Page:The practice of typography; correct composition; a treatise on spelling, abbreviations, the compounding and division of words, the proper use of figures and nummerals by De Vinne, Theodore Low, 1828-1914.djvu/309

From Wikisource
Jump to navigation Jump to search
This page has been proofread, but needs to be validated.
Variable cost of proof-reading
295

Ordinary news work, for the most part, receives but one reading. Sometimes the errors marked on the first proof are revised on another proof, but this proof is not always re-read. Sometimes revising is done in the metal. In the cheapest forms of hurried auction-catalogue printing the composition is not even proved on paper: the copy-holder reads aloud from the copy while the corrector follows him, reading from the type on the galley and correcting, as he proceeds, the grossest errors only. Reading so done is unavoidably imperfect, but the scamped method saves time and largely reduces cost. This is one way to produce cheap composition.

Every book of reference or authority should be read on three or more proofs. When the author's proof has to be read by many experts, as is usual, duplicates are taken after each correction of the previous proof, and each duplicated proof receives an entirely new reading. The cost of reading and revising with this care is large, usually about one half as much as that of the first type-setting. Some pages will cost more than the first type-setting.

The slighted catalogue reading which costs about one tenth that of type-setting, and the careful dictionary reading which costs "more than first typesetting, are the extremes of book-work. The cost of the reading of the ordinary novel or descriptive book cannot be fixed at any definite point between the extremes: it is small when it is a strict reprint, large when it is in manuscript and not entirely in