Revisiting content management: Just build what you need!

Back in 2001 when I got laid-off from my job and started learning what this new .Net thing was all about, my first attempt at building something was a CMS that could do "everything." When I look back at it, well, it pretty much did do everything. You could drop in a UserControl that did whatever you need. It was like Sharepoint before Sharepoint.

And it took a long time to get it right. It has been reliably running CoasterBuzz for nearly five years, but it never needed to be what it was. What do you suppose was in there 95% of the time? Just a bunch of HTML.

So I started to hack out something new, given my vastly more developed skill set, and it's a fraction of the code, mostly the same functionality, and took almost no time to build. Looking back, I can see that I've spent so much of my time trying to account for every scenario I can think of instead of just addressing the one sitting right in front of me.

I can attribute this mindset in part because I've always wanted to build stuff that other people can use, which was largely a mistake because I never really did sell much of it. Building a product for everyone often satisfies no one, and I should know that more than anyone since I've always been the primary user of the stuff I've built.

We get so wrapped up in frameworks and "solutions" and other nonsense and it does little more than bog you down. I hate feeling like that. You can build things in a "smart" way without resorting to curing cancer.

It's liberating, and a lot easier to maintain. 

5 Comments

  • So true, so true. I've worked at 6 different companies with 6 different custom CMSs and I think in 5 out of the 6 cases the CMS was more trouble than it was worth.

  • Nice post. In 2001, it made sense to build your own CMS for ASP.NET. But in 2007, it doesn't make sense in most cases:

    Considering the available ready-to-use CMS systems and their price, you can hardly justify the development costs of even the simplest CMS system from scratch.

    Yes, you will hardly find a CMS that has all the stuff you can think of. The point is to choose an open CMS with most required features that allows you to modify the default functionality and add your own code when necessary.

  • Actually, I don't agree with that at all. The .NET CMS systems out there are almost universally bloated and don't scale. My point was that building what you need, and only what you need, is the best path to take.

  • Actually, I don't agree with that at all. The .NET CMS systems out there are almost universally bloated and don't scale. My point was that building what you need, and only what you need, is the best path to take.

  • I think this applies to any app, not just CMS.

Comments have been disabled for this content.