English spelling and the computer
Mitton, Roger (1996) English spelling and the computer. Harlow, Essex: Longman Group. ISBN 0 582 23479 4.
The first half of the book is about spelling, the second about computers. Chapter Two describes how English spelling came to be in the state that it’s in today. In Chapter Three I summarize the debate between those who propose radical change to the system and those who favour keeping it as it is, and I show how computerized correction can be seen as providing at least some of the benefits that have been claimed for spelling reform. Too much of the literature on computerized spellcheckers describes tests based on collections of artificially created errors; Chapter Four looks at the sorts of misspellings that people actually make, to see more clearly the problems that a spellchecker has to face. Chapter Five looks more closely at the errors that people make when they don’t know how to spell a word, and Chapter Six at the errors that people make when they know perfectly well how to spell a word but for some reason write or type something else. Chapter Seven begins the second part of the book with a description of the methods that have been devised over the last thirty years for getting computers to detect and correct spelling errors. Its conclusion is that spellcheckers have some way to go before they can do the job we would like them to do. Chapters Eight to Ten describe a spellchecker that I have designed which attempts to address some of the remaining problems, especially those presented by badly spelt text. In 1982, when I began this research, there were no spellcheckers that would do anything useful with a sentence such as, ‘You shud try to rember all ways to youz a lifejacket when yotting.’ That my spellchecker corrects this perfectly (which it does) is less impressive now, I have to admit, than it would have been then, simply because there are now a few spellcheckers on the market which do make a reasonable attempt at errors of that kind. My spellchecker does, however, handle some classes of errors that other spellcheckers do not perform well on, and Chapter Eleven concludes the book with the results of some comparative tests, a few reflections on my spellchecker’s shortcomings and some speculations on possible developments.
|Additional Information:||We are indebted to the following for permission to reproduce copyright material: David Holbrook for samples of young people’s writing in Appendix Three from his English for the Rejected (CUP, 1964); Dr P.D. McLeod for figure 6.2 based on a diagram from ‘Overlapping mental operations in serial performance with preview: typing – a reply to Pashler’ by P. McLeod and M. Hume in Quarterly Journal of Experimental Psychology 47A (1): 193-9, 1994;|
|School:||Birkbeck Schools and Departments > School of Business, Economics & Informatics > Computer Science and Information Systems|
|Date Deposited:||19 Feb 2007|
|Last Modified:||17 Apr 2013 12:33|
Additional statistics are available via IRStats2.