I think this needs some context. I felt much the same way when I decided on a college, and made my decision based on that. I won't exactly say I was wrong, but 7 years later I've got a better perspective on the tradeoffs involved.
Understand that your raw intellectual prowess is a depreciating asset. I'd guess that I peaked in programming ability at about 19, and 18-20 seems a pretty common age range based on famous programmers. I'm certainly not as productive a programmer as I was at my first professional job at 19, and my 35 year old coworkers have said that there's no way they can keep up with what I do now at 26.
You continue gaining experience for a while after that, so it's not like you're over the hill at 20. The mid/late 20s seems to be the "sweet spot" where you still have decent technical prowess and yet enough experience to avoid blind alleys. But if you fritter away your early 20s, you lose the benefits of high ability and don't have the experience to show for it. You're starting behind and have to play catch-up to more youthful and more energetic competitors, which is possible but can be kinda discouraging.
OTOH, trying out a variety of things gives you a lot of perspective on what you do or don't want to pursue. I was certain I wanted to be a physics major and then a theoretical physicist when I entered college. I flirted briefly with majoring in philosophy or sociology when it turned out physics was much harder than I bargained for. I tried out fiction writing and found I wasn't any good at it. I made all-state violin in high school, but again realized that I didn't have the passion for it needed to be a professional.
The one downside of having all those options open is that I never finished anything. There was always some other avenue that looked more attractive. It was only in my senior year, when I was practically flunking out of physics and had a bunch of hobbies that weren't all that useful, that I decided I was gonna close off all the other avenues and focus on computers. And it was only then that I started actually producing stuff that was useful.
So, I'll agree with the parent post that it's important to grow creatively, socially, and intellectually in college. But understand that there are doors that going to Stanford/Harvard/MIT opens, and those doors tend to slam shut afterwards. It's possible to prop them back open, but it takes a lot more effort, because you're working against people's expectations. Similarly, there're doors that open from focusing deeply on one field, and it's much harder to achieve mastery in a subject than it seems in high school.
I remember thinking in high school "I've got 70 years left in my life, that's plenty of time to accomplish everything." But you don't - you've got maybe 10-15 years (between the ages of 20 and 35) to really make a mark on the world, and then the rest is for raising kids, writing books, passing on your wisdom, and generally enjoying the fruits of those 10-15 years.
(I do reserve the right to change my mind in 10 years and say "Pish-posh. Life's not over at 35 - keep an open mind and you can keep doing cool stuff until the day you die." But this is how it looks from the mid-20s.)
Sadly, the attitude evident in the last two paragraphs of your comment runs deep in young people. Zuckerberg at YC Startup last year, age discrimination suit at Google, etc. Logan's Run in Silicon Valley.
It's also upside down compared to other fields like medicine and writing. In medicine, you make your "mark on the world" after you go through significant training, which ends around 30 or later.
As far as making it entrepreneurially:
1) If it depends purely on intelligence or hacking ability, PlentyOfFish.com and other success stories wouldn't have made it. Even if it's somewhat true, consider that the decrease of your ability with age should be less than the interpersonal variability of ability. (50 yo Einstein > 20 yo Homer Simpson)
2) If it depends on luck, don't worry about age.
3) If it depends on connections, you get more with age.
4) If it depends on others' perception of your youth.... you may be in trouble.
Markus Frind is quite a bit more intelligent than many people give him credit for, eg. he discovered the then-longest arithmetic progression of prime numbers in 2004. Same goes for many other entrepreneurs that people assume were just lucky - James Hong, Marc Andreesen, Brad Fitzpatrick, etc. And the variation due to age may be much less than the variation due to person, but remember that there are many more smart people than there are millionaires. Empirically, it seems like you need to be smart, young, and lucky.
Also - I doubt few people would dispute that succeeding with a tech business requires the ability to concentrate intensely for long periods of time. (Something I'm obviously failing at, having been back to news.YC 3 times so far today.) Certainly many young people lack that, but for a given person, it seems likely that they're better able to do that at 20 or 25 before things like family, business connections, or reputation intercedes.
It definitely is upside down compared to medicine or writing, but that's because computer programming is nearly in unique among human fields in being subject to lots of chaotic effects. Almost every other body of knowledge is additive, but in software, small changes to requirements at the base layer can have ripple effects that invalid your whole system's design. That's why the software industry seems to undergo major tectonic shifts every 10 years or so, each of which creates lots of opportunities for young entrepreneurs who haven't learned the now-obsolete ways of "how things are done".
