Gartner on the New Enterprise Architecture

A Gartner special report explores the four elements of a new foundation in enterprise architecture:

– The “grid”
– Architectural styles that represent key business processes
– Patterns, which are logical technology models
– Core technology building blocks, which we call “bricks,” that link to the grid

Elaborates Gartner (I like some of these research firms’ reports because they introduce a nice variety of new terms, besides giving the occasional good idea):

The notion of the grid has its roots in the trickle-down of the Internet, using its concepts of access, componentization and interoperability for the interconnection of multiple enterprises. It is an interoperability platform. Some of the core components of the grid include the “multienterprise nervous system,” a security and availability management system, information and application management, data exchange through XML and its tools, governance rules and development platform methodologies. The multienterprise nervous system manages network traffic, connects, manages, monitors, translates protocols, integrates and does real-time status management – all facilitating dynamic system interconnection among enterprises. It is the smart network all grown up. Our future vision shows the grid becoming more of a living system, by being self-aware and able to self-heal, self-reconstitute and self-manage.

Bricks are foundational architectural elements, such as operating systems or databases that link to the grid to provide technology function. These bricks are the basic elements for building systems. They can have varying levels of granularity from specific components, such as gateways to platforms. Think of the grid as providing the “juice” that enables these bricks across enterprises to communicate.

Patterns are groupings of bricks – essentially logical models of technology. The concept of patterns is useful because patterns can be reused and can help create a shared vocabulary around specific design and implementation considerations. Some types of two- and three-tier computing models are examples of patterns. So are hub-and-spoke and message warehouses. The key is to leverage patterns so that they enable cross-enterprise component specification and use. For example, logical extensions to the user interface, business logic and data layers enable departmental, enterprise and cross-enterprise architectures.

Architectural styles, the last of the key elements, bring the business domain into the architecture process. Key business process elements – for example, the need to manage shared inventories, or allow multienterprise postings against a supply database by collaborative-commerce partners – require a different style of computing: a collaborative style in the first, and a transactional style in the second. Think of these styles as templates for viewing how common business processes can be standardized and then devolved into patterns and bricks. We have defined five styles: transaction processing, real time, analytical, collaborative and utility. Each of these business process examples is best-served by a different architectural approach.

New terms learnt: multienterprise nervous system, governance rules, dynamic system interconnection, foundational architectural elements, cross-enterprise component specification.

Telecom’s Bleak Future

Business Week has a story asking When Will the Telecom Depression End? The answer is not heart-warming:

BusinessWeek spent a month examining the capacity for each type of telecom service, from long-distance to wireless, and comparing it to worldwide demand. The results show that capacity continues to dwarf demand. Prices in America and Europe remain under pressure. Meanwhile, rollouts of new cables promise to extend excess capacity to regions such as Asia that have been spared much of the pain to date. “We’re not seeing any turnaround,” says BellSouth Corp. CEO Duane Ackerman.

The upshot is that the crisis could last until at least 2004. In the U.S., traffic at the core of the networks is leaping ahead at 85% a year, with Europe and Asia at similar paces. Within two years, that should soak up excess capacity of networks in operation, which are running at 35% of capacity in the U.S. and Europe and at higher rates in Asia. An economic upturn, expected by the end of 2003, could spell recovery for U.S. telecom carriers six months later. Europe is expected to follow suit in late 2004.

But things could get worse. If the world economy continues to struggle or if telecom companies fail to lop off capacity and come up with lucrative new data services, this depression could continue through 2006. Even when recovery arrives, most of the once-robust telecom players are likely to perform, at best, like stolid, slow-growing utilities through the end of this decade. Growth is likely to be 2% or 3% a year, predicts Lawrence Kenny, head of the telecom practice at PricewaterhouseCoopers.

If their future is bleak, then who is going to give me cheaper and plentiful bandwidth? My office in Bombay, India has 2 connections to the Internet: a 164 Kbps/64 Kbps ADSL and 128 Kbps via cable (as a backup). We pay USD 200 per month for each of them. [You read right on both counts – the connection is in Kilobits and the charges are in hundreds of dollars.] Is my broadband future doomed?

Business Week’s e.Biz 25

Here. The print magazine (Asian edition) didnt seem to have the detailed profiles that they have on the website.

I may be wrong but none of the 25 people selected by Business Week has a blog. Maybe they should make that a mandatory criteria next week – I want to read what the individuals think of the future, not necessary an interpretation by a journalist (however good). That would surely start a blogging stampede and do a lot of good in the world in terms of transparency and ideas flow…which is just what we need. Integrity and Innovation.

By the way, the introductory para is an excellent of the wonderful and flowery hyperbole the business magazines are so good at: “In the span of just two years, e-business has morphed from capitalist cure-all to pure catastrophe. The dot-com collapse took down not only the majority of e-tailers but also a wide swath of software and hardware companies that catered to them. The ripples from that disaster had assumed tidal wave proportions by the time they hit the telecom business, where the Internet bubble fueled wild overexpansion.”

Web Services and Business Processes

From InfoWorld:

“The business process is becoming the new unit of work,” said Charles Fitzgerald, Microsoft’s general manager of platform strategies, pointing to the forthcoming BPEL4WS, a programming language that can be used to describe business processes, as one example.

He added that using Web services for integration opens up new possibilities not typically associated with more expensive integration projects.

“If we have a mechanism to get different systems to talk to each other, we can start to think about the business processes,” he said.

