Near miss at Mumbai airport: accidents waiting to happen  

D. Raghunandan

11^th June 2009 

A potentially horrendous disaster was somehow averted at Mumbai airport on May 31. Two aircraft with close to 250 passengers and crew on board simultaneously tried to take off from Mumbai’s two intersecting runways and would have collided with each other but for last minute action by both pilots and Air Traffic Control (ATC) to abort the take-offs. As is customary in such cases, an inquiry is being conducted by the Director General of Civil Aviation (DGCA) while the pilots and the air traffic controllers involved have been taken off duty pending the investigation.  No firm conclusions can therefore yet be drawn as to causes. But again as usually happens in India leaks and rumours are flying thick and fast, obscuring the truth and building up a smokescreen behind which everyone, and thus no one in particular, is to blame. Yet there are already a number of indicators pointing to structural problems in Indian civil aviation safety with many factors contributing to accidents just waiting to happen.      

The near collision              In aviation parlance, a “near miss” is an “unplanned event that did not result in casualties or damage”. Many experts however believe that the term “near collision” is a more useful term since it better captures the reality that an accident almost occurred, rather than appearing to somewhat contradictorily suggest, as “near miss” does, that an accident was almost avoided. In any case, the point is that near collisions are always closely studied worldwide because fortunately, they happen hundreds or thousands more often than actual accidents and, at no cost, they provide invaluable lessons about what could go wrong, how and why, and what corrective steps need to be taken to prevent future such occurrences especially at the systemic level of safety procedures and institutional safeguards.  The near collision in Mumbai occurred when Air India’s Airbus 310 flight 348 to Delhi was speeding down one runway (Runway 9-27) prior to take-off, while simultaneously Jet Airways Boeing 737 9W-651 to Kolkata was similarly building up speed for take-off on the other intersecting runway (Runway 14-32). IC-348 was traveling at 70-80 knots (127-145 kmph), whereas the Jet craft was traveling at 90-100 knots (164-182 kmph), dangerously close to the “decision speed” (beyond which aborting take-offs could cause a crash) of 129 knots for B-737. Fortunately, ATC noticed the Air India plane rolling and directed the pilot to abort take-off, and the Jet pilot managed to do the same, apparently on his own initiative. How both aircraft were rolling to take-off at the same time, what instructions did they receive from ATC and when, and how exactly was the accident avoided, are precisely the subjects of the mandatory investigations now on.           

Ambiguous communication        Pilots of both the aircraft insist they had received clearance for take-off from ATC. If this is found to be true from examination of the radio-communication recordings, which surely should have become available and been fully transcribed by now, then the controllers in question have a lot to answer for. Equally, ATC could not knowingly have cleared two aircraft for take-off simultaneously on intersecting runways. Clearly some miscommunication had occurred, either by the ATC or by one or both of the pilots in understanding instructions received from ATC. ATC spokesmen said take-off clearance had been given only to the Jet flight, not to the Air India flight and that, when the latter was noticed traveling down the runway, it was ordered to stop. A General Manager of the Airport responsible for air traffic and other air-side operations reiterated this position.  But a spokesman for Air India insisted that ATC had given its pilot clear instructions: “AI 348 cleared for take off, Runway 27”.  However, according to some press reports, the spokesman also stated that “the AI flight was cleared for take-off and after that the pilots were asked to hold.”  If so, this would be astonishing indeed for, once “take-off” is “cleared”, there is no question of telling the aircraft to “hold” or wait anywhere! If the controller’s intention was to tell the Air India plane to taxi to Runway 27, hold and be ready for take-off, then the instruction should not have used the words “cleared” or “take-off” but “ready”, “hold” and “departure” instead. Further, whatever instructions were issued by ATC, these would have immediately been repeated verbatim to ATC by the pilot as per standard operating procedure to confirm the instructions eg “Roger, Control, AI-348 cleared and ready for take-off, Runway 27,” which would have given ATC another chance to realize the error and stop the aircraft from rolling.  The cockpit voice recordings should clear up what instructions were issued and acknowledged. It is surprising indeed that, more than 10 days after the incident, doubts on this score have been allowed to linger despite DGCA having said that the inquiry would be completed in a “couple of days”.   Air traffic control procedures are designed to obviate misunderstandings. Given differences in nationality, language, idiomatic usage, accents etc, standard and unambiguous terminology in English has been approved by the International Civil Aviation Organization (ICAO) and is used by controllers and pilots worldwide.  In one of the worst air accidents ever, two Boeing 747 Jumbo jets collided on the runway at Tenerife airport on a Spanish island in 1977 killing over 600 passengers and crew, when one of the aircraft was “cleared for take-off” by ATC who actually meant to tell the pilot to wait for further instructions! Since Tenerife, the word “take-off” is never used except when actually setting the aircraft rolling on the runway. Similarly the words “cleared” and “approved” are never used except to signal go-ahead for final actions. In another near-miss, a pilot who was told to taxi and hold “short of the runway” could not understand the idiomatic English and thought he was being told to “cross the runway” and almost did! With pilots of different nationalities flying aircraft all over the world, including on aircraft under wet-lease with foreign pilots flying on domestic routes, there is even greater need to ensure precise unambiguous communication.    