When I was a kid, the hot new field was microcomputers, the programming language was assembly (no wait...Pascal. no wait...C), and the largest platform was Windows. When I was a teenager, the hot new field was the web, the hot programming languages were Perl and PHP and Java, and the largest platform was Netscape. Now that I'm in my twenties, the hot new field is rich Internet applications, the hot programming languages are Flash, JavaScript, Python, and Ruby, and the largest market is Internet Explorer. I can almost guarantee that the hot field, language, and platform will be different in 10 years, and all the technical skills I've learned so far will be obsolete.
Edit: On reflection, it's not really software as a whole that's chaotic so much as the leading edge of software, i.e. the edge where you can still have a viable startup. If you're making a desktop app, the tools and skillsets are only slightly different from making a desktop app in 1995. If you're making a webapp that's just a sequence of forms, the tools from 1998 will work fine. But those are mostly solved problems: either there's an off-the-shelf solution that's already been built, or you can hire one of an army of consultants that knows how to build one. The leading edge, by definition, involves tackling problems that haven't been solved before, and so there's no advantage to experience in solving them.
I didn't know about Markus Frind's intelligence (interesting info), but in PlentyOfFish's case (and successes like MySpace), it's no masterpiece of hacking.
Writing prose can be similar in intensity to writing code. John Scalzi, a recent sci-fi award winner, wrote a book for writers called "You're Not Fooling Anyone When You Take Your Laptop to a Coffeeshop." Many writers block out pure concentration time. Nonetheless, I'll concede that long stretches of time are more available to the young, unless they are out partying :)
I wonder how many big successes depend on bleeding edge software? For every Google, there are plenty more like MySpace, Ebay, Amazon, and Facebook.
I usually live off specialized applications, where you build software that focuses on a particular need for a niche population. Knowing the domain (say radiotherapy or creative writing) is really useful because of insights into how you can fashion new tools using advancing technology (e.g., faster 3D graphics or a better web framework). So while I agree that creating totally innovative approaches might be easier for the young, I think experience helps in creating domain-specific software.
My main point, though, is even if you believe you have a limited shelf-life, you might consider it's your shelf-life in only one of many possible avenues to make your mark.
Understand that your raw intellectual prowess is a depreciating asset. I'd guess that I peaked in programming ability at about 19, and 18-20 seems a pretty common age range based on famous programmers. I'm certainly not as productive a programmer as I was at my first professional job at 19, and my 35 year old coworkers have said that there's no way they can keep up with what I do now at 26.
You continue gaining experience for a while after that, so it's not like you're over the hill at 20. The mid/late 20s seems to be the "sweet spot" where you still have decent technical prowess and yet enough experience to avoid blind alleys. But if you fritter away your early 20s, you lose the benefits of high ability and don't have the experience to show for it. You're starting behind and have to play catch-up to more youthful and more energetic competitors, which is possible but can be kinda discouraging.
OTOH, trying out a variety of things gives you a lot of perspective on what you do or don't want to pursue. I was certain I wanted to be a physics major and then a theoretical physicist when I entered college. I flirted briefly with majoring in philosophy or sociology when it turned out physics was much harder than I bargained for. I tried out fiction writing and found I wasn't any good at it. I made all-state violin in high school, but again realized that I didn't have the passion for it needed to be a professional.
The one downside of having all those options open is that I never finished anything. There was always some other avenue that looked more attractive. It was only in my senior year, when I was practically flunking out of physics and had a bunch of hobbies that weren't all that useful, that I decided I was gonna close off all the other avenues and focus on computers. And it was only then that I started actually producing stuff that was useful.
So, I'll agree with the parent post that it's important to grow creatively, socially, and intellectually in college. But understand that there are doors that going to Stanford/Harvard/MIT opens, and those doors tend to slam shut afterwards. It's possible to prop them back open, but it takes a lot more effort, because you're working against people's expectations. Similarly, there're doors that open from focusing deeply on one field, and it's much harder to achieve mastery in a subject than it seems in high school.
I remember thinking in high school "I've got 70 years left in my life, that's plenty of time to accomplish everything." But you don't - you've got maybe 10-15 years (between the ages of 20 and 35) to really make a mark on the world, and then the rest is for raising kids, writing books, passing on your wisdom, and generally enjoying the fruits of those 10-15 years.
(I do reserve the right to change my mind in 10 years and say "Pish-posh. Life's not over at 35 - keep an open mind and you can keep doing cool stuff until the day you die." But this is how it looks from the mid-20s.)