HOME / articles:

Cogito Auto SumWhat less can we say? Computers have the answer.

By Karenna GorePosted Sunday, Feb. 9, 1997, at 3:30 AM ET

By Karenna Gore

RSS

Summary is the crux of civilization. Long before the Information Age, man struggled to digest the mounds of words contained in his religious, cultural, and legal canons. Fame and fortune awaited those who could reduce windy texts to bite-sized portions. One of the all-time master summarizers was Jesus, who, in the days before Cliff Notes, condensed the 592,439 words of the Old Testament into two nuggets. "On these two commandments hang all the law and the prophets," he announced. "Thou shalt love the Lord thy God with all thy heart, and with all thy soul, and with all thy mind," and, "Thou shalt love thy neighbour as thyself."

Or at least, that is what Matthew said Jesus said. Matthew might have been summarizing, too.

Despite Jesus' excellent example, wordiness not only persists but multiplies as new volumes enter the marketplace of ideas. Besieged with information and short on time, what is a person to do? Luckily, whole industries exist to serve consumers nutshells and breviloquence: Nexis, Web search engines, executive summaries, cheat sheets, and the Reader's Digest.

To that list of lifesavers, add an upstart, "AutoSummarize," a new function in the Word 97 word processor (which is made by the same company that brings you Slate). Last week, New York Times reporter Denise Caruso called AutoSummarize a "stunning" technological achievement. It is not Slate's function to tout Microsoft products, but AutoSummarize piqued our curiosity.

How does AutoSummarize work? According to Ron Fein of the Word 97 team, AutoSummarize cuts wordy copy to the bone by counting words and ranking sentences. First, AutoSummarize identifies the most common words in the document (barring "a" and "the" and the like) and assigns a "score" to each word--the more frequently a word is used, the higher the score. Then, it "averages" each sentence by adding the scores of its words and dividing the sum by the number of words in the sentence--the higher the average, the higher the rank of the sentence. "It's like the ratio of wheat to chaff," explains Fein.

AutoSummarize can summarize texts to 10 sentences, 20 sentences, 100 words, 500 words, or various percentages of the original copy.

Inspired by Caruso's example, we turned the powers of AutoSummarize on four basic documents of Western Civilization and four contemporary texts:


The Ten Commandments	The Constitution	Behind the Oval Office	Beowulf

The Contract With America	The Gettysburg Address	Martha Stewart's Salmon Recipe	Harry Shearer's Feb. 4 "O.J. by the Sea" Dispatch