More features means better software. Add this calculator. Add that tool. Give this option. Make this shiny(er). Update the layout. More features!
Oh, and buy this upgrade. (Or worse yet, let’s force it on you by crippling your older version!)
It seems the inclusion of a new feature is automatically a good thing. We hardly measure features against usefulness, and rarely do we stand it up to the test: Will this extra bloat be worth the usefulness we’re giving the user?
Software as a Service may actually kill feature creep. Then, finally, we’ll be able to tell if software is appreciated by seeing users vote with their wallets. If they keep paying, you’re on to something! Why change anything at all?