[Update: Before you read this wall 'o text, please consider the comment in the context of the update at the bottom.]
I disagree, YC is not an incubator. Well, to be fair, incubator is kind of a nebulous term.
I define incubator as: "entity that charges entrepreneurs for privileged access to ostensibly valuable [x y z]" or "entity that hires so called 'entrepreneurs in residence' to build companies under its roof in exchange for exorbitant equity."
YC is only superficially similar to whatever you would describe Harlem Biospace as.
The essential difference is in how the two types of entities derive value from their offering.
The distinction is that [YC, 500S, TechStars] invest in the upside of companies. Harlem [and its ilk] make money per unit of offering sold [units of monthly [x y z] access].
The reason "incubators" (the ones that charge entrepreneurs) are usually worse than the YC's of the world is because they adversely select for all the wrong things, as a side effect of their design.
Since an incubator ostensibly helps startups win big, why charge per unit [startup]? Why not invest in portfolio upside?
Per unit charges seem to be a good way of making money in the near term, at a constant rate (might not be sustainable if companies don't succeed in the long run.)
Upside investment seems like a good way to make money in the long run (assuming you are sufficiently able to (a) pick the right startups and/or (b) influence them to succeed.)
So investment seems to signal that you believe that you can do (a) and/or (b).
Sure, an investment portfolio may lose money on most of the companies. However, if it has a group of companies of sufficient expected success, one will likely reap massive returns.
If the value investors provide is worth enough to attract the right kinds of founders, the expected value of investing in entrepreneurs should exceed the expected value of charging entrepreneurs.
In the long run, charging per unit of offering, may signal that the long-term value of the seller's [x y z] offering is relatively low when compared to entities that are willing to invest in entrepreneurs.
Basically, if you value the unit cost higher than the investment return you are signaling that you (a) can't pick good startups and/or (b) you can't help a sufficient number succeed.
What kind of entrepreneur would want to go to that entity?
Since portfolio companies tend to return value later in their timeline, long-term value of offering is what matters. The YC's of the world are willing to eat the short-term costs and bet that they will influence some startups to return big.
The Harlem's of the world are making a constant per unit income, but are unwilling to bet on the long tail equity upside.
These incubators are not equivalent to [YC, 500S, TechStars], they are glorified co-working spaces.
[Update: If you've read this far, please see the discussion further on the thread. I unfairly characterized Harlem as an "incubator". Harlem Biospace certainly offers value. I do stand by my comment on "incubation" in the context of consumer internet companies. It's a different ball game for capital intensive industries. One aspect of Harlem is that it is affordable lab space that is group funded. Undeniably important.]