Under-staffed, under-trained, overworked ATCOs   Unfortunately, incidents involving air traffic control are becoming more common not only because of increasing air traffic, especially in India with over 20 percent growth per year, but also due to a critical shortage of well-trained air traffic controllers (ATCOs). This is a worldwide phenomenon, but India is among the worst afflicted, with several additional structural problems at the heart of aviation safety and civil aviation infrastructure and institutions. Almost the entire emphasis of government in the past decade or more under both the NDA and UPA dispensations has been on privatization of airlines and airports, and on deregulation of civilian aviation in general. As a result, air traffic has expanded exponentially, but air traffic infrastructure and safety measures have not kept pace, and scant attention has been paid to these by government authorities. Delhi and Mumbai airports are both now under private management with foreign collaboration, an idea pushed through by the UPA government in the face of stiff opposition by Left parties, Airports Authority staff and many aviation experts all of whom argued that privatization of terminal and city-side operations would not add any value to airport operations whose main problems lay in poor air-side infrastructure and air-traffic management.  Today, both these airports handle more than 700 flights daily, but with barely 100 ATCOs each. In India as a whole, there are about 1000 ATCOs compared to an estimated requirement of 2500, and the recruitment pipeline is too thin to meet demand. ATCOs are recruited directly by AAI through examinations and interviews, and then sent for a year-long training programme in the single training institute in Allahabad that can handle only about 30 trainees in each batch. Another training academy planned in Hyderabad is yet to start work. Unlike many other countries, India does not have a system of licensing ATCOs as prevails with pilots, which could have increased the available pool. And the quality of training too, with few refresher courses, leaves much to be desired.  In most advanced countries, and in several developing nations, ATCOs are considered a special type of air service provider like pilots, due to the unique and important roles and responsibilities of both, with distinct working conditions and remuneration patterns. The International Labour Organization (ILO) holds that “ATCOs... have problems which are unique to their profession, and their concern with safety could broadly be compared with that of pilots.” In fact in many European countries, ATCOs operate under an entirely separate organization as in the UK’s National Air Traffic Service. Not so in India where ATCOs are just another set of AAI employees.  The problem of inadequate numbers of ATCOs, the managerial preference for paying overtime even at the cost of ATCO fatigue rather than bear the costs of recruiting and training additional ATCOs is undoubtedly a worldwide phenomenon. But it has assumed enormous proportions in India threatening to overwhelm the entire system.  

