Hacker News Reader

Comment by boomshucka | Hacker News Reader

I'm pretty sure he was being sarcastic... It is pretty absurd to think you are going to get your head around the requirements for a project and come up with a good estimate in 10 minutes. It suggests naivety or dishonesty. You choose.

View on HN

Half sarcastic, maybe. But, genuinely curious about how they are managing the risk around scoping/estimation. Perhaps using overly conservative estimates or else assuming that on average their estimates will be close. Or maybe those Google/MIT engineers are magical.

I'm pretty sure they're planning to throw VC money at the problem until it goes away ("it" being either the problem or the money).

I don't think estimating will be that much of a problem if the right formula is used. Is this straightforward and the team a has done stuff like this before (most cases)... apply the component formula and add the markup. Is it hairy with lots of unknowns? 1.2 million dollars.

No disrespect meant, but no one with any real software development experience would suggest that a quick formula can be used to estimate the cost/time to develop a given project. Very experienced people spend lots of time up front trying to estimate cost/time and are usually not very close.

Agreed. Re-read my comment?

"Is this straightforward" can't be figured in 10 minutes for the vast majority of projects. And if you are outsourcing to devs, "has the team done stuff like this" is going to take some time too. 10 minutes simply isn't enough time. Would you have the same thought about 1 minute? 5 minutes? At what point do you say "no, that isn't enough time?". I'll take a few hours as enough time for some reasonably small but not trivial number of projects as a %.

I've built or managed the teams that built over 50 software projects, ranging from small up to about half a million dollars.

With all that experience, my estimate process is now extremely honed. It takes me, quite consistently, 4 to 8 hours for a typical web startup MVP. And that's assuming I can build the product completely with technologies my team has used before.

Reply on news.ycombinator.com