I personally use folders to organize all of the assets related to my projects - things like plans, strategies, installers, code, procedures, & information resources. It makes sense to me.
Research is a part of most of my projects. So for those I need a project to be able to reference web resources somehow. For me it's very convenient to be able to collect and organize web links while I am researching, by simply dragging the URL right into my project collection. With this work flow I am able to organize the resource right then and there, rather than waiting till some later time to re-address it. It becomes more complicated. And at that later time if I have to navigate somebody else's idea of what works organizationally and convert it to how I do things, adds more complication. And its easy to forget things. Occasionally while researching for one project I'll stumble on some web resource that helpful to another project. So by being able to drag/drop URLs to the two project folders right then is huge. Again it removes the later complication of re-visiting and re-referencing resources and trying to remember which projects need it. To me at least, flexibility and practicality are paramount because they help me be effective. It's my firm opinion that functional flexibility is one of the criteria of great application design.
In my opinion the argument of considering or supporting features based on the majority or based on some perception of the majority has pitfalls. To say that here doesn't seem reasonable - first of all there's no formal mechanism for voting on this blog...
But I think a lot of what you are saying about majority-driven design is also an oversimplification - What the majority wants can easily be distilled into that fickle herd mentality that lives for cultural fads. The "herd" doesn't necessarily know what it needs or wants; but it's got members with itching senses that are waiting to be fed by the new and the novel... often it's the blind leading the blind. Or it's a crafty few
manipulating majority trends. A lynch mob is majority-driven decision making at its worst.
But if we add time and sensibility to the equation, the notion of majority-influenced functionality can become more compelling. Then it's about value, purpose and analysis. Over time we can analyze and learn from mistakes, we can see what persists in majority culture and why. The differences between fad and value, fashion and practicality, whim and purpose become clear. We begin to see the value where it sits. Society has a way of adopting what has worked and continues to work. Those things tend to float to the top and become more permanent fixtures or standards within culture, to the extent that they become the norm, or defacto standards. But the majority seldom drives creation of great functionality; rather, it reacts to it and adopts it to some level. I'm sure that the creators of Opera get most of this...
Over the years this URL issue has been debated over and over in Opera circles. The functionality has emerged and disappeared a few times within various releases. Yet over the same period of time the functionality has arguably become a defacto standard among the major browsers. Whether it has or not, many or most browsers do offer the functionality. So then the question becomes not so much "why should Opera do it?"; but rather, "why isn't Opera doing it?".