Infrastructure & Institutional Reform    Added to this is the problem of inadequate infrastructure and delays in or failure to upgrade technologies. Air Surface Movement Ground Control Systems should have been procured and installed at all major airports but are currently available only in Delhi and Mumbai, with fog-prone Kolkata having been sanctioned one after a near collision in January this year involving an aircraft aborting landing a mere 300 feet above ground after spotting a cargo plane on the runway! It is another matter that even with this system in place the Mumbai ATC did not notice both aircraft moving towards the intersection point last month! While the AAI is starved of funds and approvals are delayed, the Civilian Aviation Ministry is mostly busy with supporting private airport operators and addressing their concerns on building convention centres and shopping malls to increase their profitability!  Justice Lahoti heading the Inquiry Commission into the mid-air collision over Charkhi-Dadri near Delhi in November 1996 killing over 550 people observed that “the workload of [the] Controller was definitely excessive…” He also observed that “the Indian Govt adopted open sky policy but the same has… resulted in… tremendous increase in air traffic without matching additions to ATC infrastructure… For proper growth of civil aviation in the country, the ATC has to be given a special place in the scheme of things or else [our] country will continue to lag behind.”  It is ironic that these conclusions should have been reached by an essentially lay person whereas civil aviation authorities have continued to ignore what stares everyone in the face. Recognizing the special function of ATCOs, the ILO recommends that “trade unions and/or the appropriate organisation concerned should be consulted” on all issues relating to ATCOs working conditions and remuneration. In contrast in India the AAI, Government and regrettably even the Courts, have refused to recognize or deal with the ATCOs Guild and have put down every attempt at collective bargaining with an iron hand. Perhaps they are following the example of former US President Ronald Reagan who cemented his reputation as a right-wing conservative by crushing the ATC Unions’ strike early on in his Presidency. Not just the ATC but the entire system of air traffic management in India is crying out for major reforms and vastly improved infrastructure. Without such an overhaul, air safety in India will continue to suffer at high cost to airlines, passengers and the country’s reputation. Two other major institutional reforms are urgently required.  

Independent Safety & Regulatory Agencies             An independent Regulatory Authority for civilian aviation has been a long-standing demand, keeping in mind the experience of other countries. The US which actually de-regulated civil aviation during the Reagan years, is now living with the consequences of the collapse of many airlines, proliferation of fly-by-night operators and heightened problems of air safety. In India too, the pro-liberalization powers-that-be have resisted calls to set up an independent Regulator to look into routes, fares, flying conditions and so on, and have left it to the DGCA to perform this function too. DGCA already acts as the authority for licensing of airlines and pilots, type certification of aircraft and oversight of air safety.  On top of it all, DGCA is also tasked with conducting accident investigations! There is clearly a clash of interests involved here since DGCA signs off on aircraft maintenance and repairs, oversees air services, advises AAI on safety issues and also reports to the Minister of Civil Aviation. How then can one reasonably expect it to conduct accident or near-miss investigations properly, identify responsibility, and also make systemic recommendations which could go against or put pressure on any of its “client” organizations?  We still do not know who was responsible for the near-miss in Mumbai involving the President’s helicopter fleet in February this year. Initial investigations pointed to “negligence”, by who is not clear, and a final report is not yet available. In fact, 70 cases of airfield accidents all over India are pending with the DGCA since 2007. As in most cases with DGCA, interim reports have been announced but final reports and follow-up on recommendations are rare. It is high time an independent Air Safety Agency is created in India. The US has the National Transportation Safety Board besides the Federal Aviation Authority, and similar organizations exist in Canada, Australia, France, the UK and most European countries, precisely in order to avoid conflict of interest between bodies responsible for implementation and those called upon to investigate and recommend procedural or legislative changes. These safety agencies abroad have responsibilities spanning air, sea and land. It is too much to expect such cross-cutting of jurisdictions in India but a separate air safety authority is certainly feasible and eminently desirable. A veteran international ATCO (not from India) lamented that although detailed and meaningful recommendations have repeatedly been made "it always takes an accident for things to change." In India, regrettably, not even accidents or near-accidents seem to change things.                                  

Prabir Purkayastha

There has been a major battle that was fought and lost on Open Document Format, when the International Standard Organisation (ISO) finally accepted Microsoft supported OOXML as another document standard. Initially, ISO had voted against this standard, but after some countries switched their vote, it was finally passed by ISO in March this year. India and many other countries voted against the OOXML as an additional document format to the already existing Open Document Format, which is already an ISO standard. India, Brazil and South Africa have formally protested against the ISO vote and filed appeals against the ISO decision, claiming numerous violations of ISO procedures. The decision on their appeal is pending and should be taken by June.

There has been a lot of critique about Microsoft's OOXML format getting adopted as an ISO standard, foremost being that it is still largely proprietary, poorly documented, does not allow interoperability, and defective by design in order to preserve bad design decisions taken by Microsoft earlier. Finally, why have multiple standards - Open Document Standard is already an ISO standard and is being used widely. Adding to this list is the attempts by Microsoft to corrupt/arm-twist the voters as was shown during the ISO voting.

