Food for Thought: Access Used Properly

2

September 21, 2009 by Mike Hillwig

The DBA community is going to have a fit with this, but I wonder if Access is as bad of an application as we make it out to be. I know I made a case for giving Access a seat at the table in the past, and I don’t want people to think I’m a fan. Not. Even. Close.

I’m not a fan of Microsoft Access. I avoid it as much as possible. But as someone who has a major voice in my company’s technology decisions, I absolutely must be fair.

In my experience with Access applications, it’s always been “developed” by someone who wasn’t a professional application developer. My previous company had a “Master Schedule” that pulled data from their finance/accounting system. It was written by the corporate controller. My current company has an application that pulls testing data from our manufacturing QA system. It was written by a test engineer in our manufacturing unit. Notice that neither of these were written by database professionals.

What if these HAD been written by actual developers. What if the data was properly normalized. What if the right data types were used? I doubt it will ever be as good as SQL server. But could we build a good application?

I’m not advocating it. But it might be worth some investigation.

  • In my old job, I created an Access application (VBA) that handled “employee errors” on the production. What the application did is keep records on what types of error employees have and the frequency of those errors. It was an application to simply identify those who needs more training on particular tasks, and also to identify what types of training the company needed to implement.

    The management liked my application. I received lots of congratulations.

    After a month of implementation, they asked our senior developer to convert my application into a stand-alone application. I was not sure of my feeling but I knew I was kind of hurt.

    The main point of the incident is companies do not really take MS Access as a serious contender when it comes to meeting enterprise needs. One of the biggest reason is its limited capabilities.

  • Rob good question. I’ve been cuoirus about that myself, but since it costs money to ramp up a new server and do performance testing, I haven’t bothered. I’ve been waiting until I see word from them that there’s a better-performing storage subsystem available, or that they’ve improved it somehow. Have you seen anything to that effect?