: 1: Increasing the number of processors.

: 2: Increasing the variety of processors.

Some of our hardware is getting pretty good, but it... doesn’t make economic sense for most people. ...[O]ur zomes and heaters and so forth do not yet compete on a dollar basis with... conventional counterparts. It’s very exciting intellectually to work with these ideas but their validity will not really be proven until they start to replace... things they’re meant to replace.

Over and over, people try to design systems that make tomorrow's work easy. But when tomorrow comes it turns out they didn't quite understand tomorrow's work, and they actually made it harder.

The only problem is [that] the accelerator for this is about 10 times more powerful... than we can currently make. So there's lots of challenges for people like me who design accelerators, to try and come up with ways of making them more and more powerful, for very good reason.

People, ideas, hardware—in that order.

Programs today get very fat; the enhancements tend to slow the programs down because people put in special checks. When they want to add some feature, they'll just stick in these checks without thinking how they might slow the thing down.

People who are really serious about software should make their own hardware.

: 3: Increasing the number of connections between processors.

By choosing to focus on computers rather than the tasks we wanted done, we inherited much of the baggage that had accumulated around earlier generations of computers. It is more a matter of style and operating systems that need elaborate user interfaces to support huge application programs. These structures demand ever larger memories and complex peripherals. It's as if we had asked for a bit of part-time help and were given a bureaucracy.

Sometimes one can improve the theories in the sense of discovering a quicker, more efficient way of doing a given calculation.

In their work, designers often become expert with the device they are designing. Users are often expert at the task they are trying to perform with the device. [...] Professional designers are usually aware of the pitfalls. But most design is not done by professional designers, it is done by engineers, programmers, and managers.

Sometimes, a deeper order—a better fit to a purpose—is achieved through simplification rather than further increases in complexity.

If you try to fiddle around with your code to make it efficient, that's the wrong thing to do. Keep on track with the latest hardware. That's the right thing to do.