I agree that a lot of it depends upon personal preference and style. I also agree that having a file browser wouldn't hurt anyone's enjoyment of their Kobo. I just diverge from your opinion because the "just one more feature" philosophy doesn't work very well.

I once read a story about Microsoft Office. Now a lot of people loved Microsoft Word 5.1 for the Mac. It is usually cited as having the features that people need, no more and not much less. If you asked people how to improve Word 5.1 for the Mac, they would usually come back with one feature request. The thing to understand is that the feature would differ from user to user, but it would not amount to a large change to Word. Yet put altogether, it would amount to a huge change. It is something that rang true to me because I could see myself in that position: there is only one feature that I wanted to see added to Word 5.1 to perfect it. (Note: I'd probably think differently today, but it was true at the time the article was written.) Now you take those individual feature requests and create a new product. The product that you end up with ends up being disliked because it is unwieldily. You ended up with Word 6.0.

When I see these requests of Kobo, I see much the same thing. Even if you ignore the bug fixes, which would improve the product without affecting how people use it, you still end up with an assortment of feature requests that have the potential to make Kobo readers much more complex to use. To make it more concrete: a full screen reading mode sounds great, a file browser sound wonderful, the ability to specify formatting is empowering, and so forth. Individually, these may make a nicer device. Altogether, they make a contraption that is so complex to use that customers would rather not use it.

So I'm not saying that I disagree with you on the minutia. I do disagree with you on the consequences. In the end, Kobo should do their market research and justify each feature based upon their customer's needs as a whole rather than fulfilling individual requests that are honest but would produce an unmarketable product.