Business lies between the lines; Bridging digital divides; Breathe life into great ideas; Make hard work pay

Books2Byte – April 2004

 Business lies between the lines

D. Murali

What makes one tick in the business of software? You might be tempted to say technology, but read on to see if you’re right.


WHAT determines success or failure in the $600-billion software industry? “Technology,” you might answer. But no, “It is the business,” says Michael A. Cusumano in The Business of Software published by Free Press ( . The preface makes the aim of the book clear: “To provide an overview of the software business for managers who are already in the business, programmers who would like to be managers, and anyone who would like to be a software entrepreneur.”

If it is just a matter of business, why a separate book, one might ask. “But software is not like other businesses.” Chapter 1 goes on to explain the differences, starting from the very first, that the goods in this business are `soft’ and `digital’.

At the `heart of the book’ there are seven questions that Cusumano poses. Also, there are three options to choose from: “Become a products company at one end of the strategic spectrum, a services company at the other end, or a hybrid solutions company in between.” What about open source? The author concedes that it is not yet clear what implications the open source movement will have for the software business.

`Best practices in software development’ are discussed elaborately in the book. Cusumano is not surprised that Indian software companies are on top of best practices and doing well in process of software development. “For a decade or more, organisations in India have been studying Japanese factory approaches and aggressively adopting SEI concepts and some synch-and-stabilise techniques. The India story could be the subject of an entire book.”

A chapter is on tips for software entrepreneurs. “Convincing any one venture capital firm to fund a particular idea appears to be an extraordinarily difficult task — far more difficult than creating a successful company, which is difficult enough.” Among the tough problems for start-ups is the `credibility gap’ — “the fear among customers that the start-up, like 90 per cent of all start-ups, will fail”.

To wrap up, the author paints “the next chasm to cross” with thoughts such as these: Good times for technology-based companies must return. Software applications today seem to have relatively few hardware limits. With five billion or so people in the world who do not own a computer, we are still in the early stages of the business of software.

A book that will show how business lies between lines of code.

Linux in baby steps


WHEN two singers meet they don’t talk but sing, it is said. When Sufi masters cross each other’s path, they laugh. So, it is not reasonable to expect `programmer-to-programmer’ interaction to be anything beyond a babble of codes and logic. But the new Wrox book by Neil Matthew and Richard Stones, titled Beginning Linux Programming, 2004 edition, published by Wiley Dreamtech ( covers the basics of Linux. The intro is enticing: “Welcome to… an easy-to-use guide to developing programs for the Linux and other Unix-style operating systems.”

Novices may ask, “What is Linux?” It is a freely distributed implementation of an Unix-like kernel, the low level core of an OS, explains chapter 1. With an increasing resistance to Gates-type proprietary systems, the world may slowly be shifting gears to embark on a four-wheel drive in the free software terrain. Would that be a blessing or a curse, you might wonder. But `curses’ is a library that gets its name from its “ability to optimise the movement of the cursor and minimise the updates needed on a screen”. And there are two levels of curses, base and extended. For the interested, there is a whole lot of documentation on curses in an appropriately-named directory under /usr/share/doc/.

The way Linux manages memory is one of the carefully-controlled illusions that applications work under. “Linux provides applications with a clean view of a huge directly-addressable memory space.” Except for a few specialised embedded applications, the rest are “never permitted to access physical memory directly.” In technical lingo, “Linux implements a demand paged virtual memory system.”

A chapter at the end is devoted to `standards for Linux’. The authors note that the standards have moved forward and Linux itself has been enhanced at an impressive speed by the community.

Linux Standard Base (LSB) covers areas such as object formats for binary compatibility, dynamic linking standards, execution environment, system initialisation and so on.

A good alternative to a full-scale course.

Run a HOMERUN check


ORGANISATIONS working for social change face lots of challenges many of which involve attitudinal transformation. However, to make their jobs easier, activists and NGOs need to convey their ideas effectively. But public communication is an oft-ignored area because the myth is that it is more important to get on with the work. No, there is use in Getting Started in Communication say Michael Norton and Purba Dutt. Their book is part of a series from Sage Publications ( that aims to provide basic practical skills to those running and managing non-profit institutions. Among the avenues explored by the authors are Web sites, Internet mailing lists and newsgroups.

“Some NGOs have brilliant websites; others have terrible sites,” observe the authors. These organisations must remember that the audience for the sites would include members, funders and supporters who may want to find out more about the organisation they are supporting; enquirers to whom the NGO can send information about work; researchers trying to find out more about the issue at stake; and casual browsers who may even turn into future funders.

