Lotus 1,2,3 Will Never Happen Again
While I was just in a meeting, I was mulling over what I was writing about below and thinking to myself that an application like Lotus 1,2,3 - a spreadsheet - will never be created again. Or at least, it'll never be popular again.
Why? Because it's so free-form. It's a blank slate. The users add the logic in order to make it work... there's no entry point and no wizards. In fact, I'd say the majority of people still using Excel use Templates more than they just start with a blank sheet and start pumping in numbers.
I'm considering this from two angles: First the user and secondly the developer. The user has gotten used to a document model or a form-based model when it comes to using their computer. They either enter info in a document like Word, or read the document like on the web or Acrobat, OR they use a form or a Wizard to do their work. You could break down almost all modern applications as Documents or Forms: Microsoft Money: Forms. Access: Forms. Powerpoint, Frontpage: Document. Flash: err. not sure.. Forms? Documents? Anyways, rarely are "normal" users going through and entering information in a free-form, create your own logic sort of way. Users expect the logic to be part of the application (again, like in Microsoft Money), that's what they're paying for.
The second side of this is from the developer's side. I'm a developer. I work every day doing programming - expressing logic as code so that the computer can process data in some way - and I NEVER do anything even romotely like a spreadsheet. I use widgets, and even then only rarely. I use the web mostly. I'm a programmer, but I'm levels away from doing any real "new" User Interface. I put together widgets that someone else has come up with and the industry has standardized on. I'm not drawing my own pixels on the screen and listening for specific user-events: I'm telling widgets where to go and listening for their reactions.
So, my point to all this is that GUIs have stopped innovating. They're forms or they're documents. And from what I've seen - except for very specialized applications, a document model works best for most items. My favorite - and I've said before, I'm biased - is Lotus Notes combination of documents AND forms which solve 99% of the things that most users want to do.