TECH TALK: The Indianised Linux Desktop: The Solution (Part 2)

  1. Integrate with the Mobile Internet: The solution may be built using lag technologies, but its important to look at what the world will be in the coming years. The creation of a pervasive wireless network is perhaps the most important development that is taking place – via cellular technologies like 2.5G (GPRS) and 3G, and Ethernet-based extensions like 802.11b and 802.11a. This helps bridge the information to the user at all times, creating the foundation for the real-time, event-driven enterprise. As a starting point, by integrating with SMS (Short Message Service, available on GSM cellphones), it now becomes possible to deliver emails, alerts, notifications and other information to the user. By making the cellphone can work as an extension of the desktop for time-sensitive information, much greater value is created for the users.
  2. Deliver Software and Content to the LAN: Bandwidth is a big challenge in India. Enterprises will need both information and software on an ongoing basis. While a centralised model with everything on servers on the Internet may work fine, it may be worth exploring alternative models which deliver software and content to a server nearer the enterprise (or on the LAN itself) for onward distribution to the users. At the other extreme, a CD could be delivered periodically with hundreds of megabytes of data which can then be copied on to the server storage. If newspapers can be delivered daily to our homes and offices, surely a CD can be delivered periodically to enterprises. This solution helps bypass the big gap we see in LAN-WAN speeds in countries like India, where anything beyond 64 Kbps connectivity to the Internet is, for most enterprises, still a dream.
  3. Target First-Time Users: The mistake that is made in most of the Linux-as-a-failed-desktop-alternative arguments is that the effort is being made to switch existing Windows users to Linux. Once people get into a comfort zone, the switching costs become very high as they look for identical replacements of everything they are doing. However, since we are targeting first-time users, there is no such barrier. This is like Amul’s Rs 20 Pizza: a new generation of users is being targeted, one which has yet to taste computing.
  4. Subscription Business Model: So, we now have the hardware costing no more than Rs 15,000 (USD 300) and software which is predominantly built out of open source components. One pricing strategy: Rs 2,000 in upfront payment for the software and Rs 1,000 for quarterly updates. Even adding a 15% maintenance fee for the hardware in the second year of operation, the costs come to Rs 18,000 in the first year and Rs 3,250 in the second year, a total of Rs 21,250 over two years. However, instead of selling the computer, one should look at renting it out to the enterprises for Rs 1,500 per month. Over two years, collections total to Rs 36,000. Obviously, some financing will be needed to make this work. A subscription model lowers the entry barrier and makes the enterprise oblivious to the technologies underneath and focusing on using the solution for its business. It also creates a billing relationship which can serve as a platform for additional services.
  5. Push from the Government: The Government can play a positive role in the deployment and build-out of the computing infrastructure. Government offices have huge information which both individuals and enterprises need to file and fetch. By making this information available electronically in multiple languages and making it mandatory for entities to submit information electronically, the government can play the role of a catalyst in this entire process.

With technology, one always yearns for the latest and greatest. But in India and countries like India, cost and not the generation (of technology) is more important. Most of the software we use has not been designed for use in India. The companies who understand the needs best (the local Independent Software Vendors) have little incentive to create since there is a very small domestic market. Yet, this is what needs to be altered. By getting the base system cost down and making it available in multiple languages, it will be possible to build a critical mass which can attract local developers. This can start a positive feedback loop, as uptake of the systems and applications complements each other.

It is time we stopped looking Westward and looked inward to create solutions relevant in the Indian context. To build out computing as a utility, the Indianised Linux Desktop is the first stop.

TECH TALK: The Indianised Linux Desktop: The Solution

The challenge: how can the cost of the desktop be brought down dramatically? The building block for the desktop is undoubtedly Linux, but much more has to be done in order to convert it from an interesting and free software to one which can serve the needs of the mass enterprise market.

Our aim is to create a computing solution which can be adopted by enterprises in India. The objective is not necessarily to target the consumer market. Between the large enterprises (who have the money to buy any solution at any price) and the consumers (who want a lot of goodies for the lowest possible price) is a huge market of small and medium enterprises who want to (and need to) use computers for their core business applications.

Here are 10 components of a solution which can increase by 10x the use of computing among Indian enterprises:

  1. Aggregate Applications: For most users, a collection of 4-5 applications is good enough to take care of all their needs. These applications include: Email, Browser, Instant Messenger, Word Processor, Spreadsheet, and of course a Graphical User Interface (Windowing system). All of these applications are available on Linux in the form of Netscape, Jabber, Star Office and the KDE/GNOME desktop environments. What is needed is that can be integrated together to provide a seamless feel that MS-Windows and MS-Office have.
  2. Run off a CD: All the applications should run off the CD on the desktop – mails, files and user preferences should be stored on the server. By eliminating the hard disk on the desktop, not only does the cost come down but also maintenance becomes much easier since there is just the motherboard to worry about. Users cannot download anything on the desktop, and each system is interchangeable. Every few months, a new CD can be sent out with the newest versions of the applications. Just put the CD in the system, and boot right off it. A new twist to the thin client philosophy!
  3. Add the Hardware: In the market today, it is almost impossible to get a computer for less than Rs 25-30,000. It will have the latest processor at 1+ Ghz. Most companies don’t need this, and this is where costs can be reduced dramatically. Because Linux can run very well on slower processors, one can use lag technologies – processors which are in the 400-500 Mhz range. It is hard to get them now, but motherboards containing them are available if you look. The hardware cost (motherboard, memory, network card) should be no more
    than Rs 10,000 (USD 200). Add another Rs 5,000 for the CD-ROM drive, keyboard, monitor and mouse and there is a fully working system for under Rs 15,000.

  4. Support Local Languages: The applications should be available in English and various Indian languages. This requires work in terms of translations and fonts (and their availability in open source). But this is the only way to take computing to a much greater penetration, and make it a utility. Linux is the perfect base on which to build on. There are various efforts to support Indian languages in India. Most of these have so far been academic in nature. There needs to be a market pull which can provide greater impetus to these efforts. This pull is not going to come from consumers but from enterprises, especially in second-level towns and cities across India and the rural areas.
  5. Create Enterprise Applications: The hardware and software as part of the desktop is only the beginning. What is needed is to create an ecosystem which encourages enterprises to deploy these systems and developers to create products and customised applications on this base. The initial set of applications is good enough as a start, but more specialised applications for the enterprise will be needed. No way can Indian companies afford to pay the prices charged by the likes of SAP, Peoplesoft and Oracle. We need the equivalent of the generics manufactured by Indian pharma companies. We need low-cost, less-featured enterprise applications relevant to our local context. These can run off servers within the enterprise, and be distributed via the Internet. To make this possible, defined interfaces are needed to attract a set of developers. In addition, the enterprise software building blocks should use the Web Services standards (XML, SOAP, UDDI and WSDL). This will make it possible to create larger structures using the blocks like Software Lego.

TECH TALK: The Indianised Linux Desktop: Views on The Linux Desktop (Part 2)

Writing in ComputerWorld, Miguel De Icaza, CTO and co-founder of Ximian, provides an overview of Linux in the corporate market.

