Jon Collins, Views from the bridge

Another day, another keynote. This time it’s Microsoft TechEd in New Orleans, and Bob Muglia is on stage being all thrilled about all those new and improved aspects of IT. Three weeks ago it was Ajei Gopal at CA World. Today, elsewhere in the US IBM is kicking off its Rational conference, and I believe from my Twitter feed that SAS has a gig, somewhere.

It’s important to remember that most normal people don’t spend their lives attending IT conference keynotes. the real attendees at such conferences will have paid through the nose to come, and likely there may have been a little lottery between team mates about who should go. I know that back in my days as a punter rather than an analyst, attending conferences was very much a one-off, not just for cost reasons but also because I would have been too busy actually running stuff. (As an aside, it’s one of the reasons analysts exist)

So, most people in the rather humid hall I am sitting in right now will not have the luxury of comparison between multiple keynotes – and indeed (based on the “time ” thing) may not have had the luxury to stop and think about some of the things that are being presented. But, for better or worse, I do – and so it is that I inevitably start to think about how this keynote compares with those in the current batch, which appear like waypoints stretching back through the history of conference time.

Now I wouldn’t be so trite as to give marks out of ten for individual keynote presenters – though I am reminded of some of those dodgy talent competitions in the Seventies. Actually, even better than that, I wonder what Simon Cowell et al would make of Messrs Muglia and Gopal. The hand would slice, the brow would furrow and the head would shake, before the flabbergastingly obvious, albeit reasonably accurate commentary. “Haven’t we heard this before?” he would ask. Well, perhaps, but Simon never did understand the idea of paradigm shifts.

Strip away the repetition, the pseudo-excitement and exec cameos, and there is generally some good, interesting stuff in the average keynote. This one is no exception – new product announcements, capabilities that are now probable (whereas in previous keynotes they were just possible), yadda yadda. Right now for example, data visualisation is having its time in the sun – and to be fair Microsoft has a good story to tell around its SQL Server tools.

However, one area that’s pretty fundamental to IT never seems to get a mention at keynotes. All these new-and-improved capabilities are based on a core premise that they can exist in some kind of isolation from whatever else is in the IT environment. The irony is of course, that they never can – even small companies have existing technology investments, and anything new will have to work with all that old, and seemingly inferior stuff. This means both an integration impact to get new working with old, and a migration impact as the lucky elements of IT get rejuvenated and or replaced. (Hat tip to Roy Illsley for mentioning licensing as well, just before I posted this.)

Even if it were possible to adopt a green-field approach, the sheer complexity of IT very quickly comes back to bite anyone (particularly in marketing) who underestimates it. I was in a conversation a few weeks ago with a systems engineer at the Symantec Vision conference (yes, there was a keynote there as well, from Enrique Salem). The engineer went through a quick summary of the complications of failing over a production environment, in terms of servers, networking and storage, the level of hard coding still required, and the potential for error if any part of the systems were even slightly different. It was a welcome reminder of just how complex things really are, in most, if not all IT environments today.

I won’t go on as the keynote is due to finish soon. But as well as “star quality” or whatever else we feel should be on the keynote scorecard, let’s have a row for “ability to deliver in existing environments”. Perhaps it is the absence of this as a metric which has led to so many great ideas, presented at keynotes past, turn into dust – or worse, be rolled out again a few years later under a new terminological banner. IT is complex, and will remain so however hard people try to present it as something that is becoming simpler. And the sooner industry figureheads can take this on board and talk about it accordingly, the better.



Leave a Reply