Interface vs. functional specs

December 31, 2005

The good folk at 37 Signals have a very interesting (and well discussed) post on their blog.  I suppose being in product marketing, I am on the other side of the aisle, but as I commented, I am not sure I completely agree with them.  The “interface as spec” paradigm might work for certain kinds of application (like the ones that 37 Signals develop), but in the world of computational modeling that I live in, it is unlikely to work as well.  A good balance is required between interface design and the functional specifications.  The former is inherently engineering-biased and the latter is driven by marketing and often suffers from feature bloat.  While there is no one size fits all solution, in a customer-centric world, the best way to do so would be to build a prototype with some core functionality that is well understood and validate it with the customer base.  The design should be one that can be modified to fit in functionality that is requested but cannot be addressed by the original interface design.  In the end, the customer calls the shots and in my experience a large number (maybe not the majority but certainly enough) would choose functionality over a fancy interface as long as it is usable.  In the end, both engineering and marketing need to work together (with sufficient customer input) to ensure that a good balance between usability and functionality is maintained.

Technorati Tags: ,

The new blog

December 31, 2005

I am moving my old blog from blogspot to wordpress. I have had some trouble importing my old posts over so for now they are going to stay there.




Some of the more relevant posts from the previous blog have been re-published here. I am going to try to keep a tighter focus from now on