With the release of WordPress 3, there's a whole new set of discussions about Drupal vs. WordPress.
I'm going to try and explain my views on UX in Drupal, once again using the lego analogy.
First, I propose that you need a defined purpose to build great UX. WordPress focuses on blogging, and thus has an excellent blogging product.
With WP3, they've even integrated their version of multisite / Aegir out of the box. BuddyPress is an example of a product that sits on top of this base infrastructure. So now WP does blogs as well as multi-user social groups / communities.
In Drupal, we've optimized for a big box of lego. And there are a LOT of pieces in there. Pirate bits, space ship bits, viking bits. And lots of blocks that fit together pretty well if you're creative and pick through looking for ones that match in colour and so on.
I could talk about the WYSIWYG editor that ships with WP as an example. They took the base piece, customized it, and now have a rich text editor piece that fits exactly into the blogging product. It's not even a block! It can't get re-used easily with all these pirate bits! Sometimes, it's OK to plug in that perfectly shaped lego piece that isn't really a block at all.
But that's really hard to do if you don't have a defined purpose. Because something that is perfect for a large community site isn't going to be great for running an intranet. And an open everything access & permissions model is great for the former, and dangerous for the latter.
So, if we want to improve Drupal UX, we can't just take out a big UX brush and improve everything. There are certainly base building blocks and flows that we can improve, but on top of that, it will be need to be optimized for different use cases.
Just over a year ago, I proposed that Drupal could be a great social community out of the box (as in, core install profile). I'm not so sure anymore, but I think we should continue to have the discussion "What should Drupal do out of the box?" so that a) new users have a reason to use core to accomplish a task beyond "it can build anything" and b) we can experiment with optimizing UX for that purpose. This will teach us lessons that can be re-used and rolled into all Drupal-based products.
And, in general, I would like to see more people trying to build nicely polished products that have a defined purpose: this is what will move us forward in the UX department. Dries has a discussion on the business model of products / distributions that is important - because we'll need to fund development over time.
Thanks to @mattfarina for kicking this loose from my brain on Twitter.