In a recent discussion, a manager kept asking me about requirements management in agile development, and it dawned on me that many traditionalist continue to focus on requirements, whereas agilists focus on features, capabilities, and stories.

Recently, a well-known IT advisory group brashly announced a "new" approach to enterprise architecture that it described as "emergent" and "nondeterministic," with central IT and its architects ceding some control and choice to constituents rather than trying to design and control everything.

"Be afraid ... be very afraid."

Many of our business processes -- operating over the Internet or otherwise -- remain wedded to a production-line mindset that is stuck in the world of the 1980s. A well-planned business process managemenht (BPM) and service-oriented architecture (SOA) strategy needs to encourage the use of services as a tool to improve processes and solve specific business problems.

In my previous two Advisors ("Selected Innovations in Cloud Products," 9 September 2009, and "SOA and the Cloud: Getting Past the Hype," 19 August 2009), I talked about the different styles of the cloud (infrastructure as a service [IaaS], platform as a

Five to seven years ago, Java versus .Net was a hot topic. At that time, many organizations were at this important crossroad. Now, they have all made up their minds. But some are reconsidering their previous decision.

Putting it very briefly, .Net currently is more common in smaller, front-end projects, where integration with the desktop is essential, while Java is more common in larger, back-end projects, where legacy platforms run the core business applications and where integration with legacy systems is crucial.

Many organizations assume that if an obligation has been stated in a contract, the provider will comply with it and no further work needs to be done. However, astute organizations do not assume compliance; they ensure it. The time to discover the provider has not done so is not when your organization is seeking to invoke the clause, as the next case illustrates.

All the recent talk in the US surrounding proposed healthcare reform has inspired me to take a look at how BI and data warehousing are making an impact on healthcare and medicine. For our international readers who've not been following the healthcare debate in the US, the political discourse has reached a level of circuslike behavior.

There was an interesting and well-written column in the New York Times editorial section a few weeks ago called "Sleepwalking Through September," by Doug Glanville (20 August 2009).

We've never seen the effects of a recession during a time when so many customers and businesses depend on IT. Companies are responding very naturally by cutting projects and cutting investments in IT.

Agile testing is, among other things, about testing early, testing often, and testing smart. How do you achieve that? A combination of actions that can take place in parallel can help you get there. You can start by increasing your QA staff's skills through training (formal or peer-based) so that they are able to not only execute automated tests but also enhance the existing test-code base and create new ones.

Efectuar pruebas de forma agile tiene que ver, entre otras cosas, con probar tempranamente, muchas veces, y astutamente. ¿Como puede lograr eso? Llevando a cabo una serie de acciones en paralelo. Puede comenzar con incrementar el nivel de conocimientos del personal de QA mediante entrenamiento (el cual puede ser formal o mediante compañeros de trabajo) con el fin de que no tan solo puedan ejecutar pruebas automatizadas sino también mejorar el código de pruebas existentes, e inclusive codificar nuevas pruebas.

In one of my recent Trends Advisors ("One Small "Oops" for Amazon, One Giant "Holy #$@%" for Mankind," 23 July 2009), I discussed the unpleasantness between Amazon and its customers who found one morning that their (now illegal) copy of George Orwell's 1984 had been somehow removed from their Kindle electronic rea

"The worst may be behind us."

So said US President Obama in August, ever the optimist. He was not alone, however. Joining him was a rather large chorus of investors who over the past few months cheered some of the economic data, driving the stock market up.

In ancient times, the Oracle (the person, not the database) was someone to whom you addressed important questions. The answers that the Oracle gave were always true, but were often given in an elaborate code. I imagined recently that I took some of my clients' questions to a modern version of the Oracle for enlightenment.

We read much these days about business process management (BPM) and service-oriented architecture (SOA) converging. Is that just hype, or does it really make sense? And if it does make sense, just what might that mean for our organization -- not just in terms of technology, but also in terms of that subtler, softer kind of thing we call "culture" -- the unwritten rules of the game?

Search is having a major influence on BI and data retrieval and analysis in general. Although the technology is still developing, the combination of BI and search is important because it can provide nontechnical business users and BI consumers with easier access to -- and the ability to analyze -- both structured and unstructured information.

Cutter is paying attention to IT governance in 2009. The September Cutter Benchmark Review (CBR) reports the results of the recent Cutter IT Governance Survey. The December Cutter IT Journal, which we're editing, is about IT governance. Recent CIO surveys done by others place IT governance in the top echelon of concerns.

Why all this attention? Simply, because effective IT governance creates IT value.

In his groundbreaking book Crossing the Chasm, Geoffrey Moore develops his interpretation of the technology adoption cycle of Everett Rogers. This cycle starts with the innovators, moving on to the early adopters, the early majority, the late majority, and finally, the laggards.

The shift from local to distributed, from physical to truly digital, is inevitable. Maintaining digital infrastructures, however commoditized, is not core to any of the beneficiaries of the Web, just as running one's own power plant makes no sense. Preteens, adolescents, and young adults don't even think about these things.

There was an interesting and well-written column in the New York Times editorial section a few weeks ago called "Sleepwalking Through September," by Doug Glanville (20 August 2009).

The product team is a cross-functional group of managers and/or individual contributors who are collectively responsible for managing the product's lifecycle. The size of the product team will vary depending on the role of the product manager, as described previously in this series.

In my last Advisor, "SOA and the Cloud: Getting Past the Hype" (19 August 2009), I introduced the main categories of cloud computing. This week, I'll cover some of the product innovations taking place in those categories.

Judging from the wildly enthusiastic international crowd of 1,400 attending the Agile2009 conference in Chicago, Illinois, USA, last week, XP, Scrum, and agile software development methodologies are becoming more mainstream. After years of pilot projects, agile is finally gaining acceptance as a legitimate approach to software development in the business community.

Outsourcing represents another way of managing the costs of an activity. Whether we outsource the mowing of our lawn to avoid the cost of owning and operating a lawn mower, or if we outsource the costs of developing an IT service, the result ends up in a net savings of direct and indirect costs.