dotcom boom – Adam Bager

I am distrustful of bandwagons, and rarely board them as they rattle noisily by. It’s not that I’m against enthusiasm, or novelty, it’s just that bandwagons are often driven with irrational exuberance, and very often they crash. I prefer the slower vehicles that come along behind. They set out in the same direction, but with more care and circumspection, and they reach their destination more reliably.

bandwagon

I remember the bandwagon of the Dotcom boom – and the bust that followed it. Whilst there’s no doubt that the internet changed our lives, I’d rather call it accelerated evolution than revolution. Its victims were more often its most ardent supporters, hell bent on delightful ideas that lacked even a modicum of commercial good sense or realism, than the ancien regime.

The advent of the internet was evolution. It simply extended what we already had, and what we already did as business IT consultants. It didn’t, despite the fears of some, invalidate what we already knew. Business systems nestle as comfortably in the internet as they previously did in their more confined circumstances. The basic problems of system integration, of manufacturing, retail, services, accounting, distribution, CRM, and the rest, are of the same type as before, as complex as ever and we, who understand them, are as valuable as ever.

During the feverish years of the dotcom boom one of my colleagues told me that if we weren’t immediately reborn as ‘Dot LLP’ (instead of LLP Group) we’d be annihilated within six months. I laughed. We didn’t, and we’re still going strong. We learned some of the new dotcom tricks, and we go on learning.

thecloud

The Cloud, I fear, is currently another bandwagon. We won’t be boarding it with too much haste, even if its direction is the right one. We’ll be waiting for the slower train that won’t go off the rails.

The Cloud is an excellent idea. It comes in many forms. As those who tout it say, it allows businesses, even software authors such as we are (systems@work) to concentrate on what we do best. The business of managing IT infrastructure, handling communications, backups, performance, security, and so on, isn’t the business that most of us are in. It’s not our field of battle, so better to leave it to the experts. No need to employ specialists if others can do the job at a competitive price and relieve us of unproductive anxiety. We must concentrate on our core business activities.

All of that is true, and if we could all dump our systems onto rented hardware at reasonable cost, why not? Sometimes it’s possible and the right thing to do.

But that’s not exactly what the Cloud is meant to be. Certainly not all it’s meant to be. It’s not just a matter of hosting the particular collection of business software that we’ve amassed and integrated, it’s also about using a standard piece of software in a ‘multi-tenanted’ environment – one size, one copy suits all. And if we use this piece of software for accounting, this piece of software for distribution, and that piece of software for manufacturing, it may be about using multiple standard pieces of software in a number of different Clouds. It’s about business software becoming a commodity that can be accesses as easily as water through a tap.

Many business worry about the security of their data. But these issues of security are solvable, even if many companies are reluctant to let the ‘professionals’ look after their data. The fact is that data are vulnerable wherever they’re located, whether in-house or hosted, and the issues of security can be solved or not as easily in one environment as the other.

It is the issue of ‘standard software’ and ‘integration’ that don’t yet fit perfectly well within the Cloud. Sometimes, if the purpose of a piece of software is such that it can stand alone and if it’s used without alteration (even if configured for a particular company’s purposes) the Cloud can be a good place to put it, but if standard software has been modified, or extended, or is integrated in complex ways with other pieces of software, and other databases, then making this work with a Cloud-based solution will be difficult, and with a ‘pure’ multi-tenant Cloud based solution (one copy of the software serving everyone) it will be well-nigh impossible. Ensuring the consistency and coherence of systems and databases that are in multiple environments that you do not fully control will be difficult.

As time goes by, new techniques for integration may make this task easier, but we we’re not yet at that destination.

So, I am cautious about Cloud-based offerings in the world of business IT. They may work for some, but for many they aren’t yet the right solution. What seems initially like a good idea founders when a business needs something special from a standard software offering, or some special way of interfacing systems, and many businesses find themselves trapped by the choice they’ve made if it’s located in the Cloud. They’ve exchanged one limitation – the anxiety of running their own infrastructure – for another – the anxiety that comes from being limited by someone else’s standard software and infrastructure.

