Andy, I agree with your opinion thoroughly. I also 'grew up' with access... though I did mine in 95/97. So perhaps it's just familiarity.

Access is an excellent tool when used correctly. It's CHEAP. Let's start there. It installs (or did, I may need a refresher) for around $250/seat. Considering once you're over 5 or more users in access you're reaching it's reasonable locking control limits, that's just fine. It comes with office, another plus. It uses the similar menus, toolbars, etc that our users are used to.

Next, it's got a built in form and reports structure, and the reports are WYSIWYG. Users can goof around a bit and get exactly what they want, instead of having to rely on developers to make it "Just So". It also encourages the idea of dev and production in our user base, as they've become 'devs', of a form.

Finally, there is plenty of juice in Access as long as you're not overwhelming it with concurrency or data volume. There's a concept that Access is 'training wheels' for design. Well, YEAH. It is. It's meant to be end user friendly and thus has a lot of tools for those end users to be able to get their hands dirty. Get past those wizards, however, and there's plenty of firepower under the hood of the application. Self contained, easily distributed, security easily controlled via file access by AD, and user friendly.

It's not the application though, that usually gets to folks. It's that some numbskull in some department somewhere starts up a secondary system to 'work around that difficult corporate app' so they can move quickly, stores half the data off in some now 'mission critical' application, and it's now our emergency to bail their asses out of alligators. Get a good, knowledgeable, and integrated with IT MS Access person on a staff to work with those small departments for their one off needs, and I believe you'd find much of the disdain to start to fall away.

That person can then know when a department is avoiding existing systems, and thus get it up the queue to the systems developers, or actually build them their one off tool in half, if not less, of the time a 'full blown' application can take. Particularly if they just need to catalog their sports card collection when the rest of the company is concerned with making bats and helmets.