Successful sharing sites are highly democratized, which means in some sense catering to the lowest common denominator.
Put another way, you need volume to gain any traction.
Reddit has a lot of crap because it has a lot of volume: but as a sharing site it has mechanisms to let people en masse filter the crap for the gems. Such as they are, in a place like /r/adviceanimals.
HN has a fair amount of quality because it caters to a very specific crowd and it self-moderates much more mercilessly than reddit; but it still has to deal with the problem of crap.
Upvotes are highly visible. Stories fail. Anyone can say anything. The crap and the site's system for reacting to it makes the process of content gaining traction _demonstrably valid_ to a visitor to the site.
If none of this is immediately obvious in your sharing site, I don't expect it to gain the critical mass it needs for success.
(You might think TED is a counter-example in the Wonderful Videos domain, but the obvious difference is that TED is a content provider, not a sharing site.)
Edit: And you've got the social bubble virus, in that your 'social network' requires people to once again set a glorified address book in order to really leverage your resources. You're too specific for people to want to spend a lot of time cultivating!
Was it really that long ago that "Build a social network for {{x}}" was regarded as a bad idea?
I have this vague memory that investors or startup advisors used to laugh at any startup idea along the lines of "We want to build a social network, but for florists!"--and that seems to have receded in the last year.
The difference is that the new niche social networks are centered around selling stuff rather than making money from ads. Niche networks can't get big enough to make money from ads. They can get plenty big enough to sell stuff though.
Put another way, you need volume to gain any traction.
Reddit has a lot of crap because it has a lot of volume: but as a sharing site it has mechanisms to let people en masse filter the crap for the gems. Such as they are, in a place like /r/adviceanimals.
HN has a fair amount of quality because it caters to a very specific crowd and it self-moderates much more mercilessly than reddit; but it still has to deal with the problem of crap.
Upvotes are highly visible. Stories fail. Anyone can say anything. The crap and the site's system for reacting to it makes the process of content gaining traction _demonstrably valid_ to a visitor to the site.
If none of this is immediately obvious in your sharing site, I don't expect it to gain the critical mass it needs for success.
(You might think TED is a counter-example in the Wonderful Videos domain, but the obvious difference is that TED is a content provider, not a sharing site.)
Edit: And you've got the social bubble virus, in that your 'social network' requires people to once again set a glorified address book in order to really leverage your resources. You're too specific for people to want to spend a lot of time cultivating!
Was it really that long ago that "Build a social network for {{x}}" was regarded as a bad idea?