“It isn’t enough to create a popular site, you also need to create a site that users like,” states the book and that may look like home truth. But a great site ought to meet HOMERUN requirements: “High-quality content, Often updated, Minimal download time, Ease of use, Relevance to users’ needs, Unique, and Net-centric design.” It is important to `offer the user value’ rather than `assuming they are captive’. A common misconception is to want to put as much information as possible because it does not add to cost. “But more is not always better.” So, think of synopsis or executive summary.” Also, “invite people to send in e-mail for full report.”

Among the mailing lists that the book mentions are: GoaNet set up in 1994 by an 18-year old and drawing support from expatriate Goan communities around the world; and India Network Foundation that started as the first Internet resource for India-related news and events in 1988.

“Bangalore-based offers `appropriate and affordable’ IT skills, products and services,” mentions the book. “One of its applications helps non-IT persons to update their websites with news/updates regularly through a simple interface.”

To further their goals, NGOs would have to exploit IT to the maximum, mainly because of the medium’s cost-effectiveness for the reach it can achieve.

Wednesday, Apr 07, 2004


Bridging digital divides

D. Murali

Did you know there are four digital divides to be taken care of? And how do you bridge them? By committing to that goal “the same intelligence and imagination that have gone into creating the technologies themselves.”


VIP convoys criss-cross the country on highways but not everybody is able to see their favourite leaders. Likewise information technology has been proclaimed as the icon of the current age, but it doesn’t touch everybody’s lives. “Despite all utopian drams, the Information Age has so far touched only a tiny majority of the world’s population,” write Kenneth Keniston and Deepak Kumar in IT Experience in India, published by Sage ( . “If we define household access to the World Wide Web as a criterion for joining the Information Age, less than 5 per cent of the world’s population of 6 billion had gained access by the year 2002.”

Therefore, we need to bridge the divide. But there are “four digital divides”, not just one: “The first is internal – between the digitally-empowered rich and the poor. The second is a linguistic-cultural gap between English and other languages and between `Anglo-Saxon culture’ and other world cultures. The next gap is underscored by disparities in access to information technology and between rich and poor nations. Finally, there is the phenomenon of the `digerati’. This is an affluent elite possessing the appropriate skills and means to take advantage of the ICTs.”

There are daunting statistics: That only 1 per cent of the country’s population have home access to computers; of that, only a half has Net facility; more than 40 per cent of the one billion are illiterate; one in two newborns is below ideal birth weight; and only around 3 per cent can afford a telephone. Priorities could be different: With 60 million Indian children not in school, “for the cost of a computer, you can have a school.”

Yet there are bold initiatives. An example: Veerampattinam, a coastal village with 98 per cent of the families involved in fishing, receives information on wave heights in the next 24 hours, downloaded from a US Navy Web site. “The information requirements in that village are focussed on the safety of fishermen while at sea, on fish/shoal occurrence near shore, and on techniques for post-harvest processing.”

Reverting to the divides, how do we bridge them? By committing to that goal “the same intelligence and imagination that have gone into creating the technologies themselves.” A simple reminder that nothing is impossible, nor any chasm that is uncrossable.

Management beans, nuts and bolts


WHAT you specify is what you wanted yesterday. That’s the pace that we’re faced with. “Increasing demand for faster development cycles combined with the desire for more functionality has left less time for building adequate application configuration and management into Java applications,” write Ben G. Sullins and Mark B. Whipple in JMX in Action, published by Dreamtech Press ( . There are the `property files’ to help programmers in specifying a set of parameters. “With more and more configurable attributes, you will quickly find yourself stuck in a mire of property files,” warn the authors. Here’s where Java Management Extensions (JMX) is relevant. “Using JMX, you can expose your application components, attributes, and configuration to management tools in a process called instrumentation.” Chapter 1 guides readers through basics: That today’s management solutions can be divided into two categories, viz. network and application; management information base (MIB) is a hierarchical representation of information about devices; and so on.

If you were a bean counter, you’d come across a special type of bean, the MBean. It is the managed bean, “a Java class that meets certain naming and inheritance standards dictated by the JMX specification”. There are the standard ones and the dynamic MBeans. A chapter in the book explains how you can manage Jini service with a dynamic MBean, and also work on a super class “to provide some code reuse for generating the metadata descriptions of future resources.”

