Application service providers may have failed to match early expectations, but the smart IT manager will note that they can solve hard problems, says Martin Veitch.
There are a lot of myths about application service providers but the biggest one is that they have failed and should now be filed alongside kipper ties in the what-on-earth-were-we-thinking-of department. That attitude doesn't make a lot of sense because ASPs are still here Ã, rumours of their death are overstated.
The problem has been that ASPs could not live up to their hype. The ASP Industry Consortium (Aspic) did a terrific job promoting its vision of a new model for IT that would see users connecting across the Internet to resources hosted in remote datacentres.
Aspic popularised the term ASP, it evangelised the concept, it organised conferences and working groups, and ASPs waited for lots of customers to sign up. And waited. And waited.
ASPs seemed a very attractive concept two years ago, and even one year ago. The IT industry was finally facing up to the fact that client/server was the god that failed, and that all the doubts about Wintel reliability were not going to go away.
It was also becoming clear that the skills shortage was biting, and people weren't knocking on the datacentre door proffering great CVs. And it was widely believed that IP was like Lily the Pink's medicinal compound: "most efficacious in every case", if not indeed the saviour of the human race.
OK, so ASP failed one crucial test Ã, it didn't begin with a lower-case 'e' like all the other miracle cures. But it did look like a smart way to build some predictability into budgets, scale up growing operations and get to market quickly. Necessity being mother to invention, ASPs looked pretty good to some desperate IT chiefs.
ASP fundamentals haven't changed, it's just that we have changed our attitude towards ASPs, and familiarity has bred contempt. The service provider model still has a lot going for it but it is emerging as more of a fallback than a full-on challenge to in-house IT.
If you're charged with deploying a project that has to be here yesterday but must come in at under the cost of the systems alone, ASPs are probably the only alternative there is.
If you don't know how big your dot-com is going to grow but you cannot afford the shiny server that looks like a refrigerator because the enterprise IT giants have stopped swapping them for equity, ASPs again offer a convenient solution.
Also, although many large organisations may not want to use their productivity applications from the Web, they may well want to take their deathly dull process tasks that way, such as CRM form-filling in call centres. And they may well be interested in expensive applications such as video conferencing if these applications are put through the ASP mincer so that they only pay for what they use.
ASPs will not go away overnight, and ASPs such as FutureLink, NetStore and Vistorm (formerly ESoft) have the funds to battle on for quite a while yet even if sales do continue to be sluggish.
What is worrying is that Aspic, despite all its drum-banging and flag-waving, has yet to come up with an assured way that a company will be able to retrieve its data and smoothly transfer to an alternative service if the ASP they use collapses.
When we have all finished examining the details of what ASPs can and cannot do, that doubt alone will probably mean that the ASP's role is that of the substitute Ã, a resource we can bring on when we need of a capability not already present in-house, or just out of plain desperation.