Whilst the provision and management of infrastructure isn’t usually the basis of a company’s competitive edge, the business software on it, often is.

As a software author (systems@work) we’re cautious. We offer configured systems in a hosted environment, and this suits customers who don’t need any software modifications and who don’t need interfaces between Cloud-based and non-Cloud systems, or who need only simple ones. And we offer on-premise installation and full-blown integrations when they’re needed.

But for now, the Cloud isn’t always the answer and we won’t be betting the business on it.

You can design the most beautiful building in the world but it’s no good if it’s too difficult to build. Think of the Sydney Opera House. It was imaginatively and brilliantly conceived with little regard for how it might be built, and the final cost of construction was astronomically higher than the original budget allowed.

You can design a software system that reflects a set of business requirements completely, and as simply as logic permits, but it’s no good if you need a doctorate in mathematical logic to use it.

Perhaps you can even build a sophisticated legal argument too brilliant for a jury to grasp. And I’m sure a surgeon can design a surgical procedure that’s far too difficult for most surgeons to execute.

Theory can enable you to design something as perfect and as simple as it might possibly be, but only experience can tell you whether you need to simplify still further to make it usable. Consulting is an art, and although the best artists are often young iconoclasts with the most up-to-date knowledge of technology, consultants generally get better with experience, and it’s often the grey-haired veterans who are the best at designing something workable. They’ve come to know what is possible in the real world.

I remember the frightening years of the Dot.Com boom when those of us who possessed even a few grey hairs were thought of as past it. The IT world suddenly belonged to the young dotcommers who thought up brilliant things that our minds, dulled by too much experience, were incapable of. If you were over thirty, you probably couldn’t raise a penny in investment capital.

One of these brilliant things was a website called boo.com which was designed to sell clothes through the internet and which would be driven by software so brilliant that users would see what clothes might look like on their own bodies by entering their vital statistics and then rotating a graphical ‘model’. It doesn’t sound so difficult now, but back then, more than fifteen years ago, it was impossible. Bandwidth wouldn’t allow it and there wasn’t enough time to write good enough software. Boo.com got through its 135 million dollars of venture capital and failed spectacularly. The real world wasn’t good enough for the entrepreneurs’ ideas. Read about it in BooHoo.

Of all my own failings as a systems design consultant none is worse than my always attempting to build something that can do everything a client wants or needs. I’ve probably got wiser over the years, but I still try too hard to design some logic for every eventuality. It’s not that the logic I design isn’t right, but rather it’s sometimes too complex. It’s a general rule that as the logic of systems becomes more complicated, so the users’ understanding of it becomes weaker.

When something is more complicated there is not only more of it that can go wrong (software is never perfect) but more importantly, more user mistakes can be made with it, and such mistakes get ever harder to correct. Given that a consultant must eventually do a disappearing act, it’s better to leave a client with something simple and manageable.

Along with this wisdom of experience comes the skill of convincing a client that ‘keep it simple’ is a good guiding principle and that although you CAN do what the client wants, it wouldn’t be wise.

Think also of the Kalashnikov rifle – simple and pragmatic. In a life or death moment you want something that can’t go wrong in too many ways.

In the film, Lord of War, arms dealer Yuri Orlov comments:

Of all the weapons in the vast Soviet arsenal, nothing was more profitable than Avtomat Kalashnikova…. more commonly known as the AK-47, or Kalashnikov. It’s the world’s most popular assault rifle, a weapon all fighters love. An elegantly simple, 9 pound amalgamation of forged steel and plywood. It doesn’t break, jam, or overheat. It’ll shoot whether it’s covered in mud or filled with sand. It’s so easy, even a child can use it – and they do.

So, one of the secrets of good design is to design something that can actually work, can actually be used. Don’t let theoretical fancy lead you into the realms of the impossible, however complete, however beautiful your idea.

Adam Bager

Occasional thoughts on IT, Business and Other Matters

Tag Archives: dotcom boom

Clouds – Nimbus, 9, or Cuckoo?

The Art of Consulting – Designing (Pragmatism)