In a chapter titled `MBeans on-the-fly’, there’s the Model MBean that has as its valuable feature the ability to persist itself.

It can “survive the cycling of the JMX agent that contains it. Using JMX relation service, you can create and remove relations, not relatives.”

Relations are objects that contain information describing the relationship between two MBeans.”

Combining Java Message Service (JMS) with JMX is discussed in a chapter on building a home theatre system.

JMS supports two models of messaging, notes the book. “Point-to-point and publish-subscribe.”

A book that can teach you a lot about management – beans, that is, if not nuts and bolts.

Hone your language


SUMMER vacation is on and you want to try out a new language. How about C++ Power Packed, by Kabir Khanna? The book, published by Tata McGraw-Hill ( , is aimed at software professionals programming in C++, not rank novices. Try this: “A popular interview question for aspiring C++ programmers has often been, `What is the return value of a constructor?’ Promptly most people answer it correctly by saying, `none’. This is then usually followed by the question, `How does a constructor then indicate failure in creation of the object?’ That’s a relatively tricky one.”

To explain, Khanna gives a code fragment that defines class Mortal, and within curly brackets sits the line, “Mortal Earthling;” Thereafter, come philosophical questions: “When does the life of an Earthling begin? When does it end? What is the status of the Earthling before it comes alive and after it dies?”

The book is an engaging dialogue for the astute coder who will be able to read between the lines and make sense. For the rest, however, meaning may lie only in those simple lines such as: “The stack is probably the busiest memory area in any program. Misuse of templates can usually cause very hard-to-read error messages on most compilers. Never mix exception logic with normal flow logic. An exception can be thrown and never caught.” There is a C++ test at the end of the book with questions on `things you ought to know’ and those `testing deeper waters’. An appendix is devoted to the new kid on the block, C#.

Khanna is not for the weak-hearted who think that Greek or Latin could be a better choice for the holidays.

Wednesday, Apr 14, 2004


Breathe life into great ideas

D. Murali

Many breakthrough ideas just don’t reach the marketplace, and of those that reach, many fail too quickly. The solution could lie in a `six-stage process.’ Read on…


CONGRATS! You’ve got a new technology!

“Sorry! You still have a long way to go,” says Roger E. Levien in Taking Technology to Market, published by Viva Management Library ( . While the stock of technology is expanding at a good rate, “the capacity to translate technological potential into products and services that can both satisfy customers and reward investors is not expanding correspondingly.” Many breakthrough ideas just don’t reach the marketplace, and of those that reach, many fail too quickly. Levien observes that the main reason for such dismal performance is “the failure to recognise that creating a successful business to market is as challenging and demanding of creative effort as the invention of the technology itself.” The panacea could lie in a `six-stage process’ that the author discusses: “Validate technology, create business proposition, create business model, develop business plan, start business, and enter market.”

To validate, for example, an electronic technology, it may be necessary to show that it operates “in a computer simulation, a broadband mock-up, a prototype, or a full-scale working model.” A temptation to resist at this stage is `to move forward as quickly as possible’ because of getting optimistic as a result of readiness experiments. Protectability is to be ensured: “At a minimum, the principal features of the technology idea should be patented and copyrighted… Apply for patents in those countries where a reasonable chance exists that a large enough market will develop.” The oft-ignored aspect, `the business plan,’ includes inputs on some harsh realities: “Investors will want to see five years of income statements, balance sheets, and cash flow statements… Are the ratios `reasonable’ compared to similar companies? How good is the return on sales?” A big shot at the future, and that would need an accountant who can read the tea leaves.

Launch is a big occasion. However, there could be negative reactions, so your preparation would have to include `counter arguments’ and plans for `future upgrades to overcome identified shortcomings.’ Thus, `marketing’ is not a bad word after all, and in importance may well equal `technology’.

Media unfolding


Sunil Saxena worked as a print journalist for about 15 years before moving to new media. “As Vice-President (Content & Services) of the New Indian Express Group’s Internet company he is responsible for generating and sustaining Internet-based revenue streams,” states the back cover of his book, Breaking News published by Tata McGraw-Hill ( .

But `online journalism’ is `still evolving’ as Saxena would state in his preface. “The most that one can do is to share the excitement, the heartburn and the pitfalls of working online.” So, that must be good fun for readers, as with newspapers.

