back
51 comments
I would argue that the only skill a business owner really needs is the skill to identify, hire, and lead talent. Everything else is probably secondary. If you're not good at programming, learning to program in order to do it yourself is ridiculous advice - unless you spend ten years learning full time, you'll end up with a mediocre product.

Delegation works. Just do it.

So many people confuse "Let others do what they're good at" with "Let others do everything", though. Including your comment. No, talent identification is not the only skill a business owner needs. If they have a skill and you don't, why should they come work for you?

A business owner needs to be skilled enough in their particular specialty to get the ball rolling and the money flowing in. Then they'll have the resources and cachet to identify and hire talent. But if all they can do is find talent and tell it what to do, why should the talent bother listening?

I'm sure I could tell go up to Marc Andreesen or Steve Jobs or Larry Page and say "I've got this great idea. Come work for me." I also suspect that if I did this, I'd get this quizzical "Who the fuck are you and why should I?" look, and then they'd walk away.

Nostra, accidentally downmodded you when I meant to upmod. Damned interface on this phone sucks at this, sorry.
If they have a skill and you don't, why should they come work for you?

Because talented people don't spontaneously self-organize. Your example is a bit silly - you're talking about people with a lot of money, stellar reputation, and great leadership skills. Most skilled people don't have any of these things. Working for someone with leadership skills is the only option they've got. If you happen to have enough charisma to get people to buy into your vision, you really don't need to know how to code.

Because talented people don't spontaneously self-organize.

Yes they do. Game recognizes game. I always want to socialize and engage with people I deem technically competent.

Myself, and most technical people I know, would not defer judgement nor respect to someone trumping "leadership skills" as a reason to follow them. In fact, someone projecting that air would be an immediate target for mockery until they displayed technical competence.

"If they have a skill and you don't, why should they come work for you?"

I would only hire people that are more skilled than me on their areas.

They should come work for me exactly because I do this :)

Hopefully you'd be more skilled than them on your area, though.

I don't mind working for someone with excellent sales or product or visual design skills + good leadership. They're contributing something that I'm weak at. It's a trade.

I do mind working for someone with only good leadership skills. What do they do that I can't do myself? I have plenty of ideas for startups too; I could just go off and start one myself.

I think the author is trying to assert the idea of "not hanging around" waiting for someone to come and do it for you.

He's not negating the idea of delegation. I guess you just have to strike a balance.

There are many people out there who will wait days for someone to make a simple html change for them on their site, where they could just as well learn to do it themselves. Its not like they have to learn C# overnight.

How do you identify a good programmer if you don't know programming ?
A track record of producing good products would be my instinctive answer were I a non-technical person.

But since I'm a technical person, I'd ask them why manhole covers are round.

And hope they laugh and refer to Feynman
It's not immediately clear to me that technical people are any better at identifying good programmers than non-technical ones.
Some technical people are probably worse at identifying good programmers than non-technical ones. But overall, I'd say that the very best people at identifying good programmers have a strong technical background. However, would agree that the situation is hardly hopeless for non-technical folks. After all, as far as I know, Jason Fried didn't have a strong programming background, and he ended up with DHH.

I also agree with your top post. While it would be nice to have 10 years of software experience or the technical background to evaluate programming talent, you can still succeed by identifying and partnering with talented people.

One of my friends had this as an interview question "What was the last thing you programmed that wasn't for work or school?" He was looking for any answer and some enthusiasm. Might not work for everybody, but it was a neat question.
If you have someone to delegate to, delegation is great. And yep something like yourflipflops.com is a mediocre website, but its a website thats up and sells a product. Too many people can't say that because they refuse to do it themselves and get stuck waiting for someone else to help them out.

When it comes to programming, just learning to program is an incredible way to actually meet other people that are already on a path to getting much better than you at programming - doesn't mean you have to commit to it full time for 10 years.

Delegation does work when there is just too much to do, but I would also argue knowing at least a _little_ about programming would help you work with your team better. And that goes for anything in addition to programming. I don't really know how to program but I understand the basics from taking a couple classes in school. I don't try to give advice, but I at least can tell when something is going to be hard or easy to accomplish and can say more cogent things in discussions about new features/changes, etc.
I agree to an extent... (I personally am actually in the process of trying to forget that I know how to do the production work so that I can focus more on delegating and bigger picture issues.)

But on the other hand intimate knowledge of what exactly you're delegating can be crucial in many ways. Even if you're not a GREAT programmer, having had the experience learning the skills and building the prototype gives a more comprehensive understanding of what's needed to accomplish your goals

What if you don't have any money to hire people?
then I'd likely resort to getafreelancer or something and get the initial stuff done. There is quite a bandwith of talent and cheap, too.

Though, I wouldn't get excited about it. For an initial version, I guess, everything you get there might suffice; but I might turn to pros asap.

Yes,

If someone happens to have programming skill, it might expand into the skill of knowing which programmer to hire. But the skills the parent identifies are inherently the core skills - anyone growing in business would want to gain them and leave their other skills by the way side (you aren't going to run your business like a twelve year doing his homework, are you?).

