12 ms·

@patio11: may we get a post-mortem on Starfighter, and more details on what you learned about startups?

simonswords82 10y ago
Not wanting to do the concept for Starfighter any injustice but I get the feeling it's as simple as they spent too much time on the tech and not enough on the business. A very easy mistake when the co-founders are technically minded.
- kbody 10y ago
  That too, but personally I think that it was targeting way to limited subset, required practically non-existant time if you already have a job, so it was mainly targeting capable people between jobs or college students. Even though I loved the subject of algotrading, there was no time for me due to work etc.
  - simonswords82 10y ago
    Totally - the business model just didn't hold up unfortunately.
  - engi_nerd 10y ago
    > required practically non-existant time if you already have a job This. I had Starfighter bookmarked to dive into this Christmas during my holiday break. No time before then.
  - r_smart 10y ago
    This was precisely it for me. I have been excited for this project forever, but just haven't had a chance to filter it up to the top of my priority queue. Sadly, sometimes things remove themselves from your todo list.
  - Bartweiss 10y ago
    I was briefly sad to see that Starfighter was done, but then I realized that despite reading the rules and bookmarking it, I had never actually played the thing. Very cool idea, but realistically I can only spend so much time on dev stuff and it wasn't going to beat out my other priorities.
- afarrell 10y ago
  But strange for this developer whom I wouldn't consider to be naive in regards to business.
  - afarrell 10y ago
    On second thought, s/strange/interesting/. Failure happens. I would be interested in reading a retrospective.
- tptacek 10y ago
  It turns out --- I'm speaking for at least Erin and I, but I suspect to an extent Patrick too --- we don't have the stomach for the contingency recruiting model. Referring someone for an interview and watching them get turned down was, for me at least, about 80% of the psychic cost of applying myself and getting turned down. Fuck that. It's a job that is possible to do and possible to do well and my life is too short (takes bite of ham sandwich) to spend it getting good at that. So: I am still in love with the concept and still pursuing it, but I have fallen totally out of love with that particular business model and am happy to blowtorch it out of my life. More writeup to come! Patrick's at Stripe now; Erin & I have more news coming. Ironically: expect more challenges from us, not less. :P
  - angersock 10y ago
    Was there a pattern in them getting turned down, like that applicants were too expensive or had the wrong skillset or something? I stopped referring people to a former employer because I saw how much they jerked around people during the pipeline. Good luck on the new news, and thanks for getting Starfighter as far as you did. :)
    - tptacek 10y ago
      We're not done yet!
      - dennisgorelik 10y ago
        Are you still working on StartFighter but with different focus and without Patrick?
    - JamesBarney 10y ago
      Totally Speculating but I bet it was that software companies have enormously high false negative rates. Each company has it's own set of weird totally random interview biases. So Thomas kept on finding awesome recruits who were getting turned down for totally dumb reasons. I imagine it's like the HomeBrew situation every day except 10 times worse because you recommended the candidate and the interviewer is your client. And unfortunately it's hard to make a business of telling your client they're doing it wrong. This is all compounded by the fact that the devs who benefited most from their service were not Christopher Hemsworth looking Phd, ex-googlers who could get an interview with a phone call. Then come in with so much positive bias that as long as they didn't poo themselves during the interview they'd get hired. It was for people who were extremely skilled, but lacked the right pedigree or interviewing skills to be able to get the jobs they were very qualified for. Curious to see what the real answer is.
tptacek 10y ago
There's one coming.
hendzen 10y ago
Perhaps Starfighter failed because the cofounders spent too much time commenting on Hacker News?
- kbenson 10y ago
  Since that amounts to some of the best marketing they could do to some good leads, I doubt it.

Not wanting to do the concept for Starfighter any injustice but I get the feeling it's as simple as they spent too much time on the tech and not enough on the business. A very easy mistake when the co-founders are technically minded.

kbody 10y ago
That too, but personally I think that it was targeting way to limited subset, required practically non-existant time if you already have a job, so it was mainly targeting capable people between jobs or college students. Even though I loved the subject of algotrading, there was no time for me due to work etc.
- simonswords82 10y ago
  Totally - the business model just didn't hold up unfortunately.
- engi_nerd 10y ago
  > required practically non-existant time if you already have a job This. I had Starfighter bookmarked to dive into this Christmas during my holiday break. No time before then.
- r_smart 10y ago
  This was precisely it for me. I have been excited for this project forever, but just haven't had a chance to filter it up to the top of my priority queue. Sadly, sometimes things remove themselves from your todo list.
- Bartweiss 10y ago
  I was briefly sad to see that Starfighter was done, but then I realized that despite reading the rules and bookmarking it, I had never actually played the thing. Very cool idea, but realistically I can only spend so much time on dev stuff and it wasn't going to beat out my other priorities.
afarrell 10y ago
But strange for this developer whom I wouldn't consider to be naive in regards to business.
- afarrell 10y ago
  On second thought, s/strange/interesting/. Failure happens. I would be interested in reading a retrospective.
tptacek 10y ago
It turns out --- I'm speaking for at least Erin and I, but I suspect to an extent Patrick too --- we don't have the stomach for the contingency recruiting model. Referring someone for an interview and watching them get turned down was, for me at least, about 80% of the psychic cost of applying myself and getting turned down. Fuck that. It's a job that is possible to do and possible to do well and my life is too short (takes bite of ham sandwich) to spend it getting good at that. So: I am still in love with the concept and still pursuing it, but I have fallen totally out of love with that particular business model and am happy to blowtorch it out of my life. More writeup to come! Patrick's at Stripe now; Erin & I have more news coming. Ironically: expect more challenges from us, not less. :P
- angersock 10y ago
  Was there a pattern in them getting turned down, like that applicants were too expensive or had the wrong skillset or something? I stopped referring people to a former employer because I saw how much they jerked around people during the pipeline. Good luck on the new news, and thanks for getting Starfighter as far as you did. :)
  - tptacek 10y ago
    We're not done yet!
    - dennisgorelik 10y ago
      Are you still working on StartFighter but with different focus and without Patrick?
  - JamesBarney 10y ago
    Totally Speculating but I bet it was that software companies have enormously high false negative rates. Each company has it's own set of weird totally random interview biases. So Thomas kept on finding awesome recruits who were getting turned down for totally dumb reasons. I imagine it's like the HomeBrew situation every day except 10 times worse because you recommended the candidate and the interviewer is your client. And unfortunately it's hard to make a business of telling your client they're doing it wrong. This is all compounded by the fact that the devs who benefited most from their service were not Christopher Hemsworth looking Phd, ex-googlers who could get an interview with a phone call. Then come in with so much positive bias that as long as they didn't poo themselves during the interview they'd get hired. It was for people who were extremely skilled, but lacked the right pedigree or interviewing skills to be able to get the jobs they were very qualified for. Curious to see what the real answer is.

tptacek 10y ago

There's one coming.

hendzen 10y ago

Perhaps Starfighter failed because the cofounders spent too much time commenting on Hacker News?

kbenson 10y ago
Since that amounts to some of the best marketing they could do to some good leads, I doubt it.