He continued that Web services projects thus far fit into two groups. The first is people integrating with partners; the second is companies using Web services to streamline business processes.

An interesting comment on this comes from Rahul Dave on John Robb’s blog post:

Whats forgotten in the whole business-process-as-a-unit argument is that business processes are human processes like other things, and employees and managers need to tweak them, and play with them, not have some developer write it once and forget.

And you dont make process writing a normal user skill by selling web services to business. You sell it on the desktop, in the way you provide for routing messages and aggregation. In other words you do it by bringing scripting closer to the average user, through templates or super simple scripting languages (even usertalk and python are too hard.., but on the right track, usertalk more so with frontier-db, and python less so with ZODB.

I like the idea of providing users greater control with scripting languages which give them the ability to decide what they see. In fact, imagine a digital dashboard comprised of RSS feeds which bring in blog posts and enterprise events into view in a sort-of “Events Horizon”, filters (in a scripting language, like what Rahul talks of) that work as event processors/routers, an information visualisation tool like Smart Money’s Market Map which allows users to converse with information. This can be a wonderful productivity enhancer.

The problem faced by people and enterprises developed markets is too much information. The problem faced by their counterparts in the emerging markets is too little information. But the advantage that the latter group has is that it is devoid of legacy. Its weakness can actually be seen as a strength in that it enables the creation of innovation software revolutions that can enable a technological leapfrog.

USB

From ZDnet: “The arrival of portable USB storage devices–so small that they fit easily in your pocket and even sport rings for easy key chain attachment–may finally put the venerable floppy to rest. These portable devices store from 8MB to 2.1GB of data and take up a fraction of the space of other storage options.”

The USB could serve two purposes in the TC-TS solution: for boot-up (we currently do so from either a floppy or a boot ROM – both of which need to be created for the TC), and for authentication (much like Sun’s idea of the JavaCard which it is doing with a smart card).

A related USB story comes from News.com. Irt writes about USB-On-the-Go, “a new technology that allows handheld devices to share files directly, without the need for a PC….Using it, a person could plug a handheld or digital camera straight into a printer to produce a photo. PDAs also could swap documents directly or back up data by connecting directly to a portable hard drive. The technology is also expected to be used in cell phones and MP3 players.”

Brain’s Management Secrets

From Business 2.0: “Your brain is the ultimate example of a complex, decentralized organization. And because we (usually) behave coherently, smoothly integrating new circumstances as they arise, the brain is also the epitome of an adaptive organization, a learning organization, a shared-vision organization — in short, the ideal modern company.” The 5 secrets, according to the article:

1. Never try to micromanage a large, complex organization.
2. Don’t let bottom-up self-organization go wild.
3. The best way to control your subordinates is to just point them in the right direction.
4. Be careful listening to the voice of experience — that voice could be your own.
5. The organization can’t succeed without passion.

TECH TALK: The Years That Were: 1998

From a Wall Street Journal special report on Entertainment and Technology (March 19,1998):

If any industry would seem to have mastered technology, its entertainment. As anybody who goes to movies, listens to music or plays video games knows, the techies are constantly topping themselves, turning yesterdays gee-whiz into todays yawn.

But, when it comes to the economics of high-tech, the entertainment industry is still very much in the dark. Movie makers, for instance, havent yet figured out how to say no to special effects or how to make sure that plat and character dont get buried beneath an avalanche of special effects. TV networks and studios dont know what to do with digital studios. Recording companies are still trying to understand how to see the Web to sell CDs. And video game creators havent yet uncovered the secret of turning hit movies into hit games.

The reasons for all the confusion are many. Part of it is just the love of doing something just a little more spectacular, a little more glitzy. Part of it is that people are making bad decisions. And part of it is just the newness of it all.

Business Weeks August 31,1998 issue was on The 21st Century Economy, with a byline on the cover that said, Volatility is here tostay, but technology and globalization will spur robust growth. Its findings:

The 90s are no fluke. Despite Asias woes, all the ingredients are in place for a surge of innovation that could rival any in history. Over the next decade or so, the New Economy so far propelled only by information technology may turn out to be only the initial stage of a much broader flowering of technological, business and financial creativity that will sweep across the world.

Call it the 21st Century Economy an economy that, driven by technological progress, can grow at 3% pace for years to come. The innovation pipeline is fuller than it has been in decades. With the advent of the Internet, the information revolution seems to be spreading and accelerating rather than slowing down. Biotechnology is on the verge of having a major economic impact, and in labs, scientists are testing the frontiers of nanotechnology, with the goal of creating new devices than can transform entire industries.

To be sure, the path from the New Economy to the 21st Century Economy will likely be a bumpy one. Each innovative surge creates economic and social ills, from recession to stock-market crashes to widespread job losses and this one wont be different. But thats the price a nation must pay to achieve the benefits of dynamic change.

Added an editorial entitled The Innovative Society in the same issue: There is, however, a price to be paid.History shows that a high-productivity, fast-growth economy is prone to greater, not fewer shocks. Nations can rise and fall as they try to adapt to new technologies. It may be that Japans unexpected decline was the first example of a major industrial country unable to adapt to information-age technologies, turning instead to a a beggar-thy-neighbor policy of currency devaluation that helped trigger the collapse of Asian prosperity. And the ultimate success of European unification may well depend on Europes ability to integrate technology across its borders.

Tomorrow: 1999

Continue reading