etaoin shrdlu

Sort of the QWERTY or Dvorak of Linotype, back when newspapers had their text set manually. These were the most-frequently appearing letters (in descending order) in English-language texts. You could sometimes see the phrase ETAOIN SHRDLU - left in by mistake - in newspaper articles; it was meant as some sort of mark or signal to someone in the composing or typesetting process.

pingouin describes etaoin shrdlu pretty well, only I've seen a linotype machine. It's not a mark or a signal -- the letters e-t-a-o-i-n, s-h-r-d-l-u appear in that order on the keyboard. Linotype operators used to twiddle their fingers across these keys to see if the machine was working properly.

In that sense, etaoin shrdlu was the predecessor to hello world.

As often happens, we each have a piece of the story, and both pingouin and Sylvar are basically correct. Linotype operators who made a typing error would often run their fingers down the two middle columns of the keyboard to cast "etaoin shrdlu" on the next line. The resulting slug was intended as a 'flag' to call the compositor's (the fellow who assembled the pages for printing) attention to the erroneous line so it (and the "etaoin shrdlu") could be discarded (actually recycled, all Linotype type was melted down and re-used), but sometimes under the pressure of a deadline both the error and the "flag" would be missed and printed in the paper.

I am (or should I say was—since Linotype machines are hard to come by these days) a journeyman Linotype operator. I am offering an additional two cents worth of information about "etaoin shrdlu" (although the actual value of these comments are admittedly subject to the judgment of the reader).

The sequence of letters occasionally appearing in print were not