[nostrademons]

Another thing to consider:

We've been about 70% complete for version 1 roughly half a dozen times now. We were 70% complete in April of last year, which we actually did release (it's getting users now, but not many). We were 70% complete in October, when we sent along a demo with our YC app. But we realized that it wouldn't work for the general case, so we went back to the drawing board. We were 70% complete in November, when I realized we were rewriting things too much with each additional feature and stepped back for a week to actually design the thing. We were 70% (well, more like 50% by my reckoning) complete in December, when we switched web frameworks. We're about 70% complete now, though if this is anything like past 70%s, it's probably more like 40%. Now I don't trust any estimate that's not backed up by Trac tickets.

Always remember the 90-90 rule: "The first 90% of the work takes 90% of the time. The remaining 10% takes the other 90%."

Many startups have to go through multiple rewrites before they have something useful - Reddit's on their 3rd (though they launched with their first), I heard Xobni had to redo their first try to be more ambitious, YouTube completely revamped the site to make it stickier, my last day job went through about 4 rewrites before we launched the product (which is the 7th product it's launched...none of them really got product-market fit), the job before that changed directions 4 times in the year I was there and completely rewrote things about a half-dozen times, never actually releasing anything.

I'd look at the stuff you do now as exploratory programming; your job is to find out more about the area, so that it eventually gets to the point where it "clicks" and you can just crank out code that works. You're still very early in the product process; a cofounder could add a lot of value down the road.

[lanceusa]

Points well taken...and as I see you are talking from experience...exactly what I was looking for to put my mind at ease. We are planning to meet in a few weeks...if all goes well I'll have a new member on the team that seems as enthusiastic as I am about the project.

I like what you said exploratory programming because with a new cofounder I can sit down and say, "look this is what I've found will work, this needs help, this needs someone to just start outright, and do you have any ideas how the hell to make this work." Thanks for replying...I'm glad you did.