Appeal 2007-2127 Reexamination Control No. 90/006,621 word processor with spelling and grammar checking. Also, Patent Owner appears to be mistaken in stating that a compiler performs spelling checking. Nothing in the description of the lexical analyzer in Aho & Ullman mentions spelling checking. The analyzer merely breaks a string of characters into groups, called tokens, which may be keywords, identifiers (variable names), symbols, and punctuation marks, and passes the string of tokens to the syntax analyzer—it does not check to see whether the keywords, for example, are spelled correctly. While, perhaps, the compiler might generate an error if a token is misspelled, it would be because the token is not recognized, not because its spelling has been looked up in a dictionary. One of the references cited in Mr. Reiffin's declaration states that collecting sequences of characters into meaningful units "can be thought to perform a function similar to spelling" (¶ 9) but this is not the same as spelling. Patent Owner argues that limitations to "words" and "sentences" are not new matter because "[h]high-level program source code comprises many ordinary English (or other natural language) words, using the same ASCII code used to represent natural language text in a computer" (Br. 95-96), such as "BEGIN," "END," "THEN," "WHILE," "DO," etc. (Br. 96; Br. 97)). Patent Owner argues that the Examiner erred in rejecting claims that contain limitations to "language code" and "natural language" because "the language used in source code of the illustrative embodiment is made up of 'words' — indeed, English language words — and so the phrase 'words of a language' is fully supported by the specification even under the Examiner's crabbed interpretation" (Br. 98) and "persons of ordinary skill would have understood that the disclosed invention was applicable to both natural and computer programming languages" (Br. 98). 139Page: Previous 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 Next
Last modified: September 9, 2013