With the proven success of Linux-based servers for mission-critical business applications, many have asked when Linux on the corporate desktop will also take off. While Linux distributions and key server technologies such as Apache and Samba power up to a third of corporate Web, database and file servers, widespread adoption of Linux for end-user desktops at the expense of Microsoft Windows has seemed like the promise of a distant future.

Until now.

The penetration of Linux into corporate desktops has already begun as research and development departments, technology companies, retail powerhouses and even national governments turn to Linux and free software for its superiority as a development environment, its easier customization and its lower cost of ownership relative to Windows-based systems. To move beyond its current 2% market share (according to a figure published by IDC in March), though, the Linux desktop must be easier to use and offer the file compatibility and suite of productivity applications that users need while simplifying the management of constantly enhanced open-source software.

Writes Stephen E. Harris, Publisher of

Linux is still seen as a geek OS running servers and networks, and there is no desktop distribution with significant brand recognition among potential users. The market space is huge — for every server there are any number of desktops and work stations — but the current advice for most Windows users is “if it ain’t broke, don’t fix it.” That could quickly change once a robust, inexpensive Linux desktop emerges.

It is often said that Microsoft Windows “owns” the desktop market, but operating systems, unlike detergents and automobiles, are peculiar commodities that are not “consumed” in the ordinary sense, nor do they wear out on their own accord. Now that the Windows OS is ubiquitous, Microsoft is forced to compete against its past success. They seem to have settled on a .NET strategy of software leasing as the best means to churn its customer base and generate a continuous revenue stream. This entails a new licensing model which strips away software ownership, and client control over the upgrade path. Shades of the old IBM hardware monopoly, when customers were obliged to lease their
tabulating machines and computers.

Microsoft will succeed in squeezing more revenues from existing clients, but at the cost of creating resistance in companies that relinquish software ownership and watch their MIS budgets spin out of control. As an open, robust, controllable, alternative that can save thousands, even millions of dollars annually, the Linux OS and desktop will inevitably appear on corporate radar screens.

Writes Bryan Pfaffenberger, Associate Professor of Technology, Culture, and Communication at the University of Virginia:

I predict that Linux’s share of the desktop market will begin to grow, slowly at first but then steadily, just as soon as:

(1) Corporate purchasers and consumers start to feel the pull of Linux. This won’t happen unless Microsoft succeeds in its bid to greatly reduce the incidence of unauthorized software duplication. I am not certain that this will occur.

(2) Ill will toward Microsoft builds to critical mass. I am certain that Microsoft executives are sufficiently out of touch with reality that this one is a safe bet.

If (1) happens (possible), and (2) happens (probable), then Law No. 3 kicks in, push magnifies pull, and the move to the Linux desktop accelerates. And as the phenomenon mounts, all the other so-called indispensable “pull” components–filesystem standardization, a growing pool of great applications, enhanced ease of use, and readily available peripherals–will fall into place. I hope.

Hidden among all these statements are the answers to how Linux can be leveraged in emerging markets like India.

TECH TALK: The Indianised Linux Desktop: The Linux Desktop – and why it has failed

The success of Linux so far has been primarily in the server market, and to a limited but growing extent, in the embedded space. Yet, Linux has been virtually non-existent in the desktop segment. This is where Microsoft products (legal and illegal copies) rule the roost. As long as piracy is easy and people have little qualm doing it, Microsoft products will continue to rule. But assuming that (a) people are not born robbers (b) there is a viable alternative available, Linux will have a chance.

But before we look at how Linux can become that alternative in the Indian context, let us first take a look some views on Linux on the desktop, and why it has not yet taken off.

Writes David Coursey of ZDnet’s AnchorDesk:

Linux will never become common as a desktop operating system, and no amount of believing will change that. It only makes adherents look stupid. Why? Because Linux is too complex, and there isn’t enough money to make it worth someone’s time to build a really great environment for desktop apps. And then software companies would need to build applications, but how large a market is there? Yes, chicken-and-egg, but that stops many things, not just desktop Linux.

Russ Mitchell, writing in Wired (October 2001), argues that Linux should give up on the desktop market and focus on the enterprise computing market (server-side) where it has done very well so far:

Desktop computer users care about what they can do on their machines. They want reliability, simplicity, access to popular software, and the ability to communicate easily with other usersLinux is highly reliable, but unfortunately that’s about all it offers the typical desktop user.

Linux has been on the industry’s radar screen since the mid-’90s, yet the vast majority of applications available for Windows and Mac don’t exist for Linux. Microsoft apps – Word, Excel, PowerPoint – don’t work on Linux because, of course, Bill Gates doesn’t want them to. Nor does Internet Explorer, which controls 87 percent of the browser market.

Adobe Photoshop on Linux? Nope. QuickBooks? No. Eudora email? Uh-uh. Sega games? Nyet. And on and on.

The Linux desktop offers very little that could be considered plug-and-play. Linux drivers, the software that connects a computer with peripherals like printers and CD burners, are in short supply.

Nontechnical users continue to have a hard time installing Linux. Every time a review appears in the press describing how damn hard it is to install Linux, and what few useful apps there are, it hurts Linux’s reputation among top executives who might be considering the OS for enterprise.

Microsoft and Apple ads are everywhere, but no one is funding major marketing campaigns for desktop Linux. No one with any clout is carrying the torch for desktop Linux. Who is Linux’s Bill Gates or Steve Jobs? Not Linus Torvalds.

Dennis Field, who failed in his efforts to get Linux working on a laptop:

Suppose you were looking for a car, and you heard about this great new sports car that got 50 Miles Per Gallon and only cost $5000! But when you went to buy one, you were told it didn’t have any tires, and there was no alternator (so you had to figure out some other way to keep the battery charged) and, oh also, don’t drive it too fast because the brakes don’t work. And if this car ever breaks, then you’ll need to find your own spare parts and try to fix it yourself, because the dealer that sells the cars refuses to work on them. Would you buy that car? More importantly, would you recommend that car to somebody that needs reliable transportation to get to work tomorrow? Does this sound far-fetched? Well, that’s exactly what many Linux vendors are telling the people who buy their softwareMaybe people are buying Windows because it works out of the box (well, mostly works, anyway), and Microsoft at least tries to offer support when it doesn’t work.

Kevin Reichard in LinuxPlanet:

As it stands Linux on the desktop is not an entity that is usable by the average PC user when it comes to accomplishing their daily work. This has nothing to do with the quality of the desktop environment, but has everything to do with how PCs are actually used: end users don’t use the environment, they use applications. And while someday we may evolve to a world where everyone’s work is done via the Web and a Web browser, we’re not there yet. Most computing work is done via third-party software tools independent of the operating system. The lack of usable software tools is really what will doom Linux on the desktop.

There are a lot of half-baked tools out there, and one of the problems in being a Linux user is the need to figure out exactly what tools are usable in their present release and which tools merely show a great amount of promise. Linux on the desktop is weak because it just doesn’t work well on the desktop yet for the work that the average people needs to accomplish. You cannot ask users to compromise when you want them to switch operating systems, and you need to be realistic about how things work. Let’s face it: at the present time there’s nothing under Linux that works as well as Microsoft Office. Period.

