The article is essentially one big tautology: don't be distracted by distractions! Well, the problem is no one knows what's a distraction and what's the next Twitter. That's the skill that matters - not ignoring "distractions" altogether.
It's perhaps important to remember that Clay Christensen remarked that nearly all disruptive innovations look like toys at first.
Imagine the Odeo team split their efforts between Twitter and their first idea? I doubt Twitter would have became the site it is today if that happened.
No one could ever disagree with the assertion that you shouldn't devote time to sub-optimal outcomes - it's strictly irrational. "Distractions" arise not from founders making irrational decisions, they come from misjudgements about the expected future value of these "distractions." I'm not sure the solution is to ignore altogether things which have uncertain future value (which I believe is what you're proposing?) because our judgement about these issues is often so wrong.
"Hell yeah or no" and the like make for great rhetoric and catchy blog posts, but they assume our off-the-cuff intuition on such matters is correct. In my experience, it usually isn't.
How true is this! I'm on the side learning Django (because I want to use its admin feature to do a lot of CRUD) and on another side stint I'm dabbling in MongoDB, which also can come in very handy for my main projects.
