Image if you will a world where computers run on cycle petro. A fill of cycle petro would get you so many clock cycles. It's super cheap though so programmers don't really need to worry about the cost as they make programs. They can focus squarely on what people want in their software. They are free to write programs like Excel that recalculates with every change and use lots of clock cycles. They can include cycle heavy features whether the user needs them or not. They want them. Features like this, while wasteful of cycle petro, are what people want and software companies sell lots of software by making sure their programs are plum full of cycle intensive features that people want.
Then as time progresses, people realize that if we're not careful in our use of cycle petro there might be an impact in our world at some point in the future. It's not necessarily clear that this is the case but we are told that we should be looking at using our cycle intensive programs sparingly for the sake of the planet. Cycle petro is still cheap though so it's our conscience that needs to direct our actions. Perhaps cut back on playing games (because you know you still need Excel). This news doesn't seem to impact software sales much though so programmers keep pumping out cycle intensive programs that users love (and buy).
Then, for reasons that are not apparently clear, the cost of cycle petro begins to go up. And up. And up some more. Cost begins to actually make an impact on the users who use programs like Excel. Software companies don't sell as much software, while cycle petro companies make record profits. Opinions start flying. Something is wrong. Obviously people are not buying software because these companies are no longer making programs they want. Obviously the people's concerns over the planet are the real reason these companies are not selling their software anymore. A conscience awakening has occurred at precisely the same time as the rising cycle petro costs. Software companies are told by very influential people that folks want Calc.exe and not Excel. Programs that use very little cycle petro and have a small impact on the planet. They are told that they have been wrong all along on what people want in a program. They are told to re-think the programs they offer and make more programs like Calc.exe That's what people want and what people will buy.
How does this story end? Don't know yet.
Software companies listen to the very influential voices and begin to re-think the programs they make. Calc.exe gets a face lift. They come out with Calc 2 that allows users to do more, but still remains true to using the fewest cycles possible. Resources previously used for Excel development (the product that they scrapped because nobody is buying anymore) is now diverted to specialized low clock cycle applications.
Software companies rebuff the influential opinions on what people want. They remain confident in their own research that people want Excel. They conclude that what people really want is Excel to use fewer clock cycles. They put all their resources on figuring out how to get Excel to work while consuming the same number clock cycles calc.exe currently does.
The issues of the current American automobile industry doesn't really map to software so well. I thought it would be fun to voice my own opinion on the topic with a metaphorical story. People don't want to drive matchboxes. They don't want to crap up the planet either but that's not the motivation driving current auto sales. We do have a conscience but we also have kids, dogs and gear and we live in areas with plenty of space between points of daily interest. The good folks in Detroit are not idiots and they have not ignored the consumer. We are not buying their cars because wealth is currently evaporating before our very eyes and they have not yet made a full size car, truck or van that gets 50 mpg. No one else has either. A fact their critics choose to ignore or haven't yet figured out.
There is a true parallel to software in all of this. Paying attention to the wants and needs of the folks who actually use your software will result in more successful projects.
In every software project there are people with more influence than others. Their opinions carry more weight and not always because they actually know about the application domain (kinda like when congressmen and senators speak from on high and tell car company CEOs the error of their ways). And as with the automobile industry, sometimes the influential voice doesn't always align with what the users are saying. Many factors determine the success or failure of a software application. In software there is one subtle but unswerving truth. If the people actually using your software application don't want to use it, success will be hard to find or very short lived if found.
When you find the voice of the influential clashing hard with that of the user, don't ignore the user. Ask him more questions.