4 ms·

Joshu - Thanks for the comment. Curious as to what exactly you mean by "cut the idea to the very minimal core to see how small you can make it, and see if you c

Joshu - Thanks for the comment. Curious as to what exactly you mean by "cut the idea to the very minimal core to see how small you can make it, and see if you can validate that minimal concept." Are we talking about hashing out some specific feature and then seeing if that, on its own, will be successful? I guess, if you could explain a little about what you mean by "validating a concept," it would make things more clear to me. Thanks

joshu 17y ago
People have a tendency to overcomplicate things. You are looking to build a system that is a) as small and as simple as possible that yet b) still retains conceptual integrity. You want the first because it's easier to build and get to market, and the second because it's not done till all the necessary parts are in. Remember that the vast majority of your understanding of the problem space is not accurate, and will remain so until you encounter customers and users. So do as little work as possible; just enough to get you to those customers and users. Then you listen to them to find out the next parts to build. (The ideal case here is that you are also customer #1, incidentally.) So no, not just some specific feature; you need the right coverage. But make sure you don't go overboard anywhere in particular, especially if it might not be needed.
- ab_initio 17y ago
  Awesome. Thanks for the clarification. I wonder if you (or anyone) think it would be more fruitful to take a holistic approach i.e. factor in ebay, twitter, HN and create an overall score, or stick with a simpler, more streamlined karma counter type thing that would be "widgetable". Or does the latter necessarily lead to the former?
  - joshu 17y ago
    You're still dealing at levels of abstraction too far away from an actual product. At a distance, all these things are similar, but up close they are quite different. How would you "factor in" ebay and twitter into one number that made any sense at all?

joshu 17y ago

People have a tendency to overcomplicate things. You are looking to build a system that is a) as small and as simple as possible that yet b) still retains conceptual integrity. You want the first because it's easier to build and get to market, and the second because it's not done till all the necessary parts are in. Remember that the vast majority of your understanding of the problem space is not accurate, and will remain so until you encounter customers and users. So do as little work as possible; just enough to get you to those customers and users. Then you listen to them to find out the next parts to build. (The ideal case here is that you are also customer #1, incidentally.) So no, not just some specific feature; you need the right coverage. But make sure you don't go overboard anywhere in particular, especially if it might not be needed.

ab_initio 17y ago
Awesome. Thanks for the clarification. I wonder if you (or anyone) think it would be more fruitful to take a holistic approach i.e. factor in ebay, twitter, HN and create an overall score, or stick with a simpler, more streamlined karma counter type thing that would be "widgetable". Or does the latter necessarily lead to the former?
- joshu 17y ago
  You're still dealing at levels of abstraction too far away from an actual product. At a distance, all these things are similar, but up close they are quite different. How would you "factor in" ebay and twitter into one number that made any sense at all?