Flashback > 10 Years Ago
JUNE 25, 1992
Maybe the reason "fuzzy logic" has been slow in catching on in the United States is because it just hasn't been taken very seriously, even though Japan has fully embraced it. Maybe it's because the name is so peculiar. Whatever the rationale for this hesitation, many U.S. companies, spanning all types of technologies, are now looking at fuzzy logic to simplify some quite complex electronic functions.
... Almost any product with moving parts that require smooth operation is a candidate for fuzzy logic, especially if there's some type of complex algorithm behind it. Basically, fuzzy logic eliminates the traditional "yes-no" decision of conventional logic. (Technology Analysis, p. 37)
Flashback > 25 Years Ago
JUNE 21, 1977
Magnetic Bubble Memories, which have only recently taken on commercial applications, will replace recorders and computer-disc and drum memories in military aircraft and spacecraft as soon as possible. Several working systems are being readied for operational evaluation.
Both the Air Force and the National Aeronautics and Space Administration envision nonmechanical, nonvolatile memories as being part of digital data recorders in spaceborne sonobuoys; airborne warning; command and control systems; missiles; and even test sleds. The Navy is eyeing them for airborne surveillance systems and as computer mass memories in both surface ships and submarines. (News, p. 34)
Flashback > 40 Years Ago
JUNE 21, 1962
The growing interest in machine-controlled speech as an approach to electronic speech synthesis was evident at a meeting of the Acoustical Society of America in New York. A pitch-dependent method of compressing and expanding human speech into desired time slots was described as being superior to earlier time-dependent and speeded playback techniques
In the pitch-dependent method, an IBM 7090 data-processing system is used to convert the analog waveform of a speech sample to a digitized waveform. One segment is removed and the remaining segments are compressed. The resulting digitized waveform is converted to an analog waveform. The identifiability and aesthetic quality of speech processed in this manner reportedly are improved with the new method. (News, p. 28)