The series "The Software Snarl" {front page, Dec. 9-12} did a good job of describing how software is developed and why the process is so difficult. I am a software developer, and I often have little success explaining my work to friends and relatives. Having said this, I must add that there is one serious flaw to the series. The last article made it appear that America has all the problems and Japan has all the answers.

This is terribly misleading. For example, the article describes flawless software as an attribute of Japanese VCRs and other electronics. IBM, among others, has shown that American companies can achieve this, too. The IBM Wheelwriter contains more than 50,000 lines of software. To date, no defect has been traced to the software. This was achieved through using a development process that emphasizes early defect detection to avoid the debugging problems the article described for Lotus 1-2-3/G. This process was an early version of Dr. Harlan Mills's "Cleanroom" development process.

There are more examples of successful development projects in this country, and researchers around the world are working on unsnarling the snarl. All American industry has a lot to learn from the Japanese about bringing the results of research into production. It is not fair, however, to give the impression that all American software is a mess and that all Japanese software is perfect. MIKE STARK Greenbelt