Recently in an article (Dutch) in Dutch publication "Webwereld", Drupal advocate and developer Bèr Kessels stated that while Drupal is an awesome CMS, it is not a good fit for government websites and other big projects. An interesting statement for someone from the Drupal camp to make. While Bèr has some valid points, his statements are a bit too generic for my liking.
I am not a fan of Drupal. Nor am I a big fan of Joomla. But both have their place within the PHP and CMS ecosystem. And there are many situations where clearly one of these tools is the best for the job. For big custom projects, obviously my choice would be either symfony (or Symfony2), or Zend Framework. Or well, any web application framework for that matter.
And this is also a (valid) point that Bèr makes in his article. For big projects, web application frameworks are usually a much better fit. They are made to fit the needs of developers, not end users, making them much easier to use for developers of big custom projects. A very valid point, if you ask me.
Less valid already is the point he is trying to make that you need to spend a lot of time to remove the functionality that Drupal offers out of the box. Sure, Drupal offers a lot of functionality out of the box that most projects might not need. Most of it can easily be disabled with a couple of clicks or a change in the configuration. Even if you'd need to spend a week on this (which is already way too much I think), given we're talking about huge custom projects (think 3-6 month development time at least), that's still not a lot of time. You might have to spend more time getting a new project in a web application project set up to a point that is comparable to the standard Drupal installation.
That is all good and fine though. I can understand both arguments, whether I can agree with them or not. Bèr concludes his article with "The right tool, for the right job". The funny thing is, though, that his article is actually advocating the opposite. He is claiming that Drupal is not a good fit for government websites, period. And that just strikes me as odd. He states you should use the right tool for the right job, but then explicitly excludes one of the available tools for a huge selection of jobs.
To Bèr and any other developer, I would like to say: Look at what you need to build, and then make your consideration of the tools carefully. Combine tools if you need to, look at the project, look at your project team, and then decide what you're going to do. That is the way any project should be approached, whether your customer is government, a bank, a startup or the little village pet shop.