intrust

Since I have somewhat mixed metaphors, I feel obliged to give you an example. Digital sound or music production might start with a task-specific microphone and end with a task-specific loudspeaker. Digital image processing might start with a task-specific camera and end with a task-specific plasma screen or printer. These devices are the first-in and last-out elements of the production process. Once pictures and sounds are captured and converted into dW3, their task-specificity makes no special demands on the hardware of the computing system¹. They share memory, keyboard and mouse, processing power, graphic interface, and a host of other functions available on the generic, general purpose computer.

Inside the computer, task-specificity is maintained by software applications designed to work respectively on sounds or pictures, on top of an operating system that is common to them both. But even at application level commonality plays its part , since many of the software modules these applications are built of are the same for both tasks.

This has not always been the case, the initial computer based production tools for calculation, word processing, preprint layout; sound editing, picture editing, CAD (Computer aided design), CAM (computer aided manufacturing), and video editing, all arrived on the market in the form of non-modular task-specific dedicated machines².

When each of these devices was developed and first introduced, they exhibited superior performance and added additional features to the previous way of “doing things by hand”. They were also better than the software-only products running on general purpose computers that were released in their wake.

But in each case, the general purpose computer, fuelled by commonalization and commoditization, made these task-specific tools obsolete, and software-only products, running on the general purpose computer, eventually drove task-specific machines out of the marketplace.

// posted by Greg FitzPatrick @ 14:13