TECH TALK: The Indianised Linux Desktop: The Missing Indian Computer

It is quite amazing that not a single PC vendor in India yet sells an “Indian language computer”. In a country of a billion people, the fact that English is pretty much the only language available on the computer explains the poor installed base (about 6 million) and sales of computers (about 1.5-2 million this year). If one looks at China, Japan or Korea, all of them have their own versions of all major software applications. Lack of localisation remains one of the key barriers to the growth of computing in India.

The challenges in creating a mass market for computers and applications in India are many. The high cost of both computers and software makes them the domain of few. The lack of a domestic software products industry leads to little innovation. The still-born Internet boom has not helped much. The result is that in a country which harbours dreams of becoming a global economic superpower, technology still remains expensive, exclusive and an afterthought.

In India, everything is expensive relative to incomes: the cost of technology (US Dollars multipled by anything from 50 to 70), the cost of telecom and travel, and the cost of real estate (in the major metros) and electricity. The only thing cheap: human resources (though this is also starting to change). Add to that the myriad government rules and regulations, and doing business in India becomes a labour of love. Export earnings are much needed but we are doing it (at least in software) at the cost of building a domestic market.

Yet, India has excellent opportunities to create a mark in the international technology arena. Indian enterprises, will over time, get integrated into global supply chains. But the basic challenge remains: how can the cost of technology be brought down significantly to make its adoption more rapid and widespread. Enterprises of all hues and sizes need to adopt infotech and streamline business processes and communications. As the Internet makes companies relook at how they interact and what they do internally (and what they outsource), tech-savvy Indian companies can hope to reap rich dividends.

One of the key elements in the computing value chain is the Desktop. With Microsoft having a global economy with its Windows and Office suite of applications, in India the choices for desktop software are only two: either pay Rs 20,000 (which is not a small amount for most companies) or indulge in piracy (which has been the preferred option). A country does not become great by stealing. So, piracy as a long-term option to greatness is ruled out.

To bring down the cost of computing on the desktop, one answer is to consider an Indianised Linux-based solution. Worldwide, the Linux desktop is a rarity, with a market share of less than 2%. But in our quest for “disruptive” solutions, Linux along with a few other innovations may just be the answer that Indian enterprises may be looking for.

TECH TALK: One Year of Tech Talk: Key Themes (Part 3)

8. India is an excellent starting point for building out the SME Tech Utility. There are 4 billion people like us and millions of companies like the ones we will find in India across the world in Asia, Eastern Europe, Latin America and Africa. One has to think along the lines of creating mass market solutions.

  • India’s Century: There are six building blocks which can serve as the platform for making the 21st century belong to India: Infrastructure, Open Economy, Technology and the Internet, Education, Global and Entrepreneurial Mindset, Leadership and Vision (May 8, 2001)

  • A Technology Agenda for India: What does it take for India to be a leader in the development and deployment of new technologies worldwide? In this context, it is imperative for India – corporates, educational and research institutions, citizens and government – to understand the impact of technology, and put in place an agenda to ensure that we can leverage it appropriately. (April 9, 2001)

  • The Indian Internet: The Internet revolution in India has almost been a still-born, and this has led to quite some disillusionment all around. How can this change? The solution lies in the leading Internet companies becoming profitable, so that investment can once again start coming into this sector. (March 19, 2001)

  • Mass Market Internet: The Internet in India is still too elitist and growth still too slow. How can we change this? How can we get 10 times the current growth rate? How can the Internet impact the lives of 100 million Indians in the next 3 years? How can the Internet remove pain from our lives, how can it make a difference to us daily? How can the Internet become a utility in our lives — just like electricity and the phone? (Nov 13, 2000)

  • A Tale of 4 Anniversaries: India batting 54, Reforms hit wicket 10, PC retired hurt 20, Internet at the non-striker’s end on 6. (August 14, 2001)

  • 10 Trends for the Indian Internet 2001: My choices: Entrepreneurship, Bandwidth, eBusiness, Venture Capital, IT-enabled Services, Consolidation, Education, Mass Market, Wireless and Government. (December 5, 2000)

  • The New War: A View from India (September 17, 2001)

9. The opportunities for Indian companies lie in building solutions to address these markets.

  • The Indian Software Factory: At the bottom of this pyramid are over 20 million small and medium enterprises in developing markets like India. How can technology, and especially software, fulfill their needs and aspirations? At the heart of the discussion is the question asked by Prahalad and Hart: How can low cost, good quality, sustainability, and profitability be combined? (July 3, 2001)

  • Building an Innovation Valley in India: India Inc’s mission should be to help Indian entrepreneurs transform their technical expertise and ideas into successful global enterprises that build cutting – edge products and services, companies that are leading the revolution. (Nov 28, 2000)

10. What is needed to accomplish this is Entrepreneurial Thinking.

  • Entrepreneurship: Entrepreneurship is a lot more than just getting an idea and starting a company. It is the tougher choice, not the easier one. But this is one journey where the joy is as much in the ride as in reaching the destination. (March 12, 2001)

  • Entrepreneurial Learnings: Entrepreneurs do not start to create failures. Yet, only 1 in 100 startups succeeds. If one were to ask entrepreneurs when the start about their chances of success, most would give themselves greater than even odds, even as high as a 60-70% chance of success. The reality though, as we know, is very different. While no one can predict success or failure, it is important to do things right to begin with. (July 16, 2001)

  • Life as an Entrepreneur an entrepreneur is not necessarily a risk taker, but a risk reducer. Each day at work, the entrepreneur seeks to make decisions to increase the longevity of the business and diminish the risk. And yet, herein lies a paradox. For an entrepreneur, lack of growth (and uncertainty) is like death. When things start becoming too predictable, the challenge ebbs away. It is then time for a new dream, a new vision. (October 15, 2001)

  • Leadership Lessons from Lagaan: While it is not often that movies offer more than entertainment, ever so often there comes a film which can make a deep impact and teach us a few things. Lagaan is about Bhuvan and his team, who, against all odds, fight for pride, land and country – and win. It is how one single person with a passion can make a difference. It is about the triumph of human spirit, the Indian spirit. (August 6, 2001)

  • Points to Ponder: A collection of interesting thoughts from various people. Many of things stated seem obvious, but it is good to remind ourselves of these every so often, and reflect and make part of our daily life.

TECH TALK: One Year of Tech Talk: Key Themes (Part 2)

4. The key enabler for tomorrow’s enterprise will be Software, which is already becoming the oxygen of eBusiness. Instead of thinking of software as a monolith system, it will need to become like Lego which can be assembled together to form more complex systems.

