01.28.05
Posted in Application Development, SOA & Web Services at 2:24 am by Tony Baer
Why is it that every new model of cell phone sports a slightly different size recharger plug or antenna mounting? With new models competing on various camera and messaging features, you’d wonder why product engineers are wasting their time redoing silly details like recharger plugs.
Much of it has to do with the engineering mindset, where reuse of somebody else’s design is often considered an admission that you couldn’t come up with anything better yourself. Consequently, software development isn’t alone with this problem, but in these days of increased scrutiny over IT budgets and productivity – not to mention the emergence of offshore competition – you’d think that software engineers would finally wise up.
Problem is, with growing time to market pressures, and with the emergence of tools that make it easier to “service-enable” old software, retrieving existing designs is often less expedient than starting from scratch. The stumbling block is that it takes time to architect, something that remains a luxury for teams under pressure to churn out code.
Not that the software field hasn’t tried addressing the reuse bottleneck before. What do you think computer-aided software engineering (CASE), object-oriented design, and component-based development were all about anyway? Then there’s ZapThink’s proposed metamodel for services, which updates these ideas for the world of services-oriented architectures.
IBM Business Consulting Services (BCS) is releasing a new Services-Oriented Modeling (SOMA) Architecture process that provides a good first stab at adding some order to the process of service enablement. First you prioritize business goals, then analyze and decompose existing components or other software assets until you get to a lowest common denominator that can be exposed as reusable services. Although based on conventional ideas, SOMA is well thought out. At this point, SOMA only deals with the design process, but BCS eventually plans to extend this to run time to manage the full life cycle of services.
Naturally, we’d love to see more. For starters, extend the company’s Rational Unified Process – the methodology for top-down corporate software development – to incorporate SOMA. More importantly, although clearly intended as another way for IBM BCS to make money, we’d love to see ” SOMA lite” thrown over to the open source community (the Eclipse organization is a logical choice). Not only would this open SOMA to valuable external input, but it would also make the process more accessible to organizations that otherwise can’t afford the high cost of IBM BCS services or necessarily require all the steps that the current process involves – not to mention increasing IBM’s mindshare in the area.
Permalink
01.14.05
Posted in Technology Market Trends at 2:23 am by Tony Baer
There was something a bit ironic about watching Craig’s List cofounder Craig Newmark address a packed room of real estate, media, and tech executives at the Waldorf Astoria the other day. The guy – a self-effacing, and by his own admission, classic nerd – discussing the secrets of the success of a sub-$10 million company that still lacks a business plan in the conventional sense. Wasn’t this the type of venture that the dot com bust was supposed to have winnowed out?
Yet, if you want to advertise things like apartment rentals or related services in the two dozen or so largest urban markets, there’s really no other place to go. We can testify to this one personally – minutes after we posted a sublet of our Manhattan apartment on Craig’s List, we were besieged with dozens of offers. Within a couple days, the deal was signed, sealed and delivered – all without paying a cent or worrying about the word count of our ad.
Of course, one could conclude that venture’s distinguishing factors, including the “culture of trust” and the site’s unassuming personality, has much to do with its success. One could also draw a connection with Google, recently profiled on 60 minutes, whose credo is “Do no harm,” or eBay, whose egalitarian ethic promotes a vision that the way to build a killer brand on the web is not to be a killer.
(Incidentally, Newmark characterized eBay’s small minority stake in Craig’s List as a fluke – the result of a former employee selling his stake.)
In large part, the success of Craig’s List has been attributable to the web’s low cost of entry. That could explain why the venture has prevailed against incumbent print media, whose stakes were more defensive. But having not followed the online classified market closely, we can’t explain why Craig’s fended off other rivals except for one reason – when you don’t charge for ads and word spreads, it kind of hard to beat the momentum or top the value proposition.
Today, the service makes its money only off nominal charges for want ads in three cities (San Francisco, New York and LA). Ironically, Newmark said that realtors are actually pushing for charges to their part of the site so as to filter out fly-by-nighters.
The success of Craig’s List, and eBay, Google, and Amazon along with it, testifies to the power of branding. In that sense, it’s little different from the rest of the economy. If you know there’s one place by the highway where everybody sees your sign or one place where you know that prices are cheapest, it’s kind of hard to justify going elsewhere. Maybe the Internet didn’t change everything after all.
Permalink
01.06.05
Posted in Application Development, IT Services & Systems Integration, Outsourcing, Technology Market Trends at 2:18 am by Tony Baer
In the 1990s, conventional wisdom was that economic cycles were relics of the past, and that the inevitable march of technology would render IT recession-proof.
The last five years have certainly shaken those assumptions. Today, IT is often lumped with mature sectors like automotives, steel, or appliances. Recently driving past a bunch of empty Rte. 128 office parks sprouting dozens of For Rent signs, we recalled similar drives a decade ago through the Ohio River Valley, where we gazed upon abandoned hulks of steel plants resembling Mediaeval ruins.
But not all mature industries stay mature. Take consumer electronics. Twenty years ago, just about every Baby Boomer had a stereo set. Yet in recent years, digital technology reinvented the field — just gaze at all the iPods, not to mention High Definition plasma TVs gracing the landscape. Significantly, the Consumer Electronics Show (CES) has replaced Comdex as the place to be seen.
A stroll through the splashy Samsung showroom at New York’s Time Warner Center reminds you that the possibilities of marrying consumer electronics and computing are limited solely by the imagination. Only time will tell as to which combinations actually morph into viable markets.
Over the past few years, emerging technologies like WiFi have helped transform mobile computing. But as upstarts like Cometa learned the hard way, WiFi wasn’t a standalone market. Blackberries have proven that there is a commercially viable mobile text messaging market — but does that mean we’re going to forsake the convenience of the tiny cell phone footprint if all you want to do is reach out and call someone?
And just how much content or communication can one actually consume, or be willing to pay for, while in motion? Obviously, this is the stuff of market trials and ergonomic studies.
The common thread is that the success of IT is increasingly reflected in its invisibility. Voice over IP (VoIP) will supplant and enhance conventional circuit-based phone service only if it looks and feels like the thing it is replacing. So you may not “see” IT, but it is becoming ubiquitous and essential. For instance, the emergence of technology-enabled services like hosted applicatitons requires domain expertise to drive it, but IT architectural competence to support it.
Ironically, the flipside is that the emergence of “IT invisible” markets may be creating demands for the very people that technology was supposed to replace.
* For every 10 Indian offshore developers, you still need people close to the customer who understand local culture and prevailing business practices.
* Voice recognition technology may automate the offshore customer support call center people who once replaced domestic counterparts. It also could create new job opportunities for Level 2 or 3 support back home, because voice recognition and remote call centers only take you so far.
* Business process outsourcing creates new opportunities for corporate veterans who toiled away for decades before their companies deemed them redundant.
Maybe the one lesson we’ll learn in 2005 is that technology growth is not necessarily a zero sum game.
Permalink