3 ms·

Something that really bugs me about this is the lack of examples where this succeeds. In my mind, a good one is Intel. To that end, how many microprocessor star

Something that really bugs me about this is the lack of examples where this succeeds. In my mind, a good one is Intel. To that end, how many microprocessor startups actually ended up succeeding at the same working on some fundamental and challenging things? Also, the thing about hard startups is there is a lot of luck involved. Counter to what the author is saying, you do need a degree or some level of measurable expertise to start a hard startup, because not only do you need the expertise, you also need the credibility to get people to work with you and you need that solid team to get funding. The reality is, many "hard" startups actually were things birthed by experts in a field that got lucky enough to have the right expertise at the right time, but also managed to derisk many aspects of the business by being experts.

ummonk 7y ago
It seems odd, given that highly successful startups overwhelmingly were "easy" startups (including one of the examples he mentions - Instagram).
- arkades 7y ago
  He didn't say easy startups couldn't be successful - just that Insta had to stand up to several orders of magnitude more competition than a 'hard' startup was/would have been. As a matter of survivorship bias, it would be shocking if there wasn't some surviving 'easy' photo-sharing startup. But looking at it prospectively, I think his point was: you could be fending off 9 competitors rather than 999 competitors, not knowing in advance if you're going to be the 1 survivor.
jpmattia 7y ago
> Counter to what the author is saying, you do need a degree or some level of measurable expertise to start a hard startup, because not only do you need the expertise, you also need the credibility to get people to work with you and you need that solid team to get funding. Interestingly, this criterion causes Microsoft to be in the "Easy Startup" column. (Facebook too, but that seems more correct.) The Elizabeth Holmes example is also interesting to consider. Lack of degree in a hard field was no barrier to raising a ton of money. > In my mind, a good one is Intel. To that end, how many microprocessor startups actually ended up succeeding at the same working on some fundamental and challenging things? Nit: Intel's first products were memory, not CPUs. https://www.google.com/search?q=intel+first+product&oq=intel+first+pro&aqs=chrome.2.0j69i57j0l6.5384j0j7&sourceid=chrome&ie=UTF-8 https://www.google.com/search?q=intel+first+product&oq=intel... )
- rafiki6 7y ago
  Elizabeth was doing a degree in a hard field and had the credibility of her supervisor to lean on, plus a whole host of questionable motivations from those who invested in Theranos. Funny enough, the degreeless founder in a hard startup turned out to be a fraud.

ummonk 7y ago

It seems odd, given that highly successful startups overwhelmingly were "easy" startups (including one of the examples he mentions - Instagram).

arkades 7y ago
He didn't say easy startups couldn't be successful - just that Insta had to stand up to several orders of magnitude more competition than a 'hard' startup was/would have been. As a matter of survivorship bias, it would be shocking if there wasn't some surviving 'easy' photo-sharing startup. But looking at it prospectively, I think his point was: you could be fending off 9 competitors rather than 999 competitors, not knowing in advance if you're going to be the 1 survivor.

jpmattia 7y ago

> Counter to what the author is saying, you do need a degree or some level of measurable expertise to start a hard startup, because not only do you need the expertise, you also need the credibility to get people to work with you and you need that solid team to get funding. Interestingly, this criterion causes Microsoft to be in the "Easy Startup" column. (Facebook too, but that seems more correct.) The Elizabeth Holmes example is also interesting to consider. Lack of degree in a hard field was no barrier to raising a ton of money. > In my mind, a good one is Intel. To that end, how many microprocessor startups actually ended up succeeding at the same working on some fundamental and challenging things? Nit: Intel's first products were memory, not CPUs. https://www.google.com/search?q=intel+first+product&oq=intel+first+pro&aqs=chrome.2.0j69i57j0l6.5384j0j7&sourceid=chrome&ie=UTF-8 https://www.google.com/search?q=intel+first+product&oq=intel... )

rafiki6 7y ago
Elizabeth was doing a degree in a hard field and had the credibility of her supervisor to lean on, plus a whole host of questionable motivations from those who invested in Theranos. Funny enough, the degreeless founder in a hard startup turned out to be a fraud.