  • Web Services: the real utility of the Internet is now beginning to dawn. It goes by the moniker of “Web Services”. It is about business applications communicating with each other automatically through the Internet. It is about information sharing across enterprises. The promise: seamless communication across the extended enterprise, reduction in working capital and inventory costs with just-in-time information, and leverage of the legacy software that companies have installed. Is the Web Services revolution for real? (October 29, 2001)

  • Enterprise Software: What would happen to the market if the enterprise software suite cost a fraction of the USD 500,000 average that most of the “best-of-breed” enterprise software solutions cost? There are perhaps 25 million companies in the world. Can the market be expanded dramatically if the price were dropped dramatically? Can the software entry point be used to offer additional services to these enterprises, thus dramatically expanding the market? (March 28, 2001)

  • Messaging: Email is at the core of what we all do. More than browsing the Internet, we spend time communicating. We are inherently dependent on others to get our work done, so this means that we need to interact with others, we need to communicate. Email is helping us be more productive, communicate with an ever greater number of people and keep an ever increasing number of conversational threads on at any point of time. This modern-day wonder is keeping us more glued to our desks and PCs than any other application. (Feb 12, 2001)

  • Email: During the last 5-6 years, email has become a “mission-critical” business tool. Executives spend at least 1-2 hours every day in email. In many ways, email has become the “front-end”, the user interface to the Internet. But email still has certain problems. (June 26, 2001)

  • Collaboration: We are all part of groups. We need to work together with other people to accomplish our tasks. These people may be within our company or outside. This “collaborative” facet of work has not changed since time immemorial. What has changed are the tools and technologies we use to collaborate. (July 9, 2001)

  • News Refinery: We spend a lot of our time reading everyday – newspapers, magazines, websites. We are in search of news and information. This “information absorption” activity is perhaps the largest consumer of our time after email. It is therefore important to be able to get the news and information without missing anything and from all over, and perhaps faster. (May 21, 2001)

  • Gaming: As computing power on the desktop increases and connectivity to the Internet becomes better and faster (through wireless, broadband and devices), Gaming has the potential to be the next killer app. (April 2, 2001)

5. The world does not just consist of the Big Companies. There are millions of small and medium enterprises (SMEs), the majority of which are in emerging markets like India. The SMEs are the corporate of the world, the bottom of the enterprise pyramid. They use little technology today, but also have suppliers, customers and employees. They also need software.

  • Software and SMEs: Computers and the Internet have been the biggest revolutions in the past 20 years. While advances in semiconductors and networking have powered these changes, it is software that has been the invisible factor which has been common to both. Software has become the new physical infrastructure of the information age. (January 29, 2001)

6. Starting with the software needs of SMEs offers a great opportunity to think innovatively and disrupt the Enterprise Software market in the world. The objective needs to be to “delight” SMEs.

  • Innovation: What’s common to Napster, Lagaan, Amul’s Rs 20 pizza and Google? (July 31, 2001)

  • Disruptive Technologies: Disruptive Technologies bring to market a fundamentally different value proposition than what has been previously available, based on a different set of product/service attributes. This leads to the emergence of new products, the re-definition of an industry’s value network and a shake-up of its competitive structure. (August 20, 2001)

7. The solution lies in thinking of an SME Tech Utility, which makes software like electricity and water, available to all and at a low price-point.

  • SME Tech Utility: Create a Technology Utility company, to offer a one-stop solution for SMEs globally covering business and technology needs in four areas: Computing (Management of SMEs local IT infrastructure – the remotely managed desktop), Communications (A Local Server on the SMEs LAN with “intermittent connectivity” to the Internet), Software (An integrated e-business suite of applications covering ERP, CRM and SCM), and Services (Outsourced IT-enabled services from an Indian base in core business areas) – March 1, 2001

  • Alt.Software: Software powers business and is at the heart of enterprises more than at any time before. The investments in technology over the past decade have helped managers get better insights into their businesses, and get them faster. The Internet has only helped accelerate this trend. eBusiness, the new mantra, is more about software than anything else. Why then is legal Software so expensive? (October 1, 2001)

  • Tech@Enterprise: Creating a Virtuous Cycle: Even as the twin drivers of the Mobile Internet and Web Services promise to create a ubiquitous, utility-like and a much more business-oriented Internet II, the key question that needs to be addressed is: how can the penetration of technology in enterprises in countries like India be increased? (November 12, 2001)

TECH TALK: One Year of Tech Talk: Key Themes

As we look back over the past year, I thought it will be a good idea to summarise some of the key ideas that I have talked about: India and the Internet, Enterprise Software (especially for small and medium enterprises in emerging markets like India) and Entrepreneurship. There are, in my view, inter-linked. India’s future lies with entrepreneurs who can leverage the Internet and Software to carve out a niche in global markets. This means thinking globally (and beyond IT services), building products, creating brands.

I have classified the articles I have written over the past year into 10 themes.

1. Consider the trends in computers, communications and software as an aggregate. Together, they are helping define the next Internet – a Global Technology Grid, providing pervasive connectivity and a real-time infrastructure.

  • The New Internet: A new Internet is being built – an Internet that is helping glue together great progress in computing, communications and software. (May 28, 2001)

  • Envisioning the Future: IT and the Internet, Low-cost Computing and Communications, Business Process Re-engineering, Software as a Utility, Services Shift, Emerging Markets (April 16, 2001)

  • Technology Themes: To think through the impact going ahead, it is useful to look at some themes (Embedded Internet, Innovation, eBusiness, Services) brought about by these technologies. No single step may be revolutionary, but taken together their implications are. (March 5, 2001)

  • Opportunities in Tomorrow’s World: “The Intelligent Enterprise” and “Connected People” are the two themes for the world of tomorrow. (November 29, 2000)

  • TechnoWonders of the Modern World: My choices: Computer, Internet, Email, Cellphone, Cable TV, Fibre Optics and Streaming. (December 20, 2000)

  • Changing Times: Life has changed a lot in India in the past decade. Perhaps the biggest changes have been due to the compression of time and space – “the death of distance”. The speed of communications, the velocity of life, the frequency of interactions – all have gone up. At the heart of this has been the increasing penetration of technology and the opening up of the Indian economy. (July 23, 2001)

2. The Real Internet Revolution lies ahead. This one is going to be focused not as much on what we can do as consumers, but the efficiencies enterprises can bring out in interactions between them. The real benefit on the Internet in being able to cut communications and transactions costs will now be evident.

  • The Intelligent Enterprise: Integrating ERP, SCM and EIP: The Internet can make a difference to an enterprise and its communities – customers, employees and suppliers. (January 3, 2001)

  • Business to e-Business: 10 Transformations: Manufacturing Economy to Information Economy, Mass Production to Mass Customisation, Vertical Integration to Virtual Integration, Internal Teams to Outsourcing, Conversation to Conference, Request-Reply to Publish-Subscribe, Client-Server to Peer-to-Peer, Going to Work to Doing Work, Hierarchy to Business Processes, Products to Services (January 15, 2001)

  • Harnessing Information: Information naturally resides in silos. In memories, in notebooks, in writing pads, in diaries, in computer files, in email folders, in the RAM of cellphones. Information, like dust particles, naturally scatters. And just like it is impossible to aggregate the dust particles again, so too it is with information. We may be in the Information Economy, but it is still the Dark Ages. In both our personal and corporate lives, the amount of information around us has increased exponentially. What can be done to manage and harness information better? (October 22, 2001)

3. The focus now is on the Holy Grail of business: the Real-Time Enterprise. This enterprise is event-driven, information-intensive and extended to include suppliers, customers and employees. Data will need to be entered just once, and it will be available across the extended enterprise immediately.

