(For a catalog of this site, see Detailed contents.)

Welcome to the NIST Reference on Constants, Units, and Uncertainty, one of the technical activities of the Fundamental Constants Data Center of the NIST Physics Laboratory. The contents of this site have been prepared by Barry N. Taylor of the Data Center in close collaboration with Peter J. Mohr of the Physics Laboratory's Atomic Physics Division.

This site was designed, edited and programmed by Michael Douma under the direction of Drs. Taylor and Mohr.

All material has been authored or co-authored by Drs. Taylor and Mohr, with the exception of sections otherwise identified. The bibliographic database portion was designed and programmed by Michael Douma and Dr. Mohr in collaboration with Dr. Taylor, and is maintained by Drs. Mohr and Taylor. This web server is maintained by the Office of Electronic Commerce in Science and Engineering Data (ECSED) of the Physics Laboratory of the National Institute of Standards and Technology (NIST), an agency of the U.S. Department of Commerce.

For details about how this site has changed, see version history.

We welcome your feedback about this site and invite you to e-mail us with any comments or suggestions you may have.
 

Search Algorithms
The misspelling-forgiving search algorithms we use are based on phonetic and approximate matches. The fundamental constants search uses both these techniques, while the searchable bibliography uses only approximate matches. Phonetic matches were made possible using a modification of an algorithm called "SoundEx," which has been used for nearly a century to consolidate disparate spellings of surnames in census reports (see C code). Approximate spelling matches utilize the agrep algorithm by Udi Manber at the University of Arizona, widely available as C source code for unix platforms.

Acrobat PDF files
Most acrobat (pdf) files were authored in TeX in a unix environment, and Adobe Pagemaker and Illustrator on a Power Macintosh. Wherever possible, the typefaces are set in Times Roman to minimize file sizes by avoiding subset font files. In TeX, this laborious process is accomplished using MathTime fonts available from Y&Y which allow TeX to use postscript fonts in place of its default bitmapped fonts.

Searchable bibliographic database
Citations are stored in BibTeX format, with separate files for each citation. We generate html output by making customized BibTeX calls in a faux LaTeX .aux file, and parsing it with BibTeX on the fly.

Colophon
The logo and headline are set in Adobe Myriad, a Multiple Master typeface. Text is usually set in <FONT FACE="Myriad Roman, Helvetica, Arial, Avante Garde, GillSans, Geneva"> or <FONT FACE="AGaramond, Palatino, TimesRoman, Times"> in html, and Times Roman in the pdf files. HTML is edited with GoLive, BBEdit and vi, and graphics are edited with Adobe Illustrator and Photoshop on a Power Macintosh. CGI scripting is based on Korn shell, sed, awk, and perl scripts on an IBM RS/6000 AIX Apache server.