Tuesday, April 29, 2008

Simply Not Simple Enough

Ludovic, the man that originally pushed me to seriously pursue the brand comparison startup project, has been begging me for a real live website. He's sick of talking about the project, seeing people get excited, and then having nothing but a dummy site to show them.

He's right. We've been working on this for quite a while now. We should have something up and running that people can actually USE.

I think the big reason we don't is that I made a rookie mistake in the initial development process last fall and winter.

I didn't strip the idea down to its barest bones.

When I put together the first round of technical specifications, I didn't hold back nearly enough. There were pages I didn't draw and functionality I didn't describe, but I left levels of complexity in the requirements that simply weren't necessary to a concept proving prototype.

I described to the developers what was basically a fully functional site, and I assumed that, despite their other commitments, they could get the whole thing done in three months.

The couldn't, and I take the blame 100%. I distracted them with unnecessary details and caused them to neglect some aspects of the fundamental user experience.

I was overoptimistic. I was careless. I was unable to see my own lack of focus.

I see it now, however. It's frustratingly clear. We we have a prototype site that, while hugely useful for demonstration, interest piquing, and feedback gathering, is just simply not ready for users.

We're working on that, of course. We're considering the possibility of a rebuild using different technology, and we're considering the possibility of making renovations to the foundation we've already built and turning it into something usable.

It's a bummer to have to be operating this way. I'd much rather have a couple of hundred friends banging on an oversimplified site and complaining about what it can't do. But what's done is done. I made a mistake, and I have to live with the holdup I've created.

Hopefully I've learned something. Hopefully there's a moral to the story. Maybe this will only apply to my own personal weirdness, but here goes:

If you think you've simplified an idea, you probably haven't. Simplify it again, radically.

The fact that Wiley and I built a usable first version of twodaddy.com in 10 days tells me I'm making progress. Wiley will no doubt tell you I'm still a huge pain in the ass, a wealth of totally unnecessary ideas. I'd be surprised if he wouldn't concede, however, that I'm much more tolerable than I was six months ago.


wiley said...

You're right on track with this line of thought, and yes its true Jake, you really are way more able to compromise on the details than you ever were before.

My evolving observation is that developing a totally custom, basic data-driven website should not take more than a week (if you're using a framework like rails or django). In some of my projects (notably tinytrans, the "underground" project) two weeks of prototyping and showing the prototype to friends led me to discover that I really needed to totally re-think core concepts. Imagine if I had dotted every i and crossed every t, before showing it off or trying it out. I may not have been able to get that core feedback till I had spend months of hardcore development.

Unknown said...

Wiley, I knew I liked you for a reason :)

For the number of times I've been told (and told others) to make it much much simpler, you'd think it would be easier to do, but it never is...

Jake de Grazia said...

Wiley's pretty much my hero. Been some feint whispers about doing a rebuild in Python (as opposed to a .NET renovation or a PHP or Ruby rebuild), which means I might be able to throw a little more work his way.

Simplification is the worst. I try to do it, and I end up FURTHER complicating half the time.

I feel like I need an electric shock collar or something to keep myself under control.

Maybe I'll hire Wiley to administer the shocks.