  • The Intelligent, Real-Time Enterprise: The New Enterprise thus needs to be both electronic and extended: electronic, because it has to move information in the fastest possible manner among its employees, suppliers, partners and customers; and extended, because it has to work with the other enterprises as though they were one, single enterprise. (June 12, 2001)

TECH TALK: One Year of Tech Talk: Introspection (Part 2)

For me personally, it has been a wonderful experience writing these columns (and sometimes reading the older ones to see the evolution in thought). Writing has helped me think more clearly about the impact of technology and especially the Internet. I have seen the Web from the frontlines right from its early days, since the launch of IndiaWorld in March 1995, and been part of its roller-coaster ride.

The need to be continually aware of what’s happening in the world of technology, especially infotech, has definitely made me a voracious reader. But that’s only part of the story. The other element has been to ensure that I can write for others to read and understand, which at times is the harder part! I used to do a lot more of this earlier in 1995 (check for a complete list), but then had stopped writing because I found I was repeating myself! Hopefully, that hasn’t yet happened here.

Perhaps the most important side-effect for me has been the development of a view of the future and how technology can make a difference. My knowledge of the world outside the consumer Internet and portals was very limited when I started writing in November 2000. But over time, there has been a growing realisation (as has happened for many of the entrepreneurs worldwide) that the Internet is indeed a fundamental revolution and will take many years to play out (we are at the “end of the beginning”), and the impact on businesses will be substantially greater in the years to come.

Living in Mumbai, my perspective is definitely somewhat different than what it is sitting in Silicon Valley. My conviction is that the Internet and Software can make a big difference for both consumers and enterprises in emerging markets like India. When I see the inefficiencies in businesses (especially the small and medium enterprises), there is little doubt that there are huge opportunities to leverage the latest technologies to (a) make businesses much more competitive, and (b) raise the standards of living of the mass market.

The solution lies not in replicating ideas from the US, Europe or Japan but thinking deeply about what the real impact of the new technologies are and applying them to the problems in the developing countries. Our problems are unique, and so need to be our solutions. For example, Microsoft’s MS-Windows and MS-Office at Rs 20,000 per computer is not going to take computing to the masses. Neither will software piracy. The solution lies in thinking entrepreneurially about how each of us can make a difference. There is a great opportunity for us in India to create innovative, disruptive technologies, make them work in India first, and then take them to other countries and other markets like us. There are 4 billion like us in the world. These are the markets we need to go after. At its heart, that is what Tech Talk is about.

TECH TALK: One Year of Tech Talk: Introspection

Last week, Tech Talk completed one year. When I began writing last November, I started by getting the first 15 columns (3 weeks of matter) ready. Back then, I wasn’t sure I had much to say beyond a few weeks! But then, as time has passed, the joy of writing and sharing my thoughts has ensured that not a single weekday has passed without an update in the past year. Yes, some weeks, it has been harder to think of what to write, but the thoughts and ideas have always come.

At the start, I used to think of Tech Talk as a daily column, somewhat on the lines of Red Herring’s Catch of the Day. But then I realised that not being a journalist and the pressures of running a company make it almost impossible to sit and write on a daily basis. So, I’ve evolved a mechanism where I take a theme and then cover that in 1-2 weeks. Most of the week’s writing is done in one or two sittings on a Sunday morning, so there is continuity in the flow. It takes about 30-40 minutes to write one day’s column, so the week’s writing takes on an average about 3 hours. (And for those weeks when I find it almost impossible to write, I’ve fallen back on a collection of thoughts from others in the two “Points to Ponder” sections – only realising later that abstracting from what others have said takes as much time if not more than writing one’s own!)

I have also received many emails from you all, most of which have been published on Saturdays in the Readers’ Comments section.

This interaction has also fueled some of the thinking which have become columns. Thanks to everyone who has written in and taken the time to share some of your thinking.

The writings may not be as journalistically refined as they ought to have been, but they are the writings of a “technopreneur” – one who believes strongly that a mix of vision, optimism and entrepreneurial thinking is what the world needs. Even though internationally we are going through a technology slump, I cannot think of a better time to think about tomorrow’s world and build companies – the challenges and opportunities are immense. This is what I have tried to convey though my writings and the collection of quotes from various others (who say some of the things much better than I could have!)

The past year has been a difficult period for many: as companies have slowed down on technology investments, many tech companies and jobs have disappeared and venture capital for new ideas is almost non-existent. This, though, in no way diminishes the importance of technology in the years to come. If anything, the push is towards greater (and now, smarter) use of technology in companies.

TECH TALK: Tech@Enterprise: Creating a Virtuous Cycle: The Right Mix (Part 2)

The second building block is to create a Web Services API (application programming interface), so that other developers can build Lego-like enterprise software applications on top of the operating system. By building to a standardised base platform, the smaller software companies in India can use their domain expertise to fabricate components which have reuse value and thus can generate additional revenue for them also. This creates the hardware-software mix necessary to build a positive feedback cycle.

One aspect to be kept in mind for the applications in the notion of multi-modal access. Traditionally, information has been fetched through the same interface that it has been updated. Going ahead, information may been updated and accessed through any interface – web browser, email, instant message or an SMS via a cellphone. This portends a much bigger change than may seem apparent – it means that information, alerts, notifications can be sent to a user anytime and anywhere. This is, in the words of Vivek Ranadive, “the power of now”. This is the key to the creation of the real-time enterprise.

The third building block for the enterprise solution is a software subscription business model. Instead of paying high upfront costs for the purchase of software, enterprises should be able rent it for monthly payments.

This takes the trend in software to its logical conclusion and can help put Indian companies at the forefront of the new emerging business (in much the same way NTT Docomo leapfrogged over much of the world with its i-mode service). Software subscriptions are the only way software can be made available and affordable to the mass market of small and medium enterprises.

The fourth building block is the NTT Docomo-like Service Operator, who as the aggregator and the intermediary can pull all the players together and build this into a virtuous value chain. The enterprises servers are like the handsets (or the game consoles) and the software vendors are like the content providers (or the third party game developers. The service operator builds the relationships with the SME, with a key aspect of this being the ability to collect monthly payments for the services. The operator is the entity which needs to define the Web Services APIs. This is where the control, lock-in and long-term value lies for the operator.

There are over 25 million SMEs in the world. More than half of the world’s workforce is employed in these enterprises. SMEs are the real engine of growth for the world economy. By learning from some of the consumer success stories and envisioning the changes happening in the world of computing, communications and software, it is possible to create a profitable business meeting the evolving needs of the world’s enterprises and building the next-generation powerhouse: the SME Tech Utility.

