Hacker News Reader

Hacker News

Comment by liara_k | Hacker News Reader

Parent Full thread

liara_k

> unless google+apple conspire to regulate the market to prevent others from entering

Not to pick on you specifically, but this view is way too wrong to be so common.

The information economy is fundamentally different from the industrial economy for (at least) two reasons:

- Extremely high capital intensity

- Positive reinforcement through network effects (same and cross-side)

As a result, market concentration in information and technology markets is extremely high compared with other industries. It's not a conspiracy, and it's not something anyone can change. It's just the nature of the product itself.

There are a few strategies to combat it (open source, adversarial interop, etc...) but "just build your own" isn't one of them. Because it's almost always impossible in practice.

View on HN

Is it really high capital intensive? I make websites and i m not rich. Network levels are strong and global, but they do come and go every few years, its unlikely that facebook will be as long lived as disney or mcdonalds.

A mediocre engineer is worth on the order of a million bucks a year to one of these big companies... one engineer can create a reasonably large amount of 'capital' - that doesn't make it cheap.

Yeah, you could build that capital for yourself rather than for a big company, if you wanted to; that doesn't mean it wasn't expensive.

That, and I think that big successful projects take a lot more than just someone who can build a decent website. Take friendster... one of my favorite examples 'cause it was one of the real-world cases of a company dying due to insufficient resources being put into systems administration and scaling. They were in the right place, at the right time, with the right product to become what facebook is now; they got traction, but due to performance and reliability issues, they ended up leaving an opening for Facebook to come in and eat their lunch.

And even getting to the point where facebook was is really hard.

I mean, it happens; Craigslist is a famous example, but usually you need more than one engineer worth of capital to build one of these things, and you need more than just engineers.

> if you wanted to; that doesn't mean it wasn't expensive.

it wasn't, at all, if you re comparing it to almost any other industry sector

I dunno. every time I step out into other industries, I'm shocked at how little even skilled people are paid. I dunno much about animation, but I'm pretty sure you could setup, say, a small publishing house and support it for a few years on a million bucks. I'm certain you could start up a restaurant on that kind of scratch, and if you were the business type, you could turn that into a chain. I mean, sure, you'd need more for promotion/advertising, but that's just as true of the tech industry.

I mean, sure, a business starting with a million bucks is unlikely to become a multi-billion dollar business, but... I think it's a reasonable amount to start a reasonable business... and I think that's how McDonalds, at least, got it's start.

> Is it really high capital intensive? I make websites and i m not rich.

If you work in the industry and aren't familiar with concept yet, it's something that you would greatly benefit by reading up on. Capital intensity is about the RATIO of capital costs to operating costs. Compare the cost of developing a website with the cost of operating it. In general, the former are much higher than the latter--though there are some exceptions.

Actually, the total (absolute, not the ratio) cost is related to a third factor I didn't even list--constant or increasing returns to scale. This means that products can be VERY expensive to develop before an extra dollar of investment won't result in more than one additional dollar of profit. No surprise that there's a synergistic relationship here with capital intensity (very low per-unit costs). This isn't going to be super-applicable if you're building websites for small companies, but if you consider how much it would cost to develop a commercial operating system that was on par with what's currently out there (Android/iOS/Linux/MacOS/Windows/etc...) it would many millions on the low end to billions if you couldn't rely on open source.

i m not in the industry but i couldn;t find a definition such as the one you re mentioning

> Capital-intensive industries use a large portion of capital to buy expensive machines, compared to their labor costs. The term came about in the mid- to late-nineteenth century as factories such as steel or iron sprung up around the newly industrialized world.

i dont know, computers are dirt cheap, and IT is famous for being particularly easy to enter into as a sector. I don't think i m the exception

There's more to "capital" than a physical machine.

Compare the cost of creating software (often many millions) to the per unit cost of copies of that software (zero).

Compare the cost of developing a new processor architecture and setting up the manufacturing plant (billions and billions) compared with the per-unit cost.

Compare the cost of building out a physical network infrastructure to cover a city (many millions) compared with the cost of connecting a single user to the box down the street.

Compare the cost to create all of Facebook's hardware and software infrastructure to the marginal cost of adding one more user to the platform.

Reply on news.ycombinator.com