I think this picture clearly explains why making estimations is hard:
http://www.secretgeek.net/littlefishes_.asp
making estimations...
Moderator: General Moderators
I don't really see how an agile methodology would help here, since a methodology alone can't help here. Only as soon as a team commits to work (which they will deliver in the coming sprint if we talk about agile) they'll notice that some fishes aren't what they initially seemd to be.alvinphp wrote:that is pretty funny. proper estimation of a project is very important. this is why agile methodologies are gaining popularity. you only do a little bit at a time which lets you back out earlier if you need.
You might think that after a while you're going to make better estimates, but in my experience that's only true for similar fish, the real problem is that there is a lot of fish in the sea, most of which we know nothing or almost nothing about.
The other problem is that management will usually agree when they're asked to think about descoping features in place for other features but still wants *All* features to be delivered at the fixed deadline (Since they already signed a multibillion contract and just want you to make it work.. Usually they talk like: not enough time? get some extra developers then.. What they forget is that it takes some time to get an extra developer up to speed... An time is the one thing you don't have then..)