TECH TALK: Tech@Enterprise: Creating a Virtuous Cycle: The Right Mix

The consumer market has seen much more innovation in business models than has the enterprise market in recent times. The video game and the cellphone industries have built up large audiences by using a “loss leader” (the hardware) to build up content and services business that become profitable over time. The Internet portals too tried something similar, but other than AOL which established a strong franchise thanks to its billing relationship, the others have not yet proved that is possible to create a positive feedback ecosystem.

How can we create just such a virtuous cycle for enlarging the enterprise technology market? First, let us re-state the objectives. The aim is to increase dramatically the adoption of technology (hardware and software) by enterprises. This means making technology much more affordable and also meaningful to enterprises. The primary target market would be the small and medium enterprises (SMEs) in emerging markets like India. The goal is to make these SMEs globally competitive, because competition is no longer limited by geography. By using technology appropriately, these enterprises should be able to cut costs, increase sales, and build closer relationships with their suppliers, customers and employees. SMEs, which have been traditionally been islands of information, need to be bridged into the world trade (and information) networks.

Innovation is needed in building out services for the SMEs, who invariably get caught between the two other markets which are much easier to reach: consumers and large enterprises. Mass marketing helps reach consumers while direct marketing gets companies into the limited set of the large enterprises. SMEs are much harder to reach so it’s a market few have been able to tap successfully so far. By using a mix of a game console-like subsidised server to create a large installed base, by leveraging independent software vendors to create the valuable Lego-like software blocks and distributing these over the Internet, and by pricing the software and content services cheaply, it should be possible to create a mix which addresses the technology needs of SMEs.

The first building block of the enterprise solution is a Low-Cost Enterprise Server. This server is akin to the video game console and the i-mode handset. It is the platform on which services will be offered. So it is important to build up a critical mass of these to be able to attract developers to build on this platform. The Server is the Trojan Horse on the Enterprise LAN: it is the entry point for offering services to the enterprise. In emerging markets like India which are still bandwidth-challenged, the ASP model of offering services from the Net will not go too far – services will need to run off the LAN, and this is where the Enterprise Server becomes the anchor.

The Enterprise Server cost can be reduced in many ways. Firstly, one does not have to use the latest hardware – one generation lag technology will do just fine. In fact, one could look at the desktop systems as servers since there is no real need for the huge processing power that CPUs today offer. Secondly, one could even consider using systems which have been disposed off by the developed countries (where regular upgrades / replacements) are the norm. Secondly, by using open-source software like Linux, Apache, Mozilla and StarOffice, licencing fees can be eliminated. Finally, the system can be sold at a discount to speed up adoption. The cost of a headless server (without monitor, keyboard and mouse) should be brought down to about Rs 10-12,000 (USD 200-250). A leasing can ensure a payment of no more than Rs 1,000 (USD 20) a month for the system.

TECH TALK: Tech@Enterprise: Creating a Virtuous Cycle: i-Mode’s Mobile Internet

On the face of it, the value proposition of cellphones is quite straight forward: talk while you move. For the mobility and convenience, one is prepared to pay more and live with voice quality which would be downright unacceptable on a fixed line phone. Even this does not explain the tens of billions of dollars plonked down by telecom companies (especially in Europe) for the privilege of building next-generation (3G) wireless networks. All of them have hopes of doing to their markets what NTT Docomo has done in Japan with i-mode.

Introduced in February 1999, i-mode has over 25 million subscribers, one-fifth Japan’s population. It is still growing by about 1 million a month. i-mode is the world’s only successful wireless Internet service – and the Internet isn’t even mentioned in the ads for the service! Writes Frank Rose in Wired (September 2001):

Like most consumer success stories, i-mode is geared to the people who use it. Handsets are manufactured by name-brand consumer-electronics outfits like Sony and Panasonic, but DoCoMo subsidizes the phones heavily to keep the price low: A model that might ordinarily cost $600 retails for less than $350. Because DoCoMo’s wireless network is packet-switched, users are charged only for the number of data packets they send back and forth, not for the amount of time they’re connected, as on most networks outside Japan.

DoCoMo does not itself supply any of the services available on i-mode. But it’s not a dumb pipe, because in addition to the technological infrastructure, it provides the billing system that enables its partners to make money, and the marketing to sell the service to consumers.

At the heart of all this is a paradox: i-mode depends on outside providers for everything from handsets to content, yet it’s managed so carefully that nothing is left to chance.

i-mode’s success comes from a combination of many factors: technology selection (use of an HTML variation for the content), its “win-win” business model allowing big companies as well as small entrepreneurs to thrive in the services business, the billing relationship with its customers, marketing (focus is on the value provided, and not the underlying technology).

i-mode has created the virtuous cycle that Microsoft’s Xbox hopes to create: the user base now drives new content providers which drives more users. In all this, NTT Docomo has placed itself at the centre of the value chain. The handset operators make their money, but don’t get to build a brand; the content providers can get “official status” and better placement on the screens in return for a 9% fee; and consumers get almost infinite variety due to the use of open standards like HTML and Java.

Concludes Rose, “i-mode is a complex ecosystem – a self-sustaining world in which hundreds of companies feed off one another for their mutual benefit. Like water, sunlight, and soil, the elements that make up this world are everywhere. The trick, as anyone who’s ever played God can tell you, is getting the mix right.”

Getting the mix right is what we need to if we are to get a 10x increase in technology utilisation in enterprises in emerging markets like India.

TECH TALK: Tech@Enterprise: Creating a Virtuous Cycle: Gaming Consoles

On the face of it, gaming consoles and next-generation wireless services may have little to do with increasing the penetration of technology in countries like India. But, as we will see shortly, they make an interesting proposition to customers – and have been hugely successful.

The video industry (including hardware and games) was worth USD 15-20 billion in 2000. The market size has tripled over the last decade. The worldwide installed base of consoles is about 120-140 million, which means that there is still plenty of room of growth. (By comparison, there are about 500 million PCs and 1 billion cellphones). The video game console is also seen as a Trojan horse: at the centre of a digital (and Internet-enabled) home entertainment system.

What gaming consoles like Sony’s Playstation, Nintendo’s Gamecube and the soon-to-be-launched Microsoft Xbox do is offer the base product at below cost-price, hoping to recover the money over time through sales of video games and royalties (from other game makers). For them to succeed, they have to build a virtuous cycle: first build an installed base of consoles, thus attracting the best game developers (the software), thus hoping to lead to an increasing base of users. The hardware (the game console) is the razor, while the video games are the blades.

Take Microsoft Xbox, for instance. The Xbox is as powerful a machine you can get in the home (an Intel 733 Mhz Pentium III processor, 64 MB memory, a 250-Mhz graphics processor, 256 audio channels, a DVD, a hard disk and a high-speed connection to the Internet). The Xbox is being sold at USD 299, which means Microsoft stands to lose about USD 125 on every box. In this market, Microsoft is the upstart, fighting Sony and Nintendo who have years of experience and an installed base running into tens of millions.