Section 1 is on `the tools’ of the trade, the craft; it includes basics of reporting, editing and headline writing, plus inputs on site design, Net style and so on. After about 100 pages of the essentials, Saxena discusses `the issues’ in the next section. Just the things one would like to brush under the carpet – Net privacy, defamation, accuracy of opinion poll, intellectual property rights, and the merciless `economics of Web publishing’. In section 3, the author ventures `into the danger zone’ – the future – where revenue streams grow beyond syndication and subscription, to encompass `paid microsites, e-books and mobile content’. Job profiles are changing; for example, `knowledge manager’ would be researching stories and suggesting hyperlinks `to related stories and backgrounders’ even as you push an old pencil stooping over a dot-matrix printout.

Gradually, old revenue triggers are turning ineffective: exam results are published by the boards on their own sites, and film listings have been hijacked by mobile phone operators. Tender notices too are more visible in cyberspace though they are yet to vanish from newspapers because of “an archaic government rule that makes it mandatory for every government department to publish a tender notice in at least two newspapers”. An instructive and entertaining read, not only for journalists but also for readers.

Path of forms


BILL Gates knows what all you need. Which is why his suite offers the facility to design and customise electronic forms that can talk to different applications. To know more, you’d have to read How to do Everything with Microsoft Office InfoPath 2003: A Beginner’s Guide by David McAmis, published by Dreamtech Press ( . “Millions of dollars are spent each year scanning and storing documents, when it is only the data contained in the document that is important,” notes the author in the intro. “That is where InfoPath really shines.” Good, so what is InfoPath? It is “a desktop application that is installed locally and can be used to design or fill out electronic forms.” And it is `the newest member’ of the Office suite. A developer can use this to create forms that are `tied into back-end systems or processes’.

When users fill in the form, the info flows in through one of three methods: XML file, database and Web service. Using the software, you can `consolidate XML files’ and `create summary documents’, which I’m worried would axe the jobs of a few number-crunchers. These forms are no dumb bunch of boxes; you can add `validation’. For example, in an expense report, the form can talk back to the user by showing an error message or highlighting the field if an amount more than what is permissible is entered. Auditors may have to rewrite their vouching checklists, please note.

Well, you don’t trust forms, but you can create a `Trusted Form’ using the RegForm utility, which is as easy as giving a simple command at the dos prompt.

This process creates an installation script (written in JScript). You can export your form, not to Timbuktu, but on to the Web, using, not HTML format, but MHT.

That is, an abbreviation for MHTML, which is MIME Encapsulation of Aggregate HTML Documents, “an Internet standard that was originally defined as a method of sending complete web pages in the body of an e-mail message.”

Now, how do you convert a large number of forms to HTML? Use `downlevel’ – a tool to access which “you need to pass a number of command-line switches, including the name of the form template, as well as a destination for the transformed files”.

If that’s something you want your developer to do, you can get busy with filling the forms on the screen. This may well become the path most travelled.

Wednesday, Apr 21, 2004


Make hard work pay

D. Murali

Whether your business is click and order or brick and mortar, focus down layer by layer, using the `onion skin’ approach. Above all, apply the power of strategic thinking.


BUILD fences around people, and you’ll get sheep.” That’s a poster-able thought, especially for the archaic `establishment’ departments that are masquerading as `human relations’. Also, it’s a belief statement of William McKnight who moulded 3M to `embrace creativity’. To this Paul M. Elkin adds a `hard’ fact: “Build walls around an organisation, and you’ll get a coffin.” Elkin’s book titled Mastering Business Planning and Strategy, published by Viva Books P Ltd ( is about `the power and application of strategic thinking’.

The first thing to hit you is the malaise that most managers suffer from: “Completely occupied by `fire-fighting’, dealing with the crises and problems that are occurring today, rather than considering what is necessary to ensure the survival and eventual success of the business.” What’s the use succeeding in tactics but failing in strategy? So, whether your business is click and order, or brick and mortar, focus down layer by layer, using the `onion skin’ approach to handle the environment, industry and corporate peels. After you have done the SBU (strategic business unit) analysis, move on to profile the competitors by drawing a matrix and developing a map of CSFs (critical success factors). Well, even as you’re busy tweaking your software, your finance man is routinely churning out budgets and plans by making `small adjustments to the level of costs incurred’ to improve `financial efficiency’. No, that won’t do, so go for `cost profiling’ – an analytical technique `to identify potential opportunities for more radical changes to the business cost base’ and improve `financial effectiveness’.

