Blown To Bits

Network effects

Monday, July 7th, 2008 by Harry Lewis
In buy free zoloft best price jelly the long term, people with untreated RA may experience worsening buy generic cafergot problems disability and joint deformity due to increased joint damage and cheap retin-a inflammation. However, over time and with treatment, individuals may find ampicillin prescription they can do these exercises again without pain. Your doctor cipro vendors will typically start by prescribing you the initial recommended dosage buy cheap zithromax online for your condition. Extreme heat may also dry a person's order viagra on internet airways, potentially causing bronchoconstriction, which is the tightening of the order discount viagra airways. If kidney stones remain untreated, they can block the online cialis urinary tubes, which increases the risk of infection and can buy cheapest cialis on line put a strain on the kidneys. People ask their doctor, tizanidine for order dermatologist, or rheumatologist's office if they know of any local psoriasis.

There is a good short article on the NYT Business page today about the ways in which Microsoft and Google have made network effects work to their advantage. A network effect is simply a situation in which having more people use your product makes it more valuable for other people to use it too, causing its popularity to snowball. Bill Gates is credited as the master of network effects, having built the Microsoft empire on the foundation of Microsoft’s operating system. Google has no such single control point, goes the argument, because of the Internet’s open standards, but has nonetheless been quite successful at exploiting “softer” network effects.

As I was cleaning up some old files I ran across a compelling example of the way network effects have changed the personal computer industry. In early 1984, as personal computers were becoming common at Harvard, I did a campuswide survey to find out what machines students had. 54 students said they owned personal computers and 32 of those said they had them at Harvard. These numbers are surely underestimates; the survey was unscientific and there was no reward for participating. But the distribution is fascinating:

8 Apple; 10 IBM; 4 Tandy; 4 Commodore; 5 Atari; 1 Zenith; 4 TI; 3 DEC; 2 Osborne; 4 Kaypro; one each HP, Sinclair, Brothers, Actrix, Corona, Ohio Scientific, Sol20, Timex, and NEC. I remember preparing the report itself on a Heathkit Z80 machine I built at home.

Now that was a Cambrian period in the evolution of the industry. This was 9 years after Microsoft had been founded, and there was still plenty of competition. But the incompatibilities made fertile ground for de facto standards to emerge, and Gates’ company tilled that earth with amazing skill.

Comments are closed.