The accidental structures of IT

IT architecture, organisation and decision making are a by-product of engineering necessity, not a conscious design to best serve the needs of business.

Last week I introduced a journey to show you a heretical view of IT in large businesses.

I want to start the journey by examining the structures we use in IT: the structure of technology, the structure of business applications, our organisational structure, and how we structure management decision making.

Historically, technology has been expensive and difficult to work with. We have developed ingenious ways of dealing with this:

• Different types of hardware to cope with different processing needs.
• Operating systems to simplify development and overcome differences between hardware types.
• High-level programming languages to overcome the difficulties of working with machine code and assembler.
• File systems and databases to manage data storage.
• Transaction processing systems and middleware to co-ordinate data processing and movement.
• Network protocols to simplify and standardise connectivity.
• Components and shareable layers, to reuse our investment and hard work.

On top of these technical structures we have built business applications. Where possible, we have built these into components and shared layers too.

These elements - hardware, operating system, programming language, database, middleware, network, components, layers - have supported the growth of IT and the spread of IT into every part of business.

These elements dictate the organisational structures we use. We split IT work into a business-technology continuum: business analysts, system analysts, programmers, system programmers. In another dimension we split our IT organisations by type of technology: different skills and different teams for different operating systems, programming languages, databases, architectural layers and business applications.

The complexities of IT also require a special approach to management and decision making:

• IT projects require the co-ordination of multiple technical components and resources, which makes IT project management as complicated as any type of project management.
• Shared hardware, databases, middleware and application systems mean that projects need to balance the needs of multiple stakeholders.
• Because we have an army of specialists, we need to schedule work carefully to avoid resourcing problems.
• Because we have built technology layers, we need to run IT-centric projects to manage upgrades and technology refreshes, and we need to balance the need for business and technology changes.
• The overall planning and monitoring of IT work can be a nightmare, and we need sophisticated approaches like project portfolio management to get a grip on the work.

IT is a complicated business. We have created technical, organisational and management structures to deal with this complexity. But most of the complexity can be traced back to our ways of dealing with the cost and difficulty of working with technology. We did not start with a blank sheet and ask "What's the best way of doing IT in large businesses?" Our approach and our structures have evolved through time as a response to the problems we have faced. They are a by-product of engineering necessity, in a sense just an accident of history.

Hopefully you are with me so far on our journey. Next week I will cover how IT structures create an alternative reality that has no meaning outside the IT organisation.

© Copyright 2007 Minimal IT Ltd. See the Minimal IT website for the original newsletter and copyright information.

Journey to the sixth circle of hell

There is an alternative view of IT that challenges nearly everything that we believe. Although we may not like its conclusions, it is difficult to see where this alternative view is wrong.

In Dante's Inferno the sixth circle of hell is reserved for heretics, those who preach against the teachings of the church.

Over the past few years I have been taking a journey through the ideas of IT in large businesses. I have found a great heresy, and I want to take you on the same journey so that you can see the heresy too.

To give you an idea of just how big this heresy is, here are some of its teachings:

• IT departments should not be involved in business change.
• Enterprise architecture and project portfolio management undermine the use of IT in the organisation.
• Technical specialists - database administrators, security administrators, system programmers - will largely disappear.
• The need for analysis, programming and testing will diminish.
• Architectural solutions such as data warehousing, service oriented architecture and business process management will fade away.
• IT project managers will have less influence.
• IT departments may decline to little more than running email.
• But in the end, there will be a huge growth in the use of IT by business and in the value they gain from it.

Before you burn me at the stake for heresy, let me defend myself. This journey is a hypothesis, a mental model for interpreting the state of IT and for suggesting the best way forward. IT is still a young subject, and we need hypotheses to explore and improve our ideas. This hypothesis may well be wrong.

But even though it is just a hypothesis, I can not see where the journey takes a wrong turn. To me, the interpretation and way forward are compelling. I want to take you on the same journey, so that you can judge for yourself.

You may not like where the journey takes us. I am not sure that I do. But do not just argue about the destination. Think about the journey: does it take a wrong turn, is the reasoning correct? Because if it takes no wrong turns, if the interpretation is correct, we have to accept the destination, however challenging and uncomfortable that may be.

So over the next few weeks I am going to carefully lead you along the journey to see this great heresy. If I am wrong, then you will be better armed to argue against this heresy in the future. If I am right, then your view of IT will be turned on its head, and much of what you hold dear about IT will be shattered. But in the end, the heresy is not bleak. Its view of the future is much more interesting, much more valuable, and much less constrained.

I will start the journey next week, by covering how our IT architecture, organisation and management are an accident of engineering history, rather than the most effective response to the problems of business IT.

© Copyright 2007 Minimal IT Ltd. See the Minimal IT website for the original newsletter and copyright information.

