#74 Road Kill on the Information Highway
This post is a summary of an internal memo written by Nathan Myhrvold in 1993 while he was working at Microsoft. The memo accurately captures the key technology trends of the time and how the PC/computing industry needs to strategise/respond to these trends.
He makes a stunning and prescient claim in this memo - "The day will come when people will say "hey, didn't Microsoft used be the company that made office software?" and the answer will be "yes, and as a matter of fact they still do, but that isn't what they're known for these days". Approximately 30 years later - Microsoft's intelligent cloud business is the most profitable and fasting growing business!!
Some of the keys points from the first section of the memo are listed below
Exponential growth in performance of computing (CPU speed, storage, RAM density) and decreasing costs for the same performance will enable solutions to hard problems.
His advice to the computing industry was to recognise which things scale with or faster than computing and what things do not. He highlights the failure of the mainframe industry to recognise the exponential growth in the performance of microprocessors and understand it's implications:
Performance of microprocessors would surpass that of mainframes and minis
Hardware should be decoupled from software because the driving forces are different
Software would be a central focus of value
He attributes the stupendous growth of Microsoft to the exponential growth curve of compute price/performance. Software can leverage the improvements in performance to increase value provided to end users - appeal to a wider audience (via improved UX), enable new applications, get better features.
He proposes a metric for Microsoft to measure its position in the market - fraction of worldwide CPU cycles consumed and data transmitted by Microsoft products.
For Microsoft to increase or maintain it's relative position in the market, he highlights what Microsoft needs to do:
Continue to bring new technology to existing products
Create new product lines to track the emergence of computing in new mass markets
He highlights the importance of having products that are relevant to the high volume segment of the market.
Another key observation that he makes is that an operating system architecture can easily last 15-20 years or possibly even more. Over that period of time the computing hardware that the OS runs on will increase in price/performance by a factor of 30,000 to a million, taking the combined platform to new realms and application areas.