GPL and Apache licenses

Discussion of programming on Linux, including shell scripting, perl, python, c/c++, mono, java. Whatever tickles your fancy.
Post Reply
arifishaq
Lance Naik
Posts: 39
Joined: Fri Oct 03, 2003 2:20 pm

GPL and Apache licenses

Post by arifishaq »

This is from the ZNet site: http://www.zmag.org/weluser.htm

The article is here: http://www.zmag.org/content/showarticle ... temID=4894

ZNet | Global Economics

Something new in Free Software?
by miradarebelde; January 26, 2004

Intellectual property rights and their protection have become the favorite battleground for global capitalistic expansion.

In developed countries, the most important input (factor) to productive processes of all kinds is know-how, meaning both technology and information workflows. From here, instructions to run and control productive processes of material goods are sent to the developing world, where most of these processes actually take place.

Patents are a means of satisfying market requirements for transforming knowledge, ('merchandise' that, due to its very nature, can be exchanged indefinitely) into a scarce resource, in such a way that market-driven exploitation and control mechanisms can be applied to them just as with any traditional goods.

This is why a central concern of the movement for global justice is to shake off the shackles that markets would like to use to imprison knowledge.

What is necessary is a model of intellectual property protection alternative to the current one, which can be effectively countered within civil society and at the level of government or political institutions. This alternative model should prevent the commercial exploitation of knowledge from creating obstacles to its further development, on the one hand, and from building up invisible monopolies, on the other.

One of the most interesting such models is the one made popular by Free Software and its philosophically less radical brother, the Open Source model. Neither get rid of the idea of copyright. Instead, they reframe it in such a way that while allowing commercial exploitation of software products, the knowledge built in them cannot be made anyone's exclusive property.

Free Software and Open Source have proved themselves viable, efficient, and effective models for software development as seen in their most visible success story: the GNU/Linux project. GNU/Linux makes up a whole computer operating system that thanks to its performance, sturdiness and security outdoes all major competing proprietary operating systems.

To make a long story short, both models require software to be made available to anyone in its 'human readable' form, i.e. by providing its source code along with executable binaries. Anyone who wishes to can modify it to suit their needs and then circulate their enhanced versions under the same model. As a result, a huge world-wide programmer community supporting the Free Software and/or Open Source models grew and succeeded in creating a complete and free (meaning both non-proprietary and no-cost) alternative to Windows and similar products. (From here on, I am going to speak of Free software for both Free Software and Open Source).

On closer analysis, though, it becomes clear that while these models have created a kind of niche where knowledge is effectively shared and collectively built upon, they have no real power to influence the dynamics that rule the computer software know-how market. On the contrary, they have made a huge no-cost reservoir of high-quality software libraries, programs, and development tools available to IT corporations. As a result these corporations can save large sums of money on software development and maintenance, also freeing themselves from dependence on other IT corporations' proprietary solutions (which is such a great risk that it has its own name: vendor lock-in).

All these advantages for IT corporations are not counterbalanced in any way. In particular, corporations retain full control of their work. Their solutions remain proprietary and protected in spite of basing them upon or using Free Software. Let's try to see why and how this happens by considering two examples.

Say that a company plans to develop a software program by relying on a Free Software component, for example, for data compression, cryptography, image processing, or a distributed communication library. This company would fully comply with the requirements of the Free Software license simply by making freely available the source code of that part of its own program that directly uses the Free Software component.

Contrary to what we might expect, this is not enough to guarantee that the knowledge built into that software program does not become the exclusive property of that company. The company may indeed adopt a very simple strategy to honor the Free Software license while preserving its right to commercially exploit that software program.

This can be accomplished by patenting an algorithm implemented in the company's own part of the program, thus effectively ensuring that no one else can use the knowledge associated with it, although it is accessible through the source code or the patent itself. Anyone else trying to use the same approach to that kind of problem, and using the same Free Software, would infringe on that patent, even the developers of the original component trying to enhance it and extend its functionality by adding features similar to the patented ones.

Let's consider another important example: that of embedded systems, i.e., systems that are made of tightly coupled hardware and software, to the extent that it is not easy for non technical people to distinguish between the two. This could be a home appliance, a printer, or any advanced peripheral device, for example the 'intelligent' ones that are presently appearing on the market.

In this case, the key is the proprietary hardware, which is not required to comply with the Free Software license and therefore can be kept secret or, of course, patented. Indeed, the software part of the system will be strongly dependent on the hardware part, to the point of being useless without it. Therefore, once again, releasing the source code does not change the fact that knowledge residing in that device is not shared at all. As a matter of fact, more and more advanced devices made by major IT corporations make use of Free Software components, and can even be based upon the entire Linux system. Nevertheless the knowledge and techniques built into those devices are actively and effectively protected from any external access, use, or modification.

Summing it all up, the Free Software and Open Source models fail to impact on the dynamics which currently underpin the privatization and appropriation of knowledge, all the fuss about its novelty and radicalness notwithstanding. This is a main shortcoming of all the efforts which surround the idea of Free or Open Source software and hints at their inadequacy as a means of altering the actual power imbalance.

This is the reason why the support for the campaign against software patents has grown in recent times and the issue has come in the focus of the global justice movement. However, this campaign has two fundamental drawbacks:

1. It can do nothing against knowledge privatization in embedded systems, which are of critical importance in future perspective;

2. It brings the fight back within the traditional arena of lobbying in order to gain improvements on existing norms and laws, thus relying on an enlightened political class that wants to take the input coming from below into consideration.

But something new is appearing on the horizon, which could in fact reinvigorate the capacity of Free Software and Open Source to change market dynamics, thanks to their intrinsic strength.

This is best explained by a concrete effort which is taking place to define a new license for the most widely used web server on earth, Apache. The actual proposal for version 2 of the Apache license, states that whosoever uses this software renounces the right to 'institute patent litigation' claiming that patents taken out on products based on any part of the Apache software are automatically null and void.

This can be seen as a sort of reciprocity clause: if you (i.e., big IT corporations using Apache for any kind of software system) take advantage of the knowledge residing in my software, then I (i.e. the whole Apache users community) can take advantage of that residing in yours. It is as simple as that: enforcing respect of the Free Software (in its many incarnations) philosophy by means of a more careful definition of the legal requirements that its use implies.

This will give a headache to most IT corporations, but it is clearly only an example, a start, and a hint at a way to give back to Free Software its potential as a concrete means to alter the balance of forces in the software field, strengthen effective knowledge sharing and fight unjust knowledge privatization.

This model could be applied in other knowledge fields as well, for example that of the traditional know-how of native peoples all around the world, which is being plundered by transnational corporations, and much more than this. It can be just the beginning of taking back what is ours.
Post Reply