To what degree to you think Rails is suited for the enterprise? Does it matter?
Rails is much more useful for the enterprise than The Enterprise. I believe the latter to be mostly about appearances, "nobody-ever-got- fired-for-X", late, expensive software that nobody loves. The Enterprise will probably never really like Rails because there are no expense accounts, no golf dates, no month-long pandering by sales agents, and none of the big-firm prestige involved.

Now the enterprise, people in need of software to do real, or even big, business will likely learn to love Rails just as much as the small-time players. It's simply good business to spend less money developing great software.

Do you think Rails could propel Ruby to being as well established as something like PHP?
It's all about different needs. Rails is never going to be a good fit for adding a few pieces of dynamic content to your home page. But yes, I do believe that Rails will help Ruby compete for those types of projects where the quality of the programming language matters.

Lots of programmers stay with PHP for too long in my opinion. They learned programming doing a few simple PHP scripts and simply kept on using it as their appetite for achievement grew. Many probably don't realise that programming could be substantially different, that some of the hurt could go away with a change of scenery.

If a developer found they could get applications out the door as quickly with something like Zope as with Rails, is there any reason why they should give up the complicated stuff Zope can do in favour of the simplicity of Rails?
Unlike Zope, Rails is not a big believer in high-level components. We don't try to abstract business logic. So there are no standardised authorisation or access control schemes. There's no content database. There's none of these specialisations because I don't believe that they work. Once a high-level component becomes big enough to be interesting, it'll take more work to configure it than it would take to build just what you needed from scratch.

So Rails tries to just work on the stuff underneath the split between infrastructure and business logic — the stuff where most people need the same solution most of the time. Where it's possible to be opinionated without restricting the kinds of applications you can make.

To what extent is performance still a concern with Rails? Do you see it as an ongoing issue or just a temporary pre-1.0 issue?
For the vast majority of applications, Rails is more than fast enough. We have applications done in Rails that handle millions of daily dynamic requests on very reasonable hardware. Rails simply scales out fantastically well.

And since the early days, we've had the good help of performance specialists like Stefan Kaes who has been improving things immensely. Today, it's simply not something that's noteworthy of concern for a lot of people.

What has been the response to the 1.0 release? It was meant to be a kind of seal of quality, have you seen people getting interested simply because of the 1.0 label?
Definitely. Rails 1.0 is a much easier sell to management and clients than Rails 0.14.3 was. No, it doesn't make sense from a technical perspective, but perception is reality. So we've certainly seen people being thrilled about finally seeing a 1.0 release out there. Truth be told, we should have realised this long ago and released 1.0 a year before we did. Then Rails 2.0 could now have been out and that would arguably have helped perception even more.

Can you talk about upcoming features, for example SwitchTower and The Conductor? Will these be coming along in 1.1?
Development for Rails itself is in some senses slowing down. We're poking at the limits of what most people need most of the time underneath the infrastructure cut with Rails. So instead of embellishing until you can spot the kitchen sink, we've turned our attention to the tools that can support a Rails-approach to thinking in other areas than just the framework.

The first example of this is SwitchTower. It was created by fellow 37signals programmer Jamis Buck to help us manage the deployment of Basecamp, Backpack, and our other applications once we moved from a one-server setup to a 10-server cluster. It encapsulates the process of pushing new versions of your application live without dropping a single request in the process and without requiring downtime, for most operations.

So SwitchTower allows us to push a new version of Basecamp 10 times a day leaving the users none the wiser, except for when they see new features and bugs fixed. It's incredibly important part of remaining sane as your application outgrows its first box and you need to scale out.

The Conductor is the next tool we've been working on for a bit. It's an application to help you develop applications. That's about as much that makes sense to share about it at this point. We'll be getting real with that one soon enough.

What kind of time scale are we looking at for 1.1 anyway?
Rails 1.1 is targeted for release fairly soon, whenever that comes around to be.