Of course, I'd prefer not to live in the world of pure delegation so I'd rather go with being hired by someone with these skills. Oddly enough, the best managers of programmers actually don't jump in and program themselves!

Determination is a good thing, but determination without direction is not. If you're good at design, and bad at programming, why insist on doing the programming yourself? A designer-programmer team will almost certainly kick your ass.

If you have a particular skill that seems to be part of a generalized skill set (for example, if you're good at meeting new people, you're likely to be good at sales), you can test it. If you know that you're bad at math, bad at logic, bad at working on one task for more than a few minutes at a time, and bad at spotting your own mistakes, you might not want to try programming (or architecture) just to prove to yourself that you have the willpower to overcome your character traits.

I think you're right, but there's a third situation that works even better:

A great designer and a great programmer will work better than a great designer who has picked up some programming, but both will be put to shame by the team of a great designer who understands the challenges the programmer has, with a programmer who understands the challenges of the designer.

In other words: It's important to learn new disciplines, even if you never get near a professional level with them, because you'll work better with the people specializing in that area.

I think that's great advice, and it's something I'd love to work towards.

I suspect I'm speaking for a lot of us here when I say that I have absolutely no idea where to even start when it comes to design, though. And I don't mean learning CSS - that's the easy part (at least once you embrace how underequipped CSS truly is vs. the problems of layout and styling).

Are there any good beginner resources available that focus more on developing good designs and less on implementing them? It seems that most of the beginning design stuff assumes that you already know what you want things to look like and that you don't know how to build it; for me, it's usually the opposite problem.

Honestly, when you have a team of people with multiple talents, which talent each has becomes rather irrelevant.

A lot of times what happens is how well the team communicates.

Whether a designer understands the challenges of programming in general won't matter as much as where the designer is willing to listen to the particular challenges and needs of the programmer working on that particular project (and vice-versa).

I actually had this exact problem. I started out in print design - newspapers to be exact - 5 years ago. About 3 years ago I realized a lot of the ideas I had and wanted to pursue needed a web developer, and a web designer. But I was surrounded by copy editors and press guys (not that there's anything wrong with that...)

So I learned Flash. Programmed some terrible, which-button-did-I-attach-that-code-to games, then took a programming class at the community college. Learned about loops, and iteration and general object-oriented theory. I made better flash programs and built a portfolio that got me a job building websites. After which I learned how to actually build a web site using CSS and XHTML instead of flash.

Now, I'm surrounded by web developers, DBAs and generally tech-savvy folks, and I'm building one of my ideas with a very talented java-programming co-worker.

It's taken longer than I wanted, but I'm still learning, and it's insanely rewarding, even if our product ends up being a complete bust. I'll never stop learning and trying new things, even if I'm not so good at them. It opens doors and creates opportunity.

Yep, and that's why I think it's still a great idea to partner with people who are good at those skills. But if you are stuck at the finding a partner phase because you don't have friends with those skills and don't have any money, why get stuck? Just going through the motion of learning something you don't even want to learn is likely going to put you into circles of people WHO DO HAVE the passion in doing those things.

I've met dozens of people stuck at that finding someone else phase. Maybe they should look at it with a Zen attitude "to find a partner, become independent". I think many people dating and looking for others recognize that already. Needy/desperate people have a hard time finding those people they need and are desperate to meet :) So make yourself independent and those relationships will come easier.

I think you're looking at binary choices: give up, or forge ahead. Whereas I tend to see it as a question of picking among infinitely many desirable options. In that case, optimizing based on what you're best at is essential.

That actually answers the Zen question, too: don't become independent by learning how to solve every problem, but by getting into a situation in which the problems you're faced with are the ones you're best at solving.

Shortly: Play to your strengths.
This was an inspiring article for someone like me with an business degree and aspirations for creating a startup but without the programming ability. Would starting out learning a language like python be recommended as a good one in general for putting an idea into action? After some iteration and traction, I'd bring on a capable coding partner to do it right. FWIW, I've also heard that Python is useful for making scripts for automating tasks or improving productivity.
This route (learn to code it yourself) is the one I took. I now have a pretty technically credible* product with actual paying customers and everything.

The language I learned first was JavaScript because it's ubiquitous. There are a million tutorials to get you going.

A lot of JS tutorials suck, which is a disadvantage, but it gives you a pretty big confidence boost to be able to read a tutorial and think "that's moronic, I know 6 other ways to do that which will work better". Python tutorials tend to break out advanced/elegant/clever stuff really quickly which has the opposite effect.

I wrote a couple of little apps, then a prototype, then it felt like I knew enough to really get going.

I then picked up Python (actually Django) because the relative maturity of server side JS platforms isn't as good. Switching languages was easy.

The total time to get to that point was about 4 months (whilst working full time), but I had the advantage of living in a place where everything shut when it went dark at 7pm.

*Except for the bugs, and the scaling issues, and the bad design decisions I made early on. As far as I can tell I'm not the only person to have problems like this though, and refactoring is a good way to spend a plane ride.

