'Hello World' for Startup Wannabes (like me)

I am considering a trial run. Specifically, I am thinking a long the lines of a one month cycle with two launches: one week on (design, develop, launch), one week off (project slack time if needed), repeat. What I propose is an exercise in self education and a test of commitment. I'm inspired by the examples like Dominiek and his ability to build a site in 24 hours (wigitize.com). I am also inspired by minimal solution sites like del.icio.us, grouphug, hacker news, and many others.

I am calling these: 'projects' or 'sites', as there is intention for them to become grandiose web properties. Rather the focus is on micro startup's (is that a meme yet?). I am thinking about (1) simple database base-backed sites, (2) using popular technologies, (3) with minimal cost. Basically the very definition of low barrier to entry on the web, other than static text. As such they are technology and process prototypes designed to mitigate perceived uncertainties. Surely such projects would be an exercise (assignment) in a class on web technology at any reasonable university, like the ML kids working on the Netfix Prize, or kids working on Facebook apps (both at Stanford incidentally).

At the risk of committing too early (or some smart ass hacker showing off: I believe in public by default and am betting on priority via obscurity), the following provides a preliminary list of 5 potential projects dredged from an old 'ideas' documents from more 18 months ago:

• Ask The Mob: A Q&A site where you post questions to 'the mob', and checkup on what the mob is saying. Inspired by the now defunct Google Answers (I loved it), and the zillions of community-centric Q&A sites (and lots of chatting with D).
  
• I Wish: A feed site, where you can post what you desire most, or expand your perspective by reading from the anonymous well of wishes. Inspired obviously by grouphug, a Melbourne guy already did it, although it is viable for any related form of the idea, e.g. I think, I confesses, etc.
  
• IMUR: (I Am. You Are) A community where you list your perceived virtues and vices, and your friends or others confirm or deny your claims, or make counter claims. The site promotes self assessment in the context of self opinion against the weighted opinion of others. Really high-level stuff like, "I think I'm smart, although my friends think I'm less so". An elaboration of HotOrNot and OkCupid Tests, and would make a good Facebook Application.
  
• Two Words: If you had just two words, how would you use them. Yet another take on the 'feed site', constrained to two words. Extensions are obvious, as are use of text visualisation techniques.
  
• Brain Storm: A community site that promotes the posting and collaboration of ideas. Provides a set of tools for extending and elaborating on other peoples ideas, perhaps with versioning, and or a thread structure. Cross-pollination of ideas, suggested reading, and recommendation systems are natural extensions. I would totally use this if it was focused on Lazyweb or LazyResearch. Inspired by google docs, google notebook, and Wiki's (and lots of chatting with Matt, hat tip).
  

They suck, but whatever. I suspect you could build some of them in ning with little or no effort, but that is beside the point. They are training exercises, 'hello world' for the startup wannabes (like me). An alternative strategy I have considered is to go the Facebook route. It is appealing if for no other reason than the present size of the active user base. I need to think about it some more, superficially such an approach seems boring.

An observation I have made both with my brainstorming (whether it is for 'startups' or for research), is that I am technology focused. It's natural because that is what interests me, whether it is a presently popular persistence framework, or the latest stochastic optimisation algorithm. This has the effect of designing systems without explicit purpose (not addressing a specific problem) which feels like a bad thing, especially for the web (unless I want to be a technology provider of course). I've been considering an approach to force a problem focus back into the mix. For example, take a standard or even obscure (although specific) problem, and come up with a new approach to addressing it every day for about week or so, then enumerate popular problems. Other than having the potential for being really boring, it may promote a real-world problem solving perspective rather than a technologists contrived problem solving approach.