In the chapter on performance measurement, the author discusses SKPIs (strategic key performance indicators), followed by the financial, operational and city KPIs. You know EVA, which Drucker called `the measure of total factor productivity’, and its cousin MVA. Another EVA is employee value added – a measure one finds in many IT company annual reports. This ratio has total added value generated by business in the numerator and total cost of employees in the denominator. A book that argues for adding the strategy software to make your hard work pay better.

Eclipse to the extreme


DAVID Gallardo had but one purpose when starting Eclipse in Action. He wanted `to introduce Java developers to Eclipse’. In due time, Ed Burnette and Robert McGovern joined him and what you have is a guide on the topic, published by Dreamtech Press ( , that goes beyond Eclipse to the plug-in terrain, and has stuff on SWT and JFace too. “The mother of invention is not necessity, it is irritation,” is a quote of Henry Petroski that Bob Foster cites in his foreword, to emphasise the role of tool-building. One feature of Eclipse that strikes Bob as `extraordinary’ is `the excellent technical support provided in the Eclipse newsgroups by the actual people who wrote the code’: “In no other open source project are developers so committed to answering any and all questions thrown at them. In many cases, questions are answered with source code written and tested for the occasion. For a programmer, it doesn’t get much better than that.” Though Eclipse is free, you can make money by extending it. Eclipse’s licence allows one to charge for his/her `Eclipse-based extensions’.

Chapter 1 introduces readers to what makes a software developer comparable to a blacksmith. “Many blacksmiths take pride in making their own tools… Using forge, anvil, and hammer, the blacksmith repeats the cycle of heating, hammering, and cooling the steel until it becomes a tool of exactly the right shape, size, and strength for the job at hand.” So? “What code has in common with metal is malleability.” Thus, Eclipse is the software developer’s equivalent to the blacksmith’s workshop. Written in Java, Eclipse is language-neutral, and also human language neutral. However, it is not strictly platform-neutral. “This is due to the decision to build Eclipse using the OS’s native graphics. Eclipse is therefore only available for those platforms to which SWT (standard widget toolkit) has been ported.” The fundamental component is the Eclipse Workbench. It has one simple job: “To allow you to work with projects. It doesn’t know anything about editing, running, or debugging Java programs; it only knows how to navigate projects and resources.”

As a software development tool, Eclipse is well suited for certain styles of programming. “Currently, the most fashionable style is XP,” write the authors. That’s not the Windows from Microsoft, but eXtreme Programming. “One of the most unique and controversial approaches advocated by XP’s proponents is pair programming: At all times, two developers sit at a single terminal while writing code.” Something auditors always insist as an internal control measure, but “more developers are probably talking about XP than doing it.” Smooth reading.

Cracking secrets


A WHOLE book of secrets is what S.C. Coutinho gives in The Mathematics of Ciphers, published by Universities Press ( . “A leisurely journey, with many stops to appreciate the scenery and contemplate sites of historical interest”, the author promises to reach the final destination – RSA system of cryptography. Since the work has grown out of lectures to first-year students of computer science, there is no presumption of mathematics knowledge. “Cryptography is the art of disguising a message so that only its legitimate recipient can understand it.” That should explain why we don’t understand many election speeches. Perhaps the `twin sister’ of cryptography could help, cryptoanalysis: `the art of breaking a cipher’. The most widely used public key cryptosystems is RSA, invented in 1978 by Rivest, Shamir and Adleman. Put simply, “every user has a personal pair of primes that must be kept secret” though the product of these primes is made public. What’s the big deal, you might ask; factor the product and you would get the two prime numbers, won’t you? “However, if the primes have more than 100 digits each, the time and resources required to factor `n’ are such that the system becomes very hard to break.” This is the trapdoor of RSA – computing product is easy, not factoring. For this, the `exact computation’ of computer comes handy. Greeks distinguished between logistics (the science that deals with numbered things, not numbers) and arithmetic (nature of numbers with the mind only).

The book is full of stories that would make you like math and computing too. For instance, geometry originated in Egypt where the pharaoh distributed land to people in rectangular plots on which he levied an annual tax. “If the Nile swept away part of the plot, the surveyors had to be called in to calculate how much land had been lost.” Because the owner would be eligible for a reduced tax, proportional to the land lost.

To find primes from the ocean of numbers, you can use the `sieve of Erathostenes’, named after a Greek mathematician born around 284 BC. He was nicknamed `Beta’ because his contemporaries believed that he hadn’t reached a truly eminent position. When you apply the sieve to a list of positive integers, composite numbers pass through but primes get retained. Good read for the vacation to sharpen your numbers.

Monday, Apr 26, 2004


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s