Yes. Python is a great learning language, and useful for many things long term.
Python and Ruby are great first languages; they'll let you do all kinds of things, from quick scripts to web programming to GUI's. There are some things for which they're not ideal, but they're still probably 'good enough'.
Python is great. O'Reilly is coming out with a book soon, "Head First Programming" that teaches programming, specifically using Python. It could be a great first start for you.
I couldn't agree with the author more. The primary skill of a successful entrepreneur is the relentless ambition to just do it. Subsequently, after trashing my alma mater for the majority of my college career, I've come to realize that the biggest lesson I learned in life was through there. Which is of course the lesson on how to learn stuff that is really hard, but necessary to accomplish a goal.
I hear ya. I'm on the fence when it comes to Jason Fried's advice about dropping out of college. He's a big fan of convincing people to quit and start working, at least in the software industry.

I went to school as a chemical engineer, and though often it seems the education's specifics are a waste to me, I learned how to approach problems rigorously and, in your words, relentlessly. I'm not sure I'd have that same drive if I wasn't challenged like that from the teachers and the education.

This is fantastic advice for people who want to learn the new things that they need to learn. However, not everyone does, and that's perfectly OK. I know we're biased towards programming here, but let's flip your example around: I know how to code, but I don't know how to do enterprise sales. Yes, it behooves me to learn something about it, but if I can hire someone to do it for me, why shouldn't I? My goal is to always hire people smarter and more competent than myself at the thing that I'm hiring them to do. I don't need to be the best at everything; I'd much rather be the best at what I'm best at :)
Right, as long as you have the money to hire someone to do enterprise sales, great. I'm all for hiring people smarter than me and to do things I don't want to do. But most people starting a new project don't have money to do those hires.

Look at Paul Graham. When they started viaweb, Paul had to play sales guy. I believe he mentions in at least one essay, that it's not like sales was something he was chomping at the bit to go and do. But a fledging company can't go out and hire sales guys, so he stepped up and learned how to sell his product, and realized he was actually pretty good at it.

So this advice is to all those out there with no money who are stuck waiting for someone else to make their dreams come true. That's a no good place to be, and I refuse to be in it.

Being relentlessly resourceful doesn't always mean "do it myself". Sometimes it means you go get funding, sometimes it means finding someone to work for equity or revenue share, sometimes it means figuring out how to make money somewhere else until you afford to pay someone to do X.
I've actually found it surprisingly easy to sell our product. When you make "Something that people want"(tm) it just sells itself.

You just stand to one side and point out that they want it.

  "I have this great idea for a *video game*, I just need someone to build it. How do I find a *coder* and build this *video game* in a couple months?".
This happens so often on gamedev-related forums it's depressing. Pure beginners often lack the gumption, or perhaps just the knowledge, to realize that they could and usually should just do everything themselves, or at least learn how to code a game or use a game making application.

In these days of 200-person AAA development teams I can understand the sentiment that developers by nature are super-specialized, and that having multiple skillsets just doesn't compute.

I was expecting something completely different by the title. I was thinking of how, when you grow older, your curiosity dies because it takes too much effort to satisfy it.

This worries me. I have a friend who was saying the same thing, but he was uninterested in doing anything about it.

I think curiosity dies out for two reasons. The first is plain ol' brain chemistry. Secondly, when people get older and they sit down at a computer, they do so to get stuff done. When a kid sits down at a machine they explore without expectation of themselves. Adults have the expectation that - what they want to get done - will get done - and in a reasonable amount of time. However, some things do take more than a "reasonable amount of time", and frustration mounts because of this expectation. Soon, learning new things becomes associated with frustration. So relax a little, and explore a lot. Then read the docs. Or read the docs - and then explore. Just don't expect to be productive in the New-Software-Product in 2 hours.
The bottom-line in this article is that you need to take the steps necessary to get a project done. Even if this means self-educating, don't let that prevent you from accomplishing your goals.

In reality, this isn't always feasible. That's why it's so important to have good judgment about these things.

It reminds me of this quote:

God grant me the serenity to accept the things I cannot change; courage to change the things I can; and wisdom to know the difference. - Reinhold Niebuhr

[UPDATED] This isn't meant to be a religious statement, I'm just crediting the original. You can just as easily replace "god" with "imagination" or "consciousness" or whatever if that makes it more palatable.

If you know the stuff, do it yourself. If you don't know it, hire someone else to do it. If you don't have money, well, you find a cofounder.
The analogy at the top is flawed. The 7-year old has zero opportunity cost. His time is worthless. If you have a great idea, but also the burden of supporting a family, holding down a day job, or whatever, hiring somebody to build your idea might make good sense.
Is Inkling using market scoring rules for liquidity? I've been wondering about that for awhile, and the Hanson reference makes me wonder more.

(The LMSR is the only way I've managed to get a market work; I'm still totally stuck in automated market makers).

Yes, it uses Hanson's market scoring rule.
That's the same Robin Hanson I've been reading at overcomingbias and lesswrong. I wasn't aware of his work on prediction markets. Most interesting.
18