Archive for January, 2009

Good business technology is basically like good pasta

January 30, 2009

 

Stringy, fatty and bland: not three words you want to associate with a good meal. They’re not three words you want to associate to your business IT, either.

 

Take spaghetti architecture, the curse of poor IT development and the most unsavoury of technology flavours

 

Such spaghetti relies on complex and tightly coupled interactions between enterprise applications. The result is a highly unsatisfactory main course that leaves users feeling empty.

 

So, what pasta should you be serving to the business? Good technology is connected, lean and tasty – it works well together and responds to user needs.

 

The chances are that you’ve created a lot of separate side dishes – off-the-shelf and bespoke systems that are used and consumed in different ways across the company.

 

These individual systems are more helpfully viewed as the fusilli; freewheeling components that serve a purpose – but could provide a tastier experience if they were more effectively integrated.

 

IT managers searching for the perfect sauce should turn to service-oriented architecture (SOA).

 

Like a good lasagne, layers of technology work seamlessly together to create a stunning feast – and SOA will provide the integration between layers, allowing the business to re-use components in new and appetising combinations.

 

But don’t just think of your SOA as a one-size-fits all strategy; different business requirements require specific modifications to the menu.

 

Layers should be added carefully and each SOA implementation should address a core business problem within a particular area.

 

And the result will be a perfect lasagne, ready for your users to devour and enjoy.

 


Digg!

Advertisements

Only a light touch will make the most of service orientation

January 26, 2009

The propaganda seems to suggest that service-oriented architecture (SOA) is a way of integrating all processes and solving all business problems.

 

Such thinking is not only unrealistic, it’s dangerous: IT managers must stop thinking of SOA as a golden bullet.

 

SOA can have a tremendous impact on business technology, but only if service-orientation is approached with a lighter touch that puts end-user priorities at the heart of IT decision-making.

 

Don’t think you can walk in to the finance director’s (FD) office and demand money for an SOA project. First, economic uncertainty equals constrained IT budgets.

 

Second, your FD is a businessman, not a technologist. The last think he or she wants to be confronted with is yet another series of TLAs (that’s ‘three letter acronyms’, for those not initiated with the art of techno-babble).

 

Return on investment has never been more important. So, find some facts and figures that prove SOA is not only worth the money, but that it will actually help your company make better use of existing assets.

 

It shouldn’t be too difficult. While SOA has traditionally been seen as innovative and leading edge, more and more firms across all sectors are beginning to look to the value created by service orientation.

 

Gartner recently found that 53 per cent of IT leaders are already using SOA in some part of their organisation. Another 25 per cent plan to invest in service orientation during the next 12 months

 

Previously slow-to-catch-on sectors, such as insurance, are now looking long and hard at SOA. But some industries – notably manufacturing, agriculture and mining – are still dawdling.

 

Gartner suggests the two major reasons for not pursuing SOA are a lack of skills and no viable business case. But a lighter touch that concentrates on key processes – notably shared services and governance processes – should help you develop an ROI and help you develop necessary in-house expertise.

 

Being careful isn’t necessarily wrong. In fact, it’s probably right – and watching how other firms have developed SOA will help you create a successful business case.

 

But as you move into SOA through 2009, make sure you investigate how your carefully layered service-oriented projects will actually solve specific business problems. Avoid the big bang approach and you’re sure to find success.

 

Digg!

Why create portals when you can create a composite application?

January 21, 2009

Why create portals when you can create a composite application?

 

As I mentioned in an earlier post on usability issues, next generation portals will be unlikely to provide useful access to information unless IT managers take control of infrastructure issues.

 

By their inherent nature, portals require users to run multiple sessions on screen; each portal application requires a different connection to the back-end infrastructure.

 

If you’re running many applications in one portal, the strain on your network and hardware can be unbearable. So, why bother with portals?

 

Well, a well-implemented portal can help present essential information to essential users. But you have to get your approach right.

 

First things first – identify the process you want to solve. Whether it’s supply-chain management or customer service, recognise the information that will help drive increased business intelligence.

 

As you strive to create the right approach, don’t think of a portal as a bunch of separate applications that are best served by in-house storage assets.

 

If you do, the aforementioned information strain on your servers is likely to be unbearable. Instead, look to client side session management and keep data in the browser, rather than on the server (see my earlier blog posting in ‘Further reading’, below, for more details).

 

Then recognise that most of your users will have a small screen estate that cannot readily support four or five open windows.

 

Even if your back-end infrastructure can stand the strain of running simultaneous applications, your users’ eyes won’t – and as soon as key executives are straining to see detail, hopes for increased usability and high efficiency start to disappear.

 

Remember, that one process should mean one application. Pushing multiple sessions is not an intelligent way to provide clarity on your key business process.

 

Don’t think of your portal as a jigsaw, where small elements create a bigger and more effective whole. Instead, start with the process in-mind, and create a composite picture that includes all the functionality needed to solve your business concern through a single mashed-up application.

 

Analyst Gartner suggests mash-ups could emerge as an alternative to horizontal portals. I would go further and suggest that composite applications are the future of portals.

 

Giving users the power to create mash-ups through the browser will increase the effectiveness of your information push and reduce the strain on your servers.

 

 

Further reading:
https://dharmeshmistry.wordpress.com/2008/07/18/client-side-session-management/


Digg!

Finance firms must embrace end user development

January 7, 2009

The bad news for finance IT professionals just got worse. Already lumbered with a credit crisis and the prospect of deep job cuts, and now research suggests the financial services sector is the most hampered by an IT skills shortage.

According to the 2008 silicon.com Skills Survey (see further reading, below), 54 per cent of finance workers say their business has IT positions to fill. The results even compare unfavourably to other sectors that have significant gaps, such as government and retail.

And that’s not even the end of the problem. Programming is the most in-demand technology skills for finance firms, as bread and butter a prerequisite for running a successful IT department as up-to-date servers and PCs.

With economic conditions worsening by the day, finance firms will need to think quickly about how to meet demand for programming skills when IT spending remains under scrutiny.

The answer, with firms continuing to slash technology positions, is to adopt more innovative approaches. Basically, if line-of-business workers demand programming, they can do it themselves – and if you think that approach is unrealistic, think again.

Research from Manchester University indicates there will only be three million professional programmers, compared with as many as 90 million end user developers, in the US by 2012.

Allowing people that are not trained or employed as programmers to create solutions might sound beyond-blue-sky thinking, but the approach – otherwise known as EUD – is actually grounded and pragmatic.

The Manchester research indicates half of respondents not involved in an IT profession have already heard of EUD and 60 per cent believe the approach is preferable to waiting for the IT department to facilitate requests.

These non-IT workers have the right idea, because talented technology workers are likely to become an increasingly scarce commodity.

Analyst TowerGroup estimates 300,000 people could have lost their jobs in the financial services sector globally by the end of 2009, with as many as 25% drawn from IT professionals (see further reading).

Finance firms should avoid fretting because the Web 2.0 era is demonstrating how users can easily create collaborative web platforms, without being a talented code cutter.

Programmers might be thin on the ground but many finance staff are already adapt at end user programming – whether they know it, or not.

Embrace EUD and foster a sense of creativity among your users now. If you’re an executive at a cost-pressured finance firm, it could be the smartest – and most important – decision you make during the next twelve months.

 

Further reading:

2008 silicon.com Skills Survey

http://www.silicon.com/financialservices/0,3800010322,39362827,00.htm

IT job losses in the financial services sector

http://www.computerweekly.com/Articles/2008/12/08/233753/tens-of-thousands-of-it-jobs-at-risk-in-global-finance.htm

 
Digg!