The Hindu Business Line : Start with intuition but...

Almost 300 tips in less than as many pages. That's Lessons Learned in Software Testing, by Cem Kaner, James Bach, and Bret Pettichord, from Wiley India (www.wileyindia.com) . With each tip comes explanation; `the heart of the book', as Tim Lister describes in his foreword.

Beginning with `the role of the tester', the book progresses to: thinking like a tester, testing techniques, bug advocacy, automatic testing, documenting testing, interacting with programmers, managing the testing project, and so on.

The first tip tells the tester, "You are the headlights of the project." Because most projects worth doing are like `driving a truck off-road, in the mountains, at night.' What does a tester do then? "Illuminate the road ahead for programmers and managers."

Lesson 5 urges, "Find important bugs fast." Means: check first fixes and updates, focus on core functions, test capability before reliability, paint popular data scenarios before esoteric ones, handle common threats before exotic ones, give preference to testing high-impact problems, and attack `the most wanted areas'.

The tester has to `run with the programmers,' say the authors. "Aim for the shortest, quickest feedback loop you can." Question everything, not necessarily aloud, else "your testing will be aimless and mechanical." Lesson 8 states, "You focus on failure, so your clients can focus on success."

But who are the clients? The project manager who has to know your process, the programmer who benefits from your bug reports, the technical writer who gets alerted to documentation errors, the technical support personnel who learn about where troubles can crop up, and also the marketing people, top management and the users.

"Testing requires inference, not just comparison of output to expected results," reads lesson 20. Because "test designer almost never has access to an authoritative guide to what should be tested, let alone what should be expected." Another lesson dins in, "A tester is more than a tourist." How? "A tester's efforts are devoted to evaluating the product, not merely witnessing it."

Isn't intuition key? "A fine beginning, but a lousy conclusion," declare the authors. Gut feel is a starting point, not a resting point. "Apart from the fact that intuition is often strongly biased, the real trouble comes when you try to get other people, such as programmers and managers, to take your bug reports and quality assessments seriously." Therefore, use your intuitions "as a guide, but not a justification."

How to test something complex, so daunting and overwhelming that one feels `intellectually paralysed'? Do it in bursts, advises the book. "Don't expect to comprehend a complex product all at once. Try throwing yourself at it for 30 minutes or an hour. Then stop and do something else. This is the method of plunge in and quit."

Lesson 40 is very important: "You're harder to fool if you know you're a fool." How so? Because "if you know in your bones that you're easy to fool, you become a little more alert." Easiest way, this is, for a novice tester to improve, say the authors. And there are more than 250 more tips!

Smart read.

Success stories in service delivery

A recent World Bank report is Reforming Public Services in India, from Sage (www.indiasage.com) . Its focus is on `documenting success and drawing lessons'. The first such success story is of `India's telecom revolution'.

It took 20 years to dismantle the monopoly of the DoT (Department of Telecommunications), notes the book. The monopoly had "provided a channel for rent-seeking behaviour, particularly by its staff." And the consumer subsidised "an over-manned Telecom Department of over 4 lakh employees, many of whom had been appointed through political patronage," as the report narrates.

One of the factors that contributed to the success in telecom was the viewing of its modernisation as "a critical element in the growth of a world-class economy in India." Another factor was the pressure of the IT industry; also helpful was "the willingness of the government to institutionalise the industry's views through the creation of a national task force."

Even as cellular operators grew as a lobby to push for reforms, it was the emergence of an independent regulator in the form of TRAI (Telecom Regulatory Authority of India) that gradually curbed rent-seeking by the Telecom Department, "by enforcing competition and fairer licensing practices."

The second case in the report is of e-Choupal-Sagar. A key legal change that finds mention is that of Madhya Pradesh Agricultural Produce Marketing Act of 1972. The 2003 amendment permitted "the issuance of a single licence, on payment of a fee of Rs 2 lakh, to transact business with farmers directly anywhere in MP," thus bypassing official mandis.

"These regulatory changes, in turn, laid the foundations for ITC to engage in direct transactions with farmers," recounts the report. And IT provided the link, "for the procurement of soya and wheat, beginning with the placing of orders at a village e-Choupal Internet kiosk to its actual purchase by ITC at one of its Sagar procurement hubs." Sagar used as benchmark `the closing price at the nearest mandi'. Also, it used electronic weighing. "The brightly painted Sagar hub with a spotless cafeteria and toilets," featured "a well-stocked department store for farmers with cash to spend."

The kiosks did more; "they disseminated information on mandi prices for soya and wheat across the state." Films showed "application of good practices in cultivation". Also, with the help of the Meteorological Department, district-level weather forecasts came through e-Choupal Web site (www.echoupal.com) , accessible at the kiosks.

The model is going places, such as Uttar Pradesh (about 1,700 kiosks), Rajasthan and Maharashtra (800 kiosks each), and Karnataka (100 kiosks). "ITC is now planning to extend the e-choupal concept to medical services: kiosks will deliver medical information online to villagers, and set up appointments to visit the dispensary at a Sagar hub for proper attention and drugs." Currently, more than 5,000 kiosks, spread in six states, reach out to 35 lakh farmers.

A chapter titled `Simplifying transactions' begins with the eSeva model of Andhra Pradesh. Under one roof, it offers "services of about 13 state and local government agencies, three central government agencies, and nine private sector organisations." The services include "payment of utility bills, provision of birth and death certificates, payment of property and other local taxes, train and bus reservation, private cell-phone bill payments, receipt of passport applications, and even the transfer of shares."

Extensive back-end computerisation supports eSeva, so transactions get instantly recorded in a central server. "eSeva counters have also been established in banks - both private and public - and its services can be accessed online through the AP portal (www.aponline.gov.in) ." Has eSeva reduced the payment of speed money? The report notes that, tactically, eSeva hasn't been spoken of as an attempt to curb corruption and reduce drudgery for department officials.

A similar initiative in Kerala is FRIENDS, short for `Fast, Reliable, Instant, Effective, Network for Disbursement of Services.' However, this promotes "single-point front-end service delivery without any real back-end computerisation." Read also about Karnataka's Bhoomi, which took IT to farmers; and Gujarat's computerised interstate checkposts.

Reassuring report.

Tailpiece

"SMS survey on polls said... "

"Press 1 for party X, and 2 for party Y?"

"Yes, and 3 for `bid now'!"

http://BookPeek.blogspot.com