Interesting... I lean more to the latter, I am usually happy to seed, as you put it, other startups I find useful, but not necessarily meeting my needs.
I think posting your startup on news.yc allows it to be given a closer look since the community here is much tighter than most sites. I'm much more willing to graze past the usual .5 second judgment of any site posted here.
In terms of seeding for my site, if let's say 100 people added just one resource and just one sticky for that resource, then I think the site would look much better.
This is something that can be done in less than 15 minutes.
There are many services for this. Forums such as digital point or software auction sites like rentacoder work well. Even paying outright for your users is very common. Numerous large social sites with money have used affiliate systems such as commission junction, usually offering from .75$ to 2$ per user. Finally, youtube, myspace, every big social site has done contests that award people who refer the most new users.
I find the ways of getting users without paying for them directly to be more interesting though.
It is so difficult to get the right users to pay for your services when the most people using the Internet believe in getting things for free.
The problems of most startups (and VCs who back them) is their lack of discipline to map out a tangible goal and direction.
How can people develop a product when they do not know the market?
Developing a product that is looking for a marketplace is a challenging way to generate revenue.
Here's something of that nature I came across some time ago: http://forumshock.com. No clue if something like that is even effective.
Many times, it's hard to show the value of your site without a populous user base. It'd be wise to look for a very specific, good use case that will help spread your service -- or you might get lucky with something like YouTube's "One Lazy Sunday" of your own.
The best is if you take your seed money from a guy whose rich enough to fund you, but poor enough that he spends all day submitting stories from fake accounts to make sure he doesn't lose his investment. :-)
So reddit faked their early users... even though they had pg's readers. A new social site must get really creative if they consider faking users not ethical (as I do).
I can't help being curious; aren't there algorithms that itentify the likelihood of a sample of text being from a given writer based differences in writing style. If so we have plenty of writing guaranteed to be from Alexis and maybe we could identify which accounts were his.
We didn't have commenting back then, so these were all link submissions. We basically stopped using these accounts after that first month or so -- once we started getting traffic.
For us, there didn't appear to be a major ethical dilemma. We used these accounts simply so that new users would see a front page of links instead of a blank website (or one with only submissions from 'kn0thing' and 'spez'). For a site like reddit, it seemed to be a key part of showing new users how the site worked. Once we got traffic (from places like YC and PG) we stopped needing these accounts and thus stopped using them.
Without writing comments it is probably Ok for me as well. Voting is still somewhat about faking other people's opinions, but there is obviously a large grey area here.
This could be done by paying people or perhaps just agreeing to seed each other's startups.