Microprocessor standards

Now that microprocessors have been with us for three or four years, we can stand back and observe some disadvantages of our free enterprise system, as well as obviously great advantages. The disadvantages tend to fall most heavily on the shoulders of the user, not the manufacturer. The proliferation of microprocessors with incompatible instruction sets, adds unnecessary chores to learning new mnemonics, and requires rewriting (at large expense) existing software. Another interesting situation can be seen in the case of the 8080 and the Z80 processors whose mnemonics differ for precisely the same instructions. The act of copyrighting mnemonics by microprocessor manufacturers places the problems in the user's laps in order to protect the proprietary interests of the manufacturer of the chip. Whenever standards are proposed, two reactions are voiced, polarizing the opinions. On the one hand we hear general disdain: "Who needs them?" or "Standards impede progress." On the other hand we hear cheers: "What took so long?" or "Lack of standards impedes progress." The field of microprocessors is not immune to this division. We will have standards in software as well as in hardware. Unlike the larger mainframe computers, where the entry costs are high enough to insure a virtual monopoly on software and interface standards by a single manufacturer, there are a half dozen or more significant contributors to microprocessor software for each chip type, and as yet none of their products are compatible with each other. There are hundreds of components designed for very few processors, and many of these cannot even coexist in the same system. Standards are inevitable because the users will not tolerate this chaos very much longer. The absence of standards for the large, batch computers today is simply due to the fact that the manufacturers fight standardization. Plus compatible peripherals are a touchy example. It is not in the mainframe companies interest to foster a standard interface. This reluctance to standardize extends to software aspects which directly affect a user such as job control streams, editor commands, disk I/O calls, tape formats, internal data representation, ASCII vs. EBCDIC, and pairing of source and re10catable files. These differences