The Bureau of Indian Standards (BIS), India's standard making body had deliberated long and hard on OOXML. It was initially rejected unanimously by the BIS Technical Committee and finally, by rejected again before the ISO vote by 13 to 5 against OOXML. To become a standard, it has to receive two-thirds of the vote, here it was rejected by almost a two third vote!

However, we understand that the ministry of information technology is now proposing to adopt multiple standards for documents in its e-governance projects. At stake are two critical issues - who owns the data that is stored in documents implemented through such standards and the other is why should the government incur the cost of proprietary software when free and open source software is available from other sources?

Importance  of Standards

Why are standards important in the industry? Before we address the issue of OOXML and multiple document standards and all that it implies, we must understand why standards exist and how they influence the market. Standards make possible interchanging of parts in an equipment from different vendors or building equipment that will interface with equipment of other companies. For example, when you go the market and buy a screw of a certain size, you know that any nut of the same size can be used with that screw without any problems. Otherwise, for every equipment, we would have to buy the nuts and screws from only the company that originally made that particular product. It is this interchangeability or ease of interfacing that makes modern industry grow: it can specify the standard that it follows for all other products that follow the same product to be compatible with it. For the user, it is the freedom from tyranny from proprietary systems. Otherwise, they would be held captive to the original suppliers or the market leaders.

As standards allow for inter-operability, they are loved by the users or the consumers. For the same reason, they are acutely disliked by the dominant monopolies. For them, no standards means everybody will either have to buy from them or they will have to conform to what the dominant monopoly player defines as their interface. Therefore, monopolies argue that their products are the standards and everybody should be happy with this. Moreover, they also have strong patent and other protection built into their products to retain not only their dominant position, but also ensure that they can extract royalties from smaller players.

However, quite often the consumers or the bigger users get together and create standards for the industry. As the industry grows, various technical bodies try and create these standards so that the industry as a whole can grow. But this process is not an easy and comfortable one. It is generally a bitter battle that is waged by various parties before a technical consensus can be reached. Each of the bigger players try and use the standardisation process to get their products specifications as standard, just as countries use their standardisation procedure to keep their domestic market as a protected space for their industries. The use of for instance 110 V AC (the US and Japan) or 240 V AC (rest of  the world) is such an instance. The US wanted to protect its electrical appliance market from others and therefore chose to standardise on a different voltage. So also its use of NTSC or SECAM format for Television as against the PAL format used everywhere else.

Standards are also used as a tool to create global monopolies. Global corporations use their muscle power such that “their” standards can be adopted as global standards. This gives them a stranglehold on the technology through either a pool of patents or through their existing monopolistic position or both.

As and when the standard gets adopted as a global standard, all users are also forced to adopt the same. The companies that have a stranglehold on a particular standard can ensure that they will derive free royalties by virtue of that standard. A very good example of such a standard is MPEG4 that is used in DVD's. For example, the Chinese manufacturers produce DVD players that cost $20 to manufacture and then have to shell out approximately $20 as patent royalties since they have to use the MPEG 4 format, for reading the content. MPEG4 has a number of patents underlying it and is the perfect example of how standards can be used to build monopolies or extract monopoly rent.

Standards and the  Software Industry

The software industry had earlier no standards. However, with evolution of the industry, more and more new standards have appeared. Microsoft initially kept away from all standard making, claiming that it was the de facto standard and everybody had to become either a sub vendor (Microsoft Partner) or an user. In their books, standards was opening their software to external scrutiny, something they were extremely loath to do. However, this initial complete opposition was increasingly difficult to maintain, as the users will always prefer standardisation to non-standard proprietary products.

The defining moment for software standards is today and is being fought over open document formats. The need for a common format for all documents, which any software vendor would have to comply with is critical for the user. Suppose I have written a book in a particular software and that software manufacturer either folds up or asks exorbitant price for the next version of his software. I have to either pay or lose whatever I have written - the book now longer belongs to me. The vendor has a lock-in on the document. Even if I want to access my own work, I still need to pay the software vendor.

Who owns the Data  and the Documents?

