This is an odd announcement considering the last I heard, 37signals would prevent new signups to phased out services, not abandon them entirely [0]. I can't imagine that Breeze was all that resource intensive to maintain, but I don't really know what goes into such a service.

[0]http://thisweekin.com/thisweekin-startups/david-h-hansson-pa...

...

Jason Calacanis: Yeah. I remember Back Pack.

David: We stopped offering to new people. Ta-Da List and Writeboard. The critical thing, I feel like, we did right was to say, “No new people could sign up for this. Everybody who’s still on it, we’re still going to support it.” We’re just going to continue to support it.

Jason C: So what was the thinking there? It doesn’t cost us that much.

David: It doesn’t cost that much to support it. What’s really expensive in running these services is you have to keep scale as your growing. You have to buy new hardware and manage the ops. You have to deal with the bugs that invariably will spill out from launching new features and all that stuff. If you stop launching new features and you stop growing then it becomes a non issue in many ways.

Jason C: Set it and forget it.

David: Exactly. Set it and forget it. It’s going to be there till… I don’t necessarily want to say the last guy closes the lights or whatever… but there’s no imperative for us to close it down. I think it’s a much more humane way to deal with…

This conversation is in direct reference to Google's decision and management of shutting down Reader