This entry was posted on Saturday, 11th October, 2008 at 9:48 am and is filed under quality. You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.
Dennis, a VBA developer who knew nothing about .NET would (probably) not try to deploy .NET. A VBA (or more likley) VB developer who didn’t know much about Excel would be just as culpable, but it seems these guys tried .NET, that didn’t work, so they tried VBA to patch it, which brought them back to where they started functionally, but probably much more complex and less robust operationally.
To me this is clearly a case of a solution being designed and implemented without any consultation with the users. The fallback onto Excel and VBA just emphasises what a suitable environment it provides for quick fixes and prototyping. Now that everyone has been through the process once and knows what they’re dealing with in terms of business process and systems, my guess is that a decent system could be built using any number of technologies, including .NET.
If you still need to email me, try again. Looks like Pipex have completely deleted/disabled my mailbox for some reason. And they don’t provide support at weekends. I knew there would be trouble the moment they were taken over by Tiscali, and I’ve had nothing but grief since then. 100% uptime has become 95%, for example.
I’ve re-routed all of my mail to go via my business hosting supplier now, so everything’s up and running again.
Actually, the problem stems from the consultant (could be any virtually any technology) assuming that everything should be a website or web service. Do not confuse a website with Excel or vice versa or the power of a desktop application with a web application. The error was assuming that a web interface would replace ALL of the functionality of Excel, and instead it ended up going full circle back to Excel and even. There is a reason that desktop applications are still relevant in the world of the web.
Personally, I think Bert’s point is wildly off the mark. The problem is misusing technologies or rather not knowing the boundaries of the technologies, as well as a lack of knowledge about how the end user will use the application.
Simon said … You’re right too that desktop apps are still relevant – I wish more people got that, I think our numbers are in decline.
Back in 1996 I was working for a major UK financial institution, one of the few left standing, and I was PM for a credit risk system. We designed a typical 3 tier client-server architecture, with a centralised database. When the design went to the architecture teams I got a lot of flak for 3 things, first for not coming up with a devolved database (has anyone ever successfully implemented such a beast?), second for not using Lotus Notes (can you believe that one), and finally (and the point) for coing up with client-server and not designing a web based app. I won the arguments in the end, but I had to battle very hard, and it took 6 months (which pushed us uup against the wire when it came to the implementation date as the clients weren’t interested in internal arguments) and was incredibly draining.
That was 12 years ago, there was a modicum of pragmatism in later years at that place, but it was clear that the architecture theorists were only at bay, they had not been defeated.