This is the question we raised in the beginning - who owns the data that the software “codes” and “decodes” for us? Any document, music or video is coded by software into binary form (zeroes and ones), generally compressed in some way and then decoded for us to be able to understand it as a document or a piece of music or a film. The meaning of the document is restored by this act of decoding. However, if the decoding is requires either patented or proprietary software, we are now at the mercy of companies who can extort money from us at will. This is the world of “Intellectual Property Rights”, through which companies extract monopoly profits from the consumers. This is what brings us to the format wars, the key question being battled today.

The Open Document Format started as an initiative to bring the data of the users under the control of the users. It was an attempt to have an open and common format for word processing, spreadsheets and database. Once this format is accepted, then all software companies producing such software, would have to read and write documents in this common format. It means that a document produced in one software could be read in another, as the format in which the document is stored is independent of the software creating or reading the document. In other words, liberty of the user from the tyranny of the software vendors. After a long and protracted process, Open Document Format was accepted as a standard by ISO and has a number of implementation.

An Open Document Format is obviously anathema to Microsoft. It would mean that any document that is produced by Microsoft could also be read by other software. The vendor lock-in - only we can read your old document created through our software -- would be become a thing of the past. For Microsoft, fully 40 per cent of whose revenue comes from its Office suite - word-processing, spreadsheets and the database called MS ACCSESS, this would be disaster. Microsoft would have to compete by being better than others and not just from its existing monopoly.

When Open Document Format was being developed, Microsoft stayed out of it hoping the standard could be killed by ignoring it. If Microsoft did not implement ODF standard, it would be another standard which would just die out naturally.

Unfortunately for Microsoft and fortunately for the users, this did not happen. At least two widely used open office products conforming to Open Document Format are widely available. Google is implementing it also as net tool, making this even more easy to use and adopt. Worse, governments, worried about their data coming under a vendor lock-in and they being forced to pay fancy prices for the Office software suite every few years, started to prescribe software conforming to Standard protocols be used for government work. From Kerala to Massachusetts, Germany to Brazil, increasingly software specifications talked about document standards.

OOXML - A  Flawed Standard

With the writing on the wall, Microsoft launched its counter offensive. It introduced its standard - what it calls OOXML - as a new open document standard, using ECMA as its vehicle. Its aim was not to introduce a real document standard but to make Microsoft's document formats  a new global standard. The problem is that as everybody knows, Microsoft products have grown completely haphazardly and therefore these formats are almost impossible to describe and even more difficult to implement. A test done by some software experts have shown that not only Word 2007 does not conform fully to the OOXML standard being pushed by Microsoft, it throws 122,000 exceptions! If this is the plight of Microsoft, what chance for any other vendor to have a compliant implementation of OOXML?

So why is Microsoft interested in this new standard? The answer is that it wants to keep other products out of its market - it is saying loud and clear that only Microsoft can read older Microsoft documents. So if anybody chooses any other software vendor and any other format, they will not be able to read Microsoft legacy documents properly. And if others are claiming to be compatible to a global standard, OOXML is what Microsoft will conform to in the future, allowing it to bid for government contracts. In any case, nobody but Microsoft is going to implement this standard, so if it is not fully compatible to the standard, it will be known only to a few people, that too mostly within Microsoft. It is pure marketing ploy - pretend your products conforms to some global standard and threaten others with losing their data if they switch to any other product.

Incidentally, this reading of Microsoft's legacy documents is a hoax. Even Microsoft's own software makes a mess of formatting when reading older documents. And Open Office, freely downloadable from the net and developed by the Free Software community is not only much more robust but also seems to do a better job of reading older documents than Microsoft's own products. Interestingly enough,  Open Office reads Microsoft formats, and therefore can read all Microsoft documents, but Microsoft does not read any document format other than its own. Even though Open Document Format is an ISO standard, Microsoft will not read any document created with this format.  Incidentally, Open Office runs under Linux as well as Windows, something that MS Office does not.

So we come to the next question, if Microsoft products are not essential for either legacy document or reading what other people are doing, why should government accept for its e-governance work, a flawed standard such as OOXML, brought in through shenanigans which India is protesting against? Specially, when Open Office is free and Microsoft Office Suite is expensive? Why should the Indian people's money be handed over to Microsoft when we can get software freely by switching to Open Office? Considering that for any income that Microsoft gets in this country, it does not even pay any tax.

