I wonder if fellow news readers can chime in as to what are the boundary conditions to let go and shut down a service. Some services have survived through waves

I wonder if fellow news readers can chime in as to what are the boundary conditions to let go and shut down a service. Some services have survived through waves of investment cycles - ie flickr etc, others vanished and are not remembered by but a few. There are couple of scenarios I am currently involved in. One is sort of bootstrapping traveler assistance service - this one is bootstrapped but with fairly large - 10 people editorial staff. The other is my small project that I am interested in using myself. The third one is a job style project that is suppose to break open a new market in the internet retail and has been running for already a month... Qualitatively there are different reason for closing a startup - boreddom, disinterest, lack of funds, the growth curve is too slow. From my understanding startup is not a company but a place to invest money in - that has a vertical market and near future potential for geometric function style growth and if it isn't happening startup isn't a startup anymore... but there are so many stories out there of people sticking it out for 2 years and others giving up in 3-6 months... It would be nice to have some sort of coherent picture when to give up and when to soldier on evolving product, adding users - trying new things... links to resources are welcome! TIA!