There are certainly startups hoping to become redwoods, but there seem to be many with goals or target markets that would likely limit them to shrubs at best. Of course, there is also a large class of "startups" that are a lot more like Christmas tree farm seedlings.

Not quite: startup = cheatgrass: see http://www.ext.colostate.edu/pubs/natres/06310.html

Potentially, yes, but I do NOT believe this definition could be applied universally.

The author's point is that both IBM and the local bakery are built to last, patiently building value over time.

Startups are too often get-rich-quick schemes whose founders want to work hard for 3 years, then cash out and spend the rest of their time on 'speeches and beaches'.

IBM wasn't "built to last." It was a rollup of a bunch of cash register companies, handed to Watson to run by the financier who bought them.

The truth is that in reality there's usually less planning than there later seems to have been. Things that turn out big were often small and tentative initially.

I am not sure how accurate your history of IBM is since I myself am not an expert.

I agree with your last 2 statements completely though.

This was exactly the point I was trying to convey :)

"Speeches and beaches" -- my man, such a desire is a wholly lazy and narcissistic dream.

In an ideal world, this is true.

However, pay attention to what the ingredients of that seed are.

That is why there are "startups" (the type I described in the article) and then there are Startups.