Microsoft pays income tax on income generated from its software licensing in all other countries in the world but does not contribute even a single penny to the Indian exchequer in terms of income tax! The total income from 1999 to 2005 for Microsoft was computed to be about Rs 2,240 crore,  and the income tax department has now levied tax liabilities on this income of about Rs 700 crore. Microsoft is fighting this judgement in the courts claiming that its income is as royalty and not from license fees.

The important issue is why are we paying Rs 2,240 crore to Microsoft when other products are available which do not cost a rupee? And why are we locking away our data within the system of a party that can and does use predatory pricing to allow you access to your own data? And why is the government proposing to aid and abet this extortion by locking up its data in such proprietary format, forcing its citizens to buy Microsoft products to read its own government's data? Why is ministry of information technology  changing its own position on OOXML, rejected by the technical community in India and now proposing to accept it as another standard for e-governance?

February 04, 2008

(Below is an abridged version of the background paper circulated at a workshop, organised to examine the notion of Science Commons – a collaborative, open source framework for promoting advances and innovations in Science. The workshop was jointly organised by Knowledge Commons, Economic Research Foundation and Delhi Science Forum.

Participants at the workshop included Samir Brahmachari (Director General CSIR), Deepak Paintal (Vice Chancellor, Delhi University), Abhijit Sen (Member, Planning Commission), Venkatesh Hariharan, Andrew Lynn, Jaijit Bhattacharya, Kiran Chandra, Satyajit Rath, C P Chandrashekhar, T Jayaraman, B K Keayla, S P Shukla, Jayati Ghosh, Prabir Purkayastha and G Nagarjuna.)

THE speed of innovation and the growth of technology are major drivers of the global economy. While a lot of discussions have taken place on the monopoly created through control over innovations by patenting, much less attention has been focussed in the way innovation takes place and the structures within which innovation is either facilitated or retarded. Does the networked world of today carry new possibilities for alternate structures of creating knowledge and innovation that are currently being retarded? Is it possible to expand the notion of “commons” to help such processes develop?

The patenting system originated in the days of the lone inventor and the need to protect his/her invention. Historically, the lone inventor has given way to large corporate or State funded research laboratories in the early twentieth century. Increasingly, institutions of scientific and technological research, have tended to duplicate the manner of working of global corporations, locating the production of knowledge in a framework premised on private profit. The Bayh-Dole legislation facilitated the adoption of this model by publicly funded science in the US. In India, as elsewhere, a similar trajectory is gaining ground. Interestingly, this is also a time in which alternate models of generating knowledge and innovation have gained ground. The Free Software Movement has shown that networked and open collaborations of “hackers” can produce software of far better quality that what the best of well-heeled corporations working in isolation can manage.

In a forthcoming book, two researchers Bessen and Meurer, have analysed revenues generated from patents as against cost of filing, maintaining and defending patents in courts. The data shows that except in the case of pharmaceuticals, patents generate far more litigation costs than revenue. Worse, the more innovative the company, more was the likelihood of it being sued. The software and business method patents fared the worst, with costs far outstripping the benefits of patenting. Even if we examine, not the broader question of whether societies benefit due to greater innovation, but the very narrow one of whether companies that are innovative, benefit from patenting, the answer is that they do not. Research also show that patents do not promote innovation -- most of the historical data from countries that had different forms of patent protection do not show significantly different rates of innovation.

HISTORICAL LOOK AT PATENTS

Patent as an incentive, gives a monopoly to the inventor for a certain period in lieu of which he/she makes the invention public. In economic terms, this monopoly allows the patent holder to extract rent from all users of the patents: it is the State allowing the patent holder the right to levy a private tax. Therefore, the question arises whether patents (or monopolies) are the best form of providing such incentives?

Even if we accept that material incentives need to be given to the inventors, patent monopolies are not the only form of incentives. Other forms of incentives that have been used with success include, for example, a royalty for the inventor from any producer who wanted to work the patent, but not a monopoly over all reproduction of the invention. Another notion of providing incentives is that of a “prize fund”, from which prizes are awarded by the State for socially useful inventions – the quantum of the prize being proportional to the value of the innovation.