What makes Microsoft prepared to lose hundreds of millions of dollars in selling consoles (even without taking the account the USD 500 million marketing blitz)? The answer lies in understanding the business model of the video games industry. Writes Henry Blodget of Merrill Lynch (March 6, 2001):

The video game business has a virtuous cycle that is similar to the one in Microsoft’s core desktop O/S software business. An analogy can be drawn between the two business models:

Initial sales of operating system software (video game consoles) create an installed base that attracts third party application (game) developers. The availability of these third party applications (games) increases the value of the operating system (console), driving further sales of the operating system (console) and further growth in the installed base, which attracts even more third party application (game) developers, and so on.

More simply, the virtuous cycle for video games can be stated as follows: Xbox sales drives growth in the Xbox installed base, which attracts more 3rd-party game developers to the Xbox platform, which increases the number of 3 rd party games for Xbox, which makes Xbox more attractive to consumers, which leads to more Xbox sales, and so on.

Before we look at how ideas from the video games business can be applied to enterprise technology, we will take a look at the approach used by NTT Docomo’s i-mode to build an incredibly successful mobile Internet business in Japan.

TECH TALK: Tech@Enterprise: Creating a Virtuous Cycle: Realities of Technology Adoption in Enterprises

This year in India, about 2 million computers will be sold. My estimate is that enterprises will buy about two-thirds of these. The aggregate installed base of computers in India is about 6-7 million. This is a tiny figure for a country which wants to play an important role in the world’s IT future. The cost of both hardware and software are inhibitors in the growth of computing in India. Unfortunately, India has little control on prices in both segments. The real losers in this are Indian enterprises for whom the cost of technology limits its adoption.

At the same time, the Internet is driving connectedness to new levels. Companies are realizing that the Internet can be a crucial ally in the elusive search for greater efficiencies. The thrust is not as much on revolutionary new business models but on seeing how the Internet can enhance existing ways of doing business.

The basic premise of the Internet is its ability to inexpensively connect customers, suppliers and employees, and enable collaboration. By automating processes, companies can reduce costs and make customers happier. But to make the Internet truly effective, it is necessary for access to computers to be available to more people within the enterprise.

Falling communication costs combined with the need to cut overhead costs and pressure to focus only in those areas where value addition is maximised is going to increase outsourcing, especially to lower-cost countries like India. Companies in India and the rest of Asia are thus no longer working in isolation. They are becoming part of supply chains.

While in other Asian countries the focus has been on outsourcing manufacturing to companies in the region, in India, the emerging focus area is on outsourcing services (software, support and other back-office functions). In both cases, the common thread is for companies to be connected to each other and to partners internationally.

For Indian enterprises, adopting technology and permeating it across the enterprise is going to become increasingly critical. Information is going to be an important aspect of business. Today, information – and hence decision-making – is highly concentrated in the hands of a few. But the spread of computing and communications can help diffuse this to the people on the front-line, who may better-placed to make decisions in fast-changing environments.

Even as the twin drivers of the Mobile Internet and Web Services promise to create a ubiquitous, utility-like and a much more business-oriented Internet II, the key question that needs to be addressed is: how can the penetration of technology in enterprises in countries like India be increased?

One possible answer lies in looking at two consumer devices which have achieved huge penetration in the past few years.

TECH TALK: Web Services: Software Lego

So, are Web Services the Next Big Thing or the Next Great Hype? A practical view on Web Services comes from Wylie Wong, writing in

The Web services concept deserves examination if only because its most vocal evangelist is Microsoft, whose marketing prowess alone can turn an obscure idea into an entire industry. Although the notion of Web services has been bandied about for years, the software giant has given the idea new popularity with its grand .Net strategy, which is taking its first steps in the marketplace with the release of the Windows XP operating system. Microsoft is selling software that companies can use to build services while also offering a set of hosted services to businesses and consumers for a fee.

Microsoft, of course, is not alone. Many of the leading players in the computer industry have all been pushing their visions of Web Services. So, like all revolutions, expect a lot of hype, but this one has a lot of hope also built in.

My answer to the hype-hope question: Web Services are close to being the next wave in the ongoing revolution in computing, and especially software. The Internet helped ease the process of communications between people and computers. What Web Services are doing is building on the Internet to take connectivity between applications, business processes and enterprises to the next level. Just as the Web was built on standards like HTTP and HTML which sparked off rapid adoption worldwide, Web Services too have the standards in place to enable quick uptake. This time, though the change is going to be less visible to us from the outside as much of the action will take place within and across organisations.

An illustration on how things will change comes from John Hagel III and John Seely Brown (Harvard Business Review, October 2001):

Loan processing is a complex procedure requiring at least six steps (data gathering about an applicant, validation of data, credit scoring, risk analysis and pricing, underwriting and closing) and involving interactions with a number of other institutions (checking an applicant’s credit rating, verifying investment and loan balances, and so on). With a traditional IT architecture, the process is usually supported by one, very complicated application maintained by an individual bank; like a Swiss Army Knife, the integrated application does a lot of things, but it may not do any of them particularly well.

With the Web services architecture, loan processing becomes much more flexible, automated and efficient. Leased lines are replaced with the Internet, and open standards and protocols take the place of proprietary technologies. As a result, the bank can connect automatically with the most appropriate institution for each transaction, speeding up the entire process and reducing the need for manual work. And rather than maintain its own integrated loan-processing system, the bank can take a modular approach, using specialized Web services provided by an array of providers.

For software developers, the change will be as dramatic as, in the words of Kevin Werbach, “the Web changed publishing and information access.The idea is the atomization and re-integration of software, moving from monolithic whole to piece-parts that can be pulled apart or put together as needed.Developers like Web services because they can make the process of creating software faster and simpler. They like the idea of re-using components, as well as the promise of reducing thorny integration and scaling challenges. Furthermore, code written by an independent developer may now take advantage of – or be called by – a huge array of existing software.”

Many of us grew up building homes, cars and airplanes out of Lego blocks. Little did we know then that one day, developing software would be quite the same!


TECH TALK: Web Services: UDDI and WSDL

UDDI (Universal Description, Discovery and Integration) is a protocol for finding Web Services in directories. It was developed by IBM, Microsoft and Ariba. Writing in, Yasser Shohoud puts this in context:

As more businesses expose their business functions as Web services, you will need a way to find those businesses and the services they expose. For example, your company may want to use the Web to purchase office supplies electronically. To do this, you need a directory of businesses that you can use to search for all companies that sell office supplies. Once you find the list of companies, you need to find out what Web services each company provides.

An HP white paper on UDDI:

UDDI is a specification that defines a service registry of available Web services, serving as a global electronic yellow pages. UDDI enables a company to publish a description of available services to the registry, announcing itself as a service provider. Service users can send requests conforming to the UDDI schema as SOAP messages to the service registry to discover a provider for services.

WSDL (Web Services Description Language) helps in describing the Web Services interfaces and functions. Thus, companies can use WSDL to gives details of services which would be listed in the UDDI directories. Like UDDI, it too was developed by IBM, Microsoft and Ariba. The specification elaborates on WSDL:

WSDL is an XML format for describing network services as a set of endpoints operating on messages containing either document-oriented or procedure-oriented information. The operations and messages are described abstractly, and then bound to a concrete network protocol and message format to define an endpoint. Related concrete endpoints are combined into abstract endpoints (services). WSDL is extensible to allow description of endpoints and their messages regardless of what message formats or network protocols are used to communicate.

