News for hackers. Stuff that matters.
Re:Ask HN: What do you do w/ a big idea, when you are workin... by jacquesm on Saturday January 23, 2010 @04:04PM (#1072411)
Do both.

Most of the time you can spare an hour or two every day to work on your 'big' idea. Spend it on research, try to find out if it has already been done and if someone else is executing right now.

99% of the time your 'oh shit, this could be huge!" of today will turn to, "oh right, that's why they failed" or "oh, great I'll never match that" or "hm, this is a lot harder than I thought", or maybe even "wow, that was really stupid".

If after three months of trying to shoot holes in it it still looks like a winner then it's time to start worrying about priorities.

Re:Ask HN: What do you do w/ a big idea, when you are workin... by andrewljohnson on Saturday January 23, 2010 @04:59PM (#1072474)
I don't really find this to be effective advice, for everyone at least. I could believe that some people can have multiple major projects going on at once, but I think for many, trying to split attention between projects will hurt both. That's been the case for me at least.

I'm not talking about some consulting gig or school paper or something. These things you can juggle. But if you are pouring your heart into a start-up, that leaves little time to build a side project.

I agree with the sentiment that one should think through good ideas and tinker with them though, particularly before changing course. I have what I think is a great business idea most weeks, and they do tend to already be done, harder than I think to build, or not as valuable as I had originally thought, ginned up on coffee some excited morning or beer some fun night.

Re:Ask HN: What do you do w/ a big idea, when you are workin... by jacquesm on Saturday January 23, 2010 @05:24PM (#1072521)
The 'do both' is not meant to last forever, just long enough to be able to decide whether your new idea really has as many 'legs' as you think it does or whether it is just folly.

You have to come to some kind of closure with stuff like this otherwise it will nag you forever.

So, after a couple of weeks/months of trying to shoot holes in it there is more than a small chance that the answer is 'it sucks'. If not then you have to decide which one of your 'children' you want to keep.

At that junction you have several options:

  - sell your current project and use the proceeds to 
    fund the next 

  - spin off the other one to a bunch of people that 
    you trust

  - kill it anyway, because your 'old' project has in 
    the meantime gotten significantly more traction.
You basically buy yourself time and and gain more knowledge to make a much more informed decision.
Re:Ask HN: What do you do w/ a big idea, when you are workin... by jayliew on Sunday January 24, 2010 @02:32PM (#1073670)
Yup, history shows this works: Twitter was Odeo's side project, Blogger was also a side project of Pyra Labs.
Re:Ask HN: What do you do w/ a big idea, when you are workin... by DanielBMarkham on Saturday January 23, 2010 @04:16PM (#1072428)
I'd like to add to this -- in my opinion this goes in cycles, and it's important to ride out a couple of cycles. The cycles are ideating -- dreaming and researching how big a change the idea encompasses, and critiquing -- brutally throwing every criticism you can up against the idea.

So most of the time I'm not a huge believer in ideas -- it's execution that matters, and great teams can make money starting with mediocre ideas. But then every now and then one comes along that makes me pay attention.

So I go through the research, digging through competitors and looking at market-entry obstacles, partnership-leverage opportunities, and technical difficulties. If this is done correctly, at the end it's kind of like whittling away wood to make a whistle: At the end I have smaller piece of something that still looks cool and might work.

Then, if I want a whistle, I execute on that some, taking it to market or otherwise throwing as many rocks at it as I can. As I execute, it makes the idea more powerful -- only to go through another cycle of adjustment and pairing as the product meets the marketplace and you take another hard look at it.

In my opinion, after a couple of cycles you begin to either understand all the reasons why it won't work (and are executing anyway) or you understand all the awesomely cool things it might be (and have decided that it's not worth the effort).

Hope I explained myself. Tough concept to cover in a few words.