Back in the days when I lived in South Africa, I had a good friend, a quite young Sangoma, he told me an ancient wisdom (Though I guess he actually made it up himself) "A good carpenter never blames his tools for a broken table". The explanation is twofold: A good carpenter does not have bad tools, a good carpenter is a good carpenter because he knows how to choose his tools. But if, with those good tools, his table still breaks, he will blame himself, either for trying to build a table that is too hard for him to build, or for simply not doing his job well. That is what made him a good carpenter: He leared from his mistakes. In a recent discussion on a Drupal mailinglist, about a post where Drupal is presented as "not the best choice for all cases", the fingers qickly pointed at the Drupal page 'is Drupal right for you'. Drupal is presenting itself as too good, to which I agree, I brought this up a while ago already.
Someone claimed, in this discussion, that «Highlighting weaknesses is not very useful as it's frequently the limitations of the users ability to use a tool that are the real weakness.»This is my point exactly, but there is one point to mind: marketing. uggh.
In Europe we have some laws to protect you from 'too much marketing'. If I sell a client a site that (my promise) will generate millions of € in banner clicks, yet I deliver a site that can never ever do that, most courts will grant that client most of the claims. But worse, and therefore a bigger showstopper for marketeers-gone-wild, is the fact that angry clients, in big numbers, will bring your sales down. Commercial companies need to find a balance between their promises and the reality.
OSS needs not bother about both, does it? Drupal can claim to be the cure for all war, or to be your golden goose. In the end, that might backfire, but we don't yet really know that. After all, no-one got a real chance to prove that all the claims Linux adepts make, that it is much more stable, much more secure, much better architectured then another OS, are untrue. Some people jokingly state 'if you are not happy with Drupal we will happily refund you'. No-one got a real proof that the claims made by some of the 'holy OSS believers' are true, when they state that OSS is far better then the alternative, other then some general consensus and a growing (happy) userbase.
Back to Drupal: I beleive Drupal is not ready for doing everything. Drupal can do a big lot. It can probably do most, in shortest time, with biggest security, of all CMSes we know in 2006, yet that is far from 'all'. Joomla does certain 'stuff' much better, maybe not under the hood, but in the end, it is 'better' for certain projects. The fact that, just like Joomla!, much more people create websites with Wordpress, indicates that Wordpress is a better choice for certain other projects too. That wordpress can be deployed a lot better in certain projects.
Back to the very beginning: Drupal can do startups a lot of harm. Sure. And so can do Word, or even Textmate. The problem is, in the end, why people chose to use it. If people choose some tool for wrong reasons, then it will do them harm. And that might backfire on the community. Its up to us to tell people when Drupal is not a good choice. And after that we can point out all the bad carpenters one by one, and tell them how bad they are, and why are bad carpenters.