Wolcott & Hilsenrath (1976) and the NBS/NTIS Encoding

Contents

1 - Wolcott & Hilsenrath

Norman M. Wolcott and Joseph Hilsenrath's 1976 paper A Contribution to Computer Typesetting Techniques: Tables of Coordinates for Hershey's Repertory of Occidental Type Fonts and Graphic Symbols is perhaps the best known study of Dr. Hershey's glyphs. It contains a brief (seven page) discussion, sixteen pages of examples typeset with the glyphs, a table of forty additional glyphs added by the NBS (but not their coordinate data), the literal coordinate data for the Hershey glyphs as a very long table of printed text, medium-sized images of each glyph in the Occidental repertory, an index to "alphabets" (really a sort of arrangement into fonts), and an identification of many, but not all, of the special characters in the glyph set.

2 - The NBS/NTIS Encoding

Note: The former US National Bureau of Standards (NBS) is now the National Institute of Standards and Technology (NIST). The NTIS (National Technical Information Service) is an entirely separate organization.

Wolcott & Hilsenrath's report contains, in printed form, data for the presumably then-complete repertory of occidental glyphs (plus images of, but no data for, several glyphs added by the NBS). The format of these printed data is quite simple, and quite appropriate for print. Each glyph is first identified by its number, followed by a colon. It's hard to tell from the report, but I suspect that eight characters are used for this, including the colon (spaces are significant in this encoding). This number is followed by a variable number of number pairs, each encoded in eight characters as a pair of right-justified printable signed decimal numbers separated by a space and followed by a colon. The first pair represents the left and right boundaries of the glyph, and the remaining pairs the coordinate pairs and special coordinates defining the polylines of the glyph (see the earlier chapter on The Glyph Grid and the NORC Encoding). Where the original NORC encoding used the value "-49" in constructing its two special coordinate pairs ((-49,0) for end-of-polyline and (-49,-49) for end-of-glyph), this NBS encoding uses "-64".

While the NBS/NTIS encoding puts the reserved coordinates ((-64,0) and (-64,-64)) outside of the 99x99 Hershey glyph grid, encoding new glyphs using the original reserved coordinates ((-49,0) and (-49,-49)) or coordinates outside of the original glyph grid would be incompatible with the original NORC encoding of the glyphs.

The NBS "Bibliographic Data Sheet" which accompanies this report indicates that the NTIS (not the NBS/NIST) distributed magnetic tapes containing these data in BCD (Binary Coded Decimal) or ASCII form. According to the remarks made by Holzmann in his USENET distribution, this distribution was identical in encoding to the printed form which appears in Wolcott & Hilsenrath's report. How this representation, which contains the colon, space, and minus sign characters, was represented in BCD I do not know.

Exploring Dr. Hershey's Typography
CircuitousRoot