How does all this fit together? Here is an example (from an IBM white paper) which provides an excerpt from a WSDL document that is part of the definition of a Stock Quote Web Service. It defines (in XML) a method called getQuote, with the associated SOAP information that enables a piece of code to find the service, invoke a method, and process the response.

< binding name="StockQuoteServiceBinding" type="StockQuoteServiceType" >
< soap:binding style="rpc" transport=""/ >
< operation name="getQuote" >
< soap:operation soapAction=""/ >
< input >
< soap:body type="InMessageRequest" namespace="urn:live-stock-quotes" encoding=""/ >
< /input >
< output >
< soap:body type="OutMessageResponse" encoding=" soap/encoding/"/ >
< /output >
< /operation >
< /binding >

It may be somewhat hard to read (and understand), but fragments like these are going to be the building blocks of tomorrow’s software.

TECH TALK: Web Services: SOAP

SOAP (Simple Object Access Protocol) is a mechanism for encoding and exchanging structured messages between distributed Web Services. In the words of Kevin Werbach of Release 1.0, “If XML is the phone system, SOAP is the common language for expressing service requests.”

Write Pip Coburn, Faye Hou and Qi Wang of UBS Warburg:

Using SOAP, information is encoded using XML and the XML is wrapped in hyper text transfer protocol (HTTP). SOAP is used to communicate information or instruction to services, objects and servers across disparate operating systems and platforms. SOAP is a potential key to linking heterogeneous – basically, different -software components and building Web services.

SOAP is a call-response mechanism, meaning that it enables a device or client to make a call to a different server, and for that server to understand and respond to the call. It is built specifically for a distributed network environment. The call and response items are encoded in XML, and an HTTP header is wrapped around the XML, or payload. SOAP allows developers to make remote calls to services running on various platforms, from Linux to Unix to MacOS to Windows.

A more technical definition of SOAP comes from

SOAP is a protocol specification for invoking methods on servers,
services, components and objects. SOAP codifies the existing practice of using XML and HTTP as a method invocation mechanism. The SOAP specification mandates a small number of HTTP headers that facilitate firewall/proxy filtering.

The SOAP specification also mandates an XML vocabulary that is used for representing method parameters, return values, and exceptions.

What SOAP does is let developers work at the API-level, rather than getting into the innards of systems. This abstraction can in the coming years form a smarter web. Amit Asaravala of Webtechniques gives a practical example of how SOAP can be used and its future vision:

Microsoft’s .Net frameworks will use SOAP messages to send information between companies that have agreed to share data. eBay has already agreed to use the .Net framework to open up its auction databases. When the technology is in place, developers from other sites will be able to write auction applications that rely on live data from eBay’s central database.

Applications and devices that implement SOAP will be able to send information to (and receive information from) other compliant applications and devices. This opens the way for the creation of smart appliances that can “talk” to one another to schedule appointments, reorder supplies, and generally automate otherwise manual tasks.

In essence, SOAP is enabling the Web that we don’t see. It’s the technology that will help us realize a semantic, invisible Web that runs in the background, doing our bidding without our constant attention. So long, Web browsers.

TECH TALK: Web Services: XML

XML (eXtensible Markup Language) is a standard for representing structured information. It is, in the words of Software Magazine, “a standard, simple, self-describing way of encoding both text and data so that content can be processed with relatively little human intervention and exchanged across diverse hardware, operating systems, and applications. Information formatted in XML can be exchanged across platforms, languages, and applications, and can be used with a wide range of development tools and utilities.”

XML, though similar to HTML (HyperText Markup Language), is actually derived from SGML (Standard Generalized Markup Language). HTML is the language in which web pages are written. However, there are 2 key differences between XML and HTML. Firstly, XML does not describe presentation of the information, unlike the HTML tags; it only focuses on the structure and content of the data. Secondly, XML, as is evident in its name, is extensible. Tags can be defined by anyone or any group, unlike HTML whose tags are laid down by the World Wide Web Consortium (W3C).

This extensibility along with the separation of form and content is making XML the universal choice for exchanging data between applications. Writes Kevin Werbach in Release 1.0, “The great thing about XML for Web services is that its attributes are self-defining. If Sam builds a stock-quote application and defines fields for price and volume, he can enclose them with and tags, making it easy for Ethel (or a system she builds) to parse what he was trying to do.”

Just as the use of HTML made web publishing easy and made Web publishing easy, XML too holds out the promise of making communications between applications easier. This facet makes it the building block for Web Services. Writes Chad Dickerson in InfoWorld:

XML is a markup specification language and XML files are data: They just sit there until you run a program which displays them (like a browser), or does some work with them (like a converter which writes the data in another format, or a database which reads the data), or modifies them (like an editor).

In other words, as much as we all love it, XML alone is more or less useless. Although XML can be wonderful for trading data among applications, applications do not magically appear around XML documents. XML does, however, function as a great point of leverage for applications, which leads us to Web services.

So, XML by itself is not the solution. But it is a good start. Writes Clay Shirky in, “XML makes it possible for businesses or developer groups to share data, provided they agree on the semantics of that data in advance. This is not to say XML is not an enormous advance. It plainly is. However, its advance lies in aiding data interoperability where shared semantics can be assumed. It does nothing at all to create semantic interoperability.”

TECH TALK: Web Services: The Standards

There are 3 entities required to put together the Web Services framework:

  • Service Providers: who offer the services on the Web and publish the APIs to provide access to the services
  • Service Brokers: who register and categorise the services offered. Think of them as clearinghouses for services.
  • Service Requesters: who need the services and use the Service Brokers to get to the Service Providers. They use a collection of services to put together their applications.

There are three Web Services operations. An IBM White Paper elaborates on these:

  • Publish/Unpublish: Publishing and unpublishing involves advertising services to a registry (publishing) or removing those entries (unpublishing). The service provider contacts the service broker to publish or unpublish a service.
  • Find: The find operation is performed by service requesters and service brokers together. The service requesters describe the kinds of services they’re looking for, and the service brokers deliver the results that best match the request.
  • Bind: The bind operation takes place between the service requester and the service provider. The two parties negotiate as appropriate so the requester can access and invoke services of the provider.

Standards now exist for each of the above elements. Service deployment (“publish”) is done via WSDL. Service Broking (“find”) is done via UDDI. The interface between service providers and service requesters (“bind”) is done via SOAP. The common language for exchanging data is XML. The underlying transport protocol for Web Services is the same as that used for accessing documents on the Internet – HTTP.

The full forms for each of the acronyms are:

  • WSDL: Web Services Description Language (for service APIs – interfaces and functions)
  • UDDI: Universal Description, Discovery and Integration (for finding services in directories)
  • SOAP: Simple Object Access Protocol (for encoding and exchanging messages between Web services)
  • XML: eXtended Markup Language (like HTML; a mechanism to communicate between components)
  • HTTP: HyperText Transport Protocol