None of you saw my site when it launched, because you were not among the 70 people who were subjected to that horror. (Those brave souls gave me many insights on how to make it better.) For the 60% of 5,000 people who visited yesterday who had never been to the site before, the past 3.5 years essentially didn't happen. To them, my site launched on Valentine's Day.
I also think people vastly, vastly overstate the degree to which you will develop a "reputation" or "brand" as being (whatever you are currently). Large companies pay Madison Avenue staggering sums of money to get people to remember them. They do it because it is inherently hard work. 99.99% of the time, a user having a negative experience with your site ends with the back button. 99.99% of the remainder, you get an email with, ahem, variable levels of politeness. Respond to it with grace and humility, and you'll avoid almost all long-term issues.
Another example, Age of Conan, great game idea, early execution was off and it set them on a path where they were never really going to get back to initial subscription numbers no matter how much effort they put into the game.
EX: I really want an improved version of PowerPoint that let's me embed editable graphs without crashing and it would be nice if the interface was cleaner. Now if I had a lot of money I might try and build a competing version of PowerPoint but what I really want is a feature, now what's the minimum product? How about a PowerPoint add on that cleans up whatever the bug(s) that causes the problem in the first place? Or a website that fixes broken PowerPoint files? Ok what's the minimum amount of work to to start to get money from this idea?
Also if you have money tucked away to survive in some cases it would be better to sit back, re factor, add/ improve features until you have something your really confident about spreading it around widely. There is a lot of talk about release early, but for smaller unknowns you may only get one chance to make a good initial impression.