I understand that at their current level of maturity the value of EBay and/or Craiglist is partially tied to the size of the marketplace and that focused-and-better may be the best or even only way to compete with them but it seems like on that particular point what works for you, the small emerging marketplace could work just as well for them, they just have a higher overhead to meet. It's not enough just to be more focused, you need to be better at your niche.
You're asking why it's a negative for eBay and Craigslist that they aren't small (i.e., that they have a lot of listings), but I think the quote you took from the article is saying the inverse - that it's not necessarily a negative for Etsy and Threadless that they are small. In fact, I think Jason does recognize that being bigger is an advantage:
>> "Clearly the value of a marketplace increases as it grows — both as a business and to the buyers and sellers."
As to your second paragraph, I agree it comes down to you meeting the needs of your niche in some way that is better than the way the big boys are doing it (that is, of course, if you are competing directly with them). There is nothing, in principle, stopping them from replicating whatever you do to create an advantage. And you certainly won't win just because you have a smaller market. But there are a lot of products (like the t-shirts on Threadless) where relative obscurity increases the value of the product. There will always be consumers who shy away from the mainstream offerings, in some cases because the offerings are mainstream.
a) Building two businesses is really scary.
b) Focus on sellers first is definitely right. You'll never have any buyers if you don't have any sellers. The worst case scenario for a seller if there are no buyers is an increase of zero / a loss of nothing. Things can only go up.
c) Pay per lead works great here. The listing fee up front is hard to do and usually penny wise + pound foolish. We have close to 500 apps. I bet if we tried to charge a small 50 dollar fee to be on cloudomatic we might have gotten it from some devs, but we'd have a lot less apps (maybe 25 total?).
d) Content is really important as a way to gain traffic on the buyer side. It worked well for Mint. I consider Mint to be a marketplace at its core (connect people (buyers) with better finance deals (sellers)).