Portable applications - interim report

It is practical to run a PC environment from a flash drive. Additional work is required to make the environment really flexible and easy to use.

A few weeks ago I wrote about my experiment with portable applications. I thought I would give an update on progress so far.

To recap, I am trying to recreate my PC environment on a USB flash drive. I want to transfer my work between Windows XP and Vista PCs easily, and make sure that I can continue my work immediately if one PC fails.

I have now largely managed to recreate my environment on a flash drive. My conclusions so far are:

• PortableApps.com provides a strong and growing collection of office products and utilities that run directly from a flash drive.
• The PortableApps version of Thunderbird, run with the Lightening calendar plug-in, provides a usable alternative to Microsoft Outlook. Because I want to be able to access and send email anywhere, I have had to find an alternative to my ISP's outgoing email service because that is only available when connected directly to them.
• Most of the development tools that I use are easy to run from a portable device because they have not been designed to be Windows-specific.
• Flash drives have different speed characteristics than disk. I have not noticed the speed differences for reading data or for writing a small number of large files. However, writing a large number of files to the flash drive is very slow indeed.
• Because of the speed characteristics, some software runs very slowly from the flash drive. I have got around this by using a version installed on the PC where available, and relying on the flash drive version only when on a borrowed PC.
• For backup, it is relatively easy to copy an entire flash drive, compress it, and write it to a CD. Recovery is simple because the entire environment is just a bunch of files. There are no registry entries or file permissions to worry about.
• Some software does not run properly from a flash drive. I have copied the installation files for these to the flash drive so that I can install the software as required.
• I found it rather unnerving moving to a portable environment. Files and programs are not in their familiar places, and I have to be paranoid about not losing the flash drive.

I still have more work to do.

• I want to recreate lost shortcuts, file associations and startup programs. This would then give me the best of both worlds - a PC that feels like it's my PC, but with all the flexibility of portable applications.
• The environment still assumes fixed locations for software and data. To make it more flexible, I want to restructure the environment into bundles of functionality and data that can be moved round. This would, for example, let me simply copy a directory to the hard drive to speed up software that does not run well from a flash drive. It would also let me split the environment over many devices - for example keeping little-used data on a different device.
• I want to create a more efficient backup, using incremental backup or synchronisation.

I will let you know as I progress.

© Copyright 2007 Minimal IT Ltd. See the Minimal IT website for the original newsletter and copyright information.

The win-win-win scenario

IT has a rich ecosystem of service providers. They are critical to bringing new products to market.

If you developed a product that really helped the management of large-scale corporate IT, and that reduced the need for expensive consultants, who would you try selling it to?

The obvious answer is IT managers in large organisations because they gain benefit from the product. You would not think of service providers because they do not directly gain benefit and could even lose business.

We tried just that with system governance. But the feedback we got back was "Good idea, not my priority right now." IT managers have to focus on the short term pressures of project and service delivery.

We tried again, stressing how our methods also help in the short term. People still thought it was a good idea, but they are so focussed on their current work that they scarcely have time to stop and think of something broader.

We thought long and hard, and turned our marketing on its head. We decided to refocus our marketing effort on the businesses that provide IT services into large companies.

Some of our motivation was simply practical. Working with service providers gives us more contacts and a larger sales force. Service providers let us deliver into many more businesses. Working with service providers lets us concentrate on what we do best: the core methods, tools and materials, and promoting system governance. For the service providers, it provides opportunities for new business and a new competitive differentiator.

As well as being tactically convenient, working with service providers makes sense to the end customers, the IT managers in large organisations. They like to work with existing providers. They look to service providers for new ideas and added value. They want the control, quality assurance and cost effectiveness that system governance provides, and having it bundled with other services makes it easy for them to adopt. Here are some examples of added value bundling that we have been discussing:

• Outsourcing providers can provide a quality control framework for outsourced systems.
• Business process transformation specialists can reduce problems with existing systems, and provide a framework for identifying and renovating systems that need to be enabled to work with new processes.
• Audit specialists can extend their offer from controls and security to a broader assessment of IT best practice.
• IT strategy consultants can include estimates of return for preventative maintenance, as well as the business case for new projects.
• Hardware and system software providers can contribute to the proactive management of IT, and become more than commodity suppliers.
• Project management specialists can manage long-term solution risks as well as short-term project delivery risks.

Our change of direction has been very successful. We have generated so much interest from the first few companies we talked to that we have had to slow down our marketing efforts.

In hindsight, we now realise that we were initially too dismissive of service providers. Working with and through providers gives our target customers easier ways to gain the benefits of system governance, as well as making good business sense to us and the providers. Working with service providers is a win-win-win scenario.

© Copyright 2007 Minimal IT Ltd. See the Minimal IT website for the original newsletter and copyright information.