5 ms·

It's definitely true, but most important is the last paragraph at the end. The best example is building enterprise vs. consumer applications. If you'd like a s

by strlen 19y ago

It's definitely true, but most important is the last paragraph at the end. The best example is building enterprise vs. consumer applications. If you'd like a safe way to get rich off a start-up, creating a consulting gig doing something "enterprisey" and then morph it into a start-up. In 5-7 years that could mean wealth. But this also means you'll be coding Java/.NET all of that time and are never going to build the next google. What you would build is something like the next Siebel: worse than what the open source community provides, but some suit somewhere will justify spending $1 mill on it as a "business decision".

donw 19y ago
I would advocate against going against the 'Enterprise' route, unless you already have contacts that can feed you your first few clients. Getting contracts at the sorts of companies that buy 'enterprise' systems is much more a matter of being connected than software quality. This should not come as a surprise. Taking a million-dollar wager on an unknown is corporate suicide. A million-dollar contract with an established partner that goes tango-uniform is just a 'bad call'.
- strlen 19y ago
  You need contacts, but you don't need to market your business (or hope your business becomes viral). You can get contacts by starting off doing consulting work. You'd only need to hire (or have as a co-founder) a business/sales person, you won't need to hire a marketing person. Nor do you begin with million dollar deals, first customers can be cheaper, just to establish a clientèle base/relationships. The fact you said it's a matter of contacts vs. software quality is a point I forgot to mention: as Ben Stein said, you don't need to be great to do this. You just need to follow directions. pg makes the same recommendation (go into enterprise) for those who do not think they are smart or driven enough to build consumer web applications.
  - donw 19y ago
    All very good points. I also think we're using different definitions for the word 'enterprise'. 'Enterprise' software is, to me, a behemoth application constructed by committee, designed to Do Everything All At Once, but which in reality does Very Little. Not all business software is enterprise software.
  - bigbang 19y ago
    "pg makes the same recommendation (go into enterprise) for those who do not think they are smart or driven enough to build consumer web applications." what makes u think that enterprise stuff doesnt need smartness? vmware? oracle? fwiw, ppl go where the money is and/or where the barrier is low, not where its smarter. for enterprise barrier is too high to get customers or even your customers are evry limited and you have to spend lots of time building your product unlike in web where there are lots of "platforms" like ror,adsense etc etc.
    - strlen 19y ago
      enterprise can benefit greatly from smartness (which is why vmware is such a success story). but more than just a smart idea, enterprise requires perseverance and a longer term commitment (you can't just quickly build a site and see if it floats or sinks in the next n months). for a consumer application, you have to offer something of immediate value (whether through a technological advance or not) or you have to heavily market it. no matter how long your commitment is to an idea, if it can't be marketed or isn't viral it won't catch on (versus with enterprise, where you are working with few high-value clients until you develop a product to match their needs / you build up a client base).
    - nostrademons 19y ago
      A bunch of my friends were early VMWare customers (1999/2000). They were almost totally a consumer/small business company then. VMWare Workstation was their first product, and they only got into server virtualization and enterprise solutions after the dot-com recovery, when folks realized "Hey, our servers are sitting idle most of the time, let's put a bunch of additional virtual servers on it." They've followed the path of many, many other successful disruptive innovations. They come out with something that's innovative & technically hard, release it as a low-margin consumer product, and then gradually move up-margin into enterprise systems as their product gets better & more widely known.
  - nreece 19y ago
    Just out of curiosity, which one of PG's essay recommends that (going enterprise)?
    - strlen 19y ago
      <blockquote type="cite"> If you don't think you're smart enough to start a startup doing something technically difficult, just write enterprise software. Enterprise software companies aren't technology companies, they're sales companies, and sales depends mostly on effort. </blockquote> See: http://www.paulgraham.com/notnot.html http://www.paulgraham.com/notnot.html

by strlen 19y ago

donw 19y ago

I would advocate against going against the 'Enterprise' route, unless you already have contacts that can feed you your first few clients. Getting contracts at the sorts of companies that buy 'enterprise' systems is much more a matter of being connected than software quality. This should not come as a surprise. Taking a million-dollar wager on an unknown is corporate suicide. A million-dollar contract with an established partner that goes tango-uniform is just a 'bad call'.

strlen 19y ago
You need contacts, but you don't need to market your business (or hope your business becomes viral). You can get contacts by starting off doing consulting work. You'd only need to hire (or have as a co-founder) a business/sales person, you won't need to hire a marketing person. Nor do you begin with million dollar deals, first customers can be cheaper, just to establish a clientèle base/relationships. The fact you said it's a matter of contacts vs. software quality is a point I forgot to mention: as Ben Stein said, you don't need to be great to do this. You just need to follow directions. pg makes the same recommendation (go into enterprise) for those who do not think they are smart or driven enough to build consumer web applications.
- donw 19y ago
  All very good points. I also think we're using different definitions for the word 'enterprise'. 'Enterprise' software is, to me, a behemoth application constructed by committee, designed to Do Everything All At Once, but which in reality does Very Little. Not all business software is enterprise software.
- bigbang 19y ago
  "pg makes the same recommendation (go into enterprise) for those who do not think they are smart or driven enough to build consumer web applications." what makes u think that enterprise stuff doesnt need smartness? vmware? oracle? fwiw, ppl go where the money is and/or where the barrier is low, not where its smarter. for enterprise barrier is too high to get customers or even your customers are evry limited and you have to spend lots of time building your product unlike in web where there are lots of "platforms" like ror,adsense etc etc.
  - strlen 19y ago
    enterprise can benefit greatly from smartness (which is why vmware is such a success story). but more than just a smart idea, enterprise requires perseverance and a longer term commitment (you can't just quickly build a site and see if it floats or sinks in the next n months). for a consumer application, you have to offer something of immediate value (whether through a technological advance or not) or you have to heavily market it. no matter how long your commitment is to an idea, if it can't be marketed or isn't viral it won't catch on (versus with enterprise, where you are working with few high-value clients until you develop a product to match their needs / you build up a client base).
  - nostrademons 19y ago
    A bunch of my friends were early VMWare customers (1999/2000). They were almost totally a consumer/small business company then. VMWare Workstation was their first product, and they only got into server virtualization and enterprise solutions after the dot-com recovery, when folks realized "Hey, our servers are sitting idle most of the time, let's put a bunch of additional virtual servers on it." They've followed the path of many, many other successful disruptive innovations. They come out with something that's innovative & technically hard, release it as a low-margin consumer product, and then gradually move up-margin into enterprise systems as their product gets better & more widely known.
- nreece 19y ago
  Just out of curiosity, which one of PG's essay recommends that (going enterprise)?
  - strlen 19y ago
    <blockquote type="cite"> If you don't think you're smart enough to start a startup doing something technically difficult, just write enterprise software. Enterprise software companies aren't technology companies, they're sales companies, and sales depends mostly on effort. </blockquote> See: http://www.paulgraham.com/notnot.html http://www.paulgraham.com/notnot.html