The question is whether the monopoly patent regime has helped in promoting innovation. Let us start with the most celebrated innovation, which in all text books is stated to be one of the key elements of Industrial Revolution: the Steam Engine. James Watt perfected his version of the steam engine for which he secured a patent in 1769. In 1775, using the influence of Mathew Boulton, his rich and influential business partner, he succeeded in getting the Parliament to pass an Act extending his patent till 1800. The major beneficiary of the advances in steam engines would have been the mining industry in Cornwall. Watt spent a large part of his life suing the Cornish miners, to prevent them from making any advances over his design. The firm of Boulton and Watts did not even manufacture steam engines then, they only allowed others to construct the engines based on Watt’s designs for which they claimed huge royalties. If we examine the increased efficiencies of steam engines and plot it against time, we find that after the initial Watts breakthrough, during the period that Watt had monopoly, all further improvements virtually stopped, starting again only after the expiry of his patents. During the period of Watt’s patents the UK added about 750 horsepower of steam engines per year. In the thirty years following Watt’s patents, additional horsepower was added at a rate of more than 4,000 per year. Moreover, the fuel efficiency of steam engines changed little during the period of Watt’s patent; while between 1810 and 1835 it is estimated to have increased by a factor of five.

Interestingly, all those who made further advances, such as Trevithick, did not file patents. Instead, they worked on a collaborative model in which all advances were published in a journal collectively maintained by the mine engineers, called the “Lean’s Engine Reporter”. This journal published best practices as well as all advances that were being made. This was the period that saw the fastest growth of engine efficiency.

There is very little concrete evidence to suggest that patents have consistently promoted innovation in the past. To the contrary, the above case of the steam engine and other instances such as the development of the blast furnace in the US and UK in the 19th century, suggest that collective innovation settings lead to a faster diffusion of technology and more innovation as opposed to the closed, patent based monopolies. It is significant that the advances in the two key elements of the industrial revolution – steam engines and steel – both came out of a non-patented and open sharing environment. The recent advances of Free and Open Source Software is not an anomaly but merely the reflection that an open model of developing knowledge is a faster and surer way to innovation than conferring state mandated monopolies.

SCIENCE AND  OPEN MODELS

The current contours of the scientific discovery are defined by a “competitive” notion of exclusive discovery. They are consistent with a reductionist paradigm where small problems can be examined in isolation. Such models are unlikely to be adequate today. Cooperation in the scientific community on a far wider scale than has been the case so far is critical if major advances are to take place.

Today, the information technology sector has shown that new technologies and methodologies can be developed by cooperative communities. It may be argued that this sector is unique in that the “reproduction costs” of the “artefacts” – the software – are relatively low. However, the question needs to be posed whether it is possible to design such approaches for other areas such as, say, the life sciences? Is it possible to have similar cooperative communities that work together to produce new products? Is it possible to envisage ways by which artefacts can be reproduced and reach the community without high costs of such “reproduction”? Are there spaces to be found in which new, more intimately cooperative modes of scientific enquiry can be initiated? What is needed is to explore new ways of establishing ‘creative commons’, in which new technologies and methodologies are developed by cooperative communities. Some examples of this are given below.

Agribiotechnology

There is little doubt that genetically engineered plants are going to create an enormous impact on agriculture in the future. There are several reasons as to why this is yet to happen. A major reason is that the science behind genetic modification of organisms is still in its infancy. The second and perhaps more important reason is that unlike the Green Revolution that came out of public domain science, the Gene revolution is coming from private domain science. The prospect of agriculture of any country passing into the hands of a few multinational companies is not a reassuring one. It is compounded by the fact that most of the successful biotech seed companies are either chemical companies such as Monsanto, Du Pont etc., while others are pharmaceutical companies – Novartis, Bayer, etc. The track record of both regarding public good has been extremely dubious. The discomfort that people have regarding their countries’ agriculture passing into multinational hands is not unjustified.

Greg Traxler, in his paper for the FAO discusses the rapid increase of transgenic crops in some countries and for specific crops. In 1996, approximately 2.8 million hectares were planted to transgenic crops or genetically modified organisms (GMO) in six countries. Adoption has been rapid in those areas where the crops address important production problems, and by 2003 the global area had risen to 67.7 million hectares in 18 countries. Six countries (the USA, Argentina, Canada, Brazil, China and South Africa), four crops (soybean, cotton, maize and canola) and two traits (herbicide tolerance and insect resistance) account for more than 99 per cent of global transgenic area.

The bulk of ‘innovative technology’ in the agri-biotech sector is currently focussed on making genetically modified crops – a technology that is patent-protected by the MNC sector. An interesting step away from this corporate model of agribiotech development has been the establishment of an ‘open source biology’platform, centred around new microbes useful for making transgenic plants. The most advanced initiative of this kind is the Australia-based CAMBIA/BIOS. This initiative focuses on freeing the basic technological tools of biotech for general use. It promotes a protected commons license for use in this regard. It also operates a web portal BioForge, similar to the SourceForge of the open-source software movement. While the BIOS initiative is not identical to the free-software idea, it appears to be the most developed initiative of this kind so far.

However, such a knowledge commons approach may still depend on the conventional manufacturing sector for delivery of the products – for example, the seeds – to the market. Also, it still involves making transgenic crops, which is a technology replete with implementation difficulties of both the political and the environmental kind.

One alternate possibility that is being discussed globally is to take advantage of the growing ability to sequence the entire genetic sequence of individual organisms at steadily declining expense. The incorporation of such a step in traditional plant breeding for advantageous traits will allow the breeding programmes to overcome some of the major obstacles in creating crop varieties with advantageous traits that breed true so that seeds can be re-used. Such a programme would be of little interest to the for-profit sector, since farmers can reuse seeds. It is a programme that would demand a large-scale cooperative global effort between breeders and scientists. Breeders would need to collect and maintain source varieties and carry out careful breeding. Scientists must, on the other hand, generate new ways of handling and interpreting the large mass of data that sequencing-assisted breeding would yield.

Open Source Drug Discovery

A similar possibility exists in the area of drug discovery. In 1995 the TRIPS agreement introduced an uniform and higher level of Patent protection across the globe. The promise that this would lead to higher levels of innovation remains a mirage. Globally, the number of New Chemical Entities (NCEs) have progressively gone down over the past decade. Further, of NCEs approved for marketing, a very small fraction – less than 3 per cent – constitute a significant advance over prevailing therapies. An overwhelming majority of new products address needs of the wealthy populations in the global North, while the disease burden is largely in the global South. While the industry researches drugs for lifestyle conditions of the affluent – obesity, erectile dysfunction, baldness, etc. – conditions such as Tuberculosis, Kala Azar, Sleeping Sickness, have to make do with decade old therapies. The last drug developed specifically for Tuberculosis, was introduced some three decades back.

Clearly the IPR based model for innovation is just not working. Strong IP protection is encouraging protectionism and is harming the way science is done. Many more Patents are taken out to stop others from working than to protect one’s own research. It is premised on very high costs of development, that are sought to be recovered through high monopoly pricing of products, thereby closing the door for research that targets conditions of the global poor who do not have pockets deep enough to afford the high prices.

Can open-source drug research and development, using principles pioneered by the highly successful free software movement, help revive the industry? An open source model would organise research around researchers across the globe, who draw from a pooled source of information to which they contribute, and to which they pledge to plough back the new developments that accrue. A decade back such a model might have appeared an utopia. Not so today when very powerful tools are available that can create virtual models, that can sequence genetic codes of humans, that can identify potential targets for interventions in the genetic code. It is possible to process genomic information and on a much larger scale, create public databases of genomic information and protein structures, identify promising protein targets, and deliver such compounds for clinical trials. It would be based on a collaborative, transparent process of biomedical development to take on health challenges that big pharmaceutical corporations have neglected in favour of what they perceive as “block-buster drugs”.

Such a model can identify new candidates at a fraction of the cost that Big Pharma claims to spend. It has been argued that the major cost in drug development relates to clinical trials that need to satisfy drug regulatory agencies. Today Big Pharma outsources clinical trials to a dispersed set of Contract Research Organisations. A collaborative open source model could use the same route, with the difference that the entire endeavour – from selection of promising candidates to marketing approval – is organised and overseen by a publicly funded entity or group that promises to place such research in public domain, without insisting on Patent monopolies. It is an idea whose time has come and has the potential to revolutionise the way research is done.