick of that battle, and I now see we were irresponsible to fight internecine battles, trying to push our own versions of Unix, while Microsoft ran away with the prize: control over the desktop, and the untold billions that comes with it.
But Linux, free from the institutional greed that killed Unix's chances, has thrived. And it has found some strong backers, such as IBM. To a lesser extent, HP, Dell, and so forth, have also jumped on the Linux bandwagon, because they all recognize that Microsoft has been skimming away most of the profits through its monopoly on the operating system.
The net result is that today one can buy a Linux-based PC for far less money than a Microsoft Windows-based PC. It will, it is pretty much guaranteed, work much better, and be less resource hungry: you don't have to throw away your computer every couple of years just because Microsoft's new OS is a resource hog.
Linux, with its compelling advantages, has put the brakes on Microsoft's attempts to dominate the server market as much as it dominates the desktop market. It is much more robust, it costs essentially nothing; if problems are found, the army of Linux enthusiasts will fix them practically overnight; and because of its Unix foundations, it is far more suited to complex networks than Windows, which had its origins in a throw-away system called QDOS, Quick and Dirty Operating System!
In addition, Linux is making waves in the world of consumer electronics: see the article 'Linux Set to Break Through in Consumer Electronics' in the New York Times, August 29, 2003. Sony, Matsushita, Philips and other majors in the world of TVs, phones, DVD players and set-top boxes find Linux's flexibility and low cost irresistible, not to mention the prospect of not having to pay Microsoft a 'tax.' Furthermore, as home networks proliferate, the consumer will want to have the electronic devices interact seamlessly with PCs, and here Linux's facility with networking will be helpful.
Can nothing stop Linux then? Well, there are some people trying to stop do just that. I used to work at AT&T Bell Labs' Unix development group: this was later sold off to Novell, then to the Santa Cruz Operation, and finally to Caldera, which renamed itself SCO. Therefore Caldera has the original Bell Labs Unix source code and the intellectual property associated with it.
Caldera, a small and relatively unsuccessful software maker from Utah, has become the dark horse in the Linux sweepstakes. Caldera apparently decided that its assets, that is the intellectual property associated with Unix source code, were being violated by Linux users because Linux developers had 'stolen' Unix code.
Caldera went ahead and sued IBM in March for $3 billion. In June, they further sent out a letter to 1,500 corporate users of Linux suggesting that they would be freed of legal problems if they bought licenses from Caldera for every copy of Linux they used. I think the price was a mouth-watering $699 a copy, but I could be mistaken.
The reaction from the developer community was predictable: they argued that Caldera's action was without merit, and that it was motivated primarily by greed. The original thinking was that Caldera was probably angling for a buyout by IBM, which would make sense from the giant's point of view: this is an annoying little distraction. For a comprehensive response from Eric Raymond, see the 'OSI Position Paper on the the SCO-vs.-IBM Complaint' http://www.opensource.org/sco-vs-ibm.html
In the middle of all this, there was additional high drama. Microsoft with a great show of probity, paid Caldera for a license which it didn't need. This did not surprise observers, who believed all along that Microsoft may be secretly funding SCO: for any FUD (Fear, Uncertainty and Doubt) about Linux would only help Microsoft push Windows. Microsoft is not averse to using all sorts of nasty maneuvers: for example see its interactions with a small company, Blast.com, according to columnist Robert X. Cringely in 'Stupid Microsoft Tricks: Why the Richest Company on Earth Feels it Needs to Cheat' http://www.pbs.org/cringely/pulpit/pulpit20030828.html
Interestingly, IBM has counter-sued SCO, and some of the claims include allegations that their code has been used in SCO's own offering of Linux (oh yes, I forgot to mention that SCO had a rather unsuccessful Linux offering in the market) and that, in any case, SCO was in violation of the GPL (or 'copyleft'). IBM has also said that it bought a permanent, non-revocable license from AT&T in the old days, which should suffice.
IBM also claims that SCO has so far not demonstrated any of the code that it claims has been illegally used by Linux, and that therefore it is hard to take them seriously. I can understand SCO's dilemma, though: assuming there is merit to what they are saying, as soon as they display the alleged violating code, the Linux folks will immediately rewrite it as SCO-free code, thus nullifying SCO's case!
In the meantime, Novell, an interim owner of Unix copyright, has said that SCO owns neither the patents nor the copyright associated with Unix; the implication is that the SCO suit has no leg to stand on.
Furthermore, in a previous instance where AT&T and Novell went to court against Cal, Berkeley regarding BSD source code, the result was in favor of the university, since it was fairly apparent that significant added value had been created by the researchers over the value of the original source code.
It is probably a fair assumption that the original Bell Labs code is by now relatively obsolete, and that a lot new innovation has been added by the free software people. Therefore, it is probable that if this comes to a court hearing, SCO will lose. But in the meantime, if enough people have been terrified into a wait-and-see attitude on Linux, or into continuing with Windows, that may be a serious blow to the open source movement.
How ironic that IBM, which originally created the strategy of using FUD as a sales tool, is now being victimized by it! The IBM of today, of course, is not the Darth Vader of computing that it used to be. (The dubious distinction must belong to Microsoft these days.) If through its buccaneer tactics, Caldera (and possibly lurking in the background, Microsoft) succeeds in damaging the momentum behind Linux, that would be a great shame indeed.
Comments are welcome at rajeevs@rediff.co.in
