I got good advice from Silicon Valley investors about Project Melville, my stealth-mode Web business, but no money. Everyone agreed my best bet is to bootstrap it. In other words, don't quit my day job, eat brown bag lunches, and pull in a few friends as roll-up-your-shirtsleeves partners.

As noted last week, I have an idea for starting a new business (see "The Kernel Of An Idea"), and I decided to run it past a few experienced investors when I was in the Bay Area. My first meeting was with Doug Renert of Tandem Entrepreneurs, an investment firm started last year that favors investments of under $1 million. What Tandem lacks in size of investment, it makes up for in involvement with its companies, helping them with market strategy and finding the right technical and management talent, for example. Tandem has invested in about a half dozen companies. They include Attassa, which is developing software to manage e-mail overload, and Zecter, focused on online document sharing.

Doug's first question was, How are you going to build up a user base quickly? Building a Web audience organically can take years. The obvious answer is to seek partnerships with established players, and while I know who they are, I'm not far enough along to approach them yet.

In fact, it's becoming clear that I need to develop something approximating a business plan. It's possible to get started without having everything figured out. See Marc Andreessen's "Why A Startup's Initial Business Plan Doesn't Matter That Much." At a minimum, though, I need to get some ideas and numbers down in writing to frame this out.

Bootstrapping might work since it's not that hard to get a Web site up and running these days at minimal cost. Doug Renert pointed me to Rentacoder.com and Elancer.com as two places to find affordable programming and design help. Tandem Entrepreneurs got a bunch of custom illustrations done for its site for only $100.

If there was any question that bootstrapping was the way to go, it was put to rest in my meeting with Bruce Fram and MR Rangaswami, two veteran Silicon Valley advisers and investors. Bruce suggested that $10,000 to $25,000 might be enough to get me started. MR challenged me to think lower -- to try to get my site started for nothing, or next to it.

With URLs selling for $10, hosting services starting at $10 a month, and freelance programmers paid by the hour, it's possible to launch a Web site for a few hundred bucks. I envision a site that melds blogging, video, photos, social networking, user-generated content, and more, which may raise the cost of entry. I've got more homework to do, but the plan for now is to build a prototype on a shoestring budget, then go from there.