Cataloging Distribution Service

Special Notes: 1) The RFP will be issued on or about Wednesday, May 21, 2008. Proposals will be required in early June, 2008. If you would like a copy of the RFP, please send an E-Mail to [email protected] with your E-Mail address. 2) Please do not use any Maryland phone number or address which may appear in this announcement. Use only Mr. Scheuchenzuber's address, phone number, and E-Mail address located at NTIS in Springfield, Virginia. 3) Atlis Systems in Silver Spring, Maryland is the incumbent. Description of Requirement The Library of Congress has been generating the annual 7,000+ page 5-volume Subject Headings using a series of mainframe-based programs that have been in place since the 1970s. The last step of this series of programs is the creation of what is called a “Videocomp Driver” which is then written to magnetic tape. This tape was formerly sent to the Government Printing Office (GPO) for processing on GPO’s Videocomp* typesetting machine. The Videocomp at GPO had been specially programmed to include customized fonts necessary to typeset the extensive and complex diacritical marks required by the Subject Headings. Videocomp was retired and the Library of Congress now needs to find another methodology to produce the Subject Headings. In addition, the Library of Congress also retired their in-house mainframe system and the software would need to be entirely rewritten if the Subject Headings were still to be produced. The Library of Congress subsequently replaced their mainframe system with an XML-based workflow drawn directly from the MARC records when this rewrite became necessary. The Library of Congress is searching for an organization with a 1970s-era Videocomp typesetting machine with the appropriate font configuration or can offer an alternate solution. Bidders should know that the meaning of some of the code sequences in the data stream are context-dependent, i.e. the same code sequence in the input data may have several different meanings depending on where it appears in the coding hierarchy. For this reason, simple lookup and replace methodologies cannot correctly convert the data stream and the conversion program must be aware of the context in which a string appears in order to process it correctly. As noted above, the data includes a variety of diacritics: grave, acute, circumflex, tilde, macron, underscore, breve, superior dot, dot below, umlaut, hacek, circle above, double acute, cedilla (accent and not just the character), right hook, double dot below, circle below, double underscore, left hook, right cedilla and tilde spanning multiple characters. In addition to the standard accents outlined above, the fonts must support a variety of accents that are relatively rare such as the pseudo question mark, upadhmaniya, cantrabindu, high comma off-center, and high comma centered. The solution must offer a high-speed batch composition system that directly processes the XML and outputs PDF files. The system must properly handle all the Unicode characters and accents, embeds all necessary fonts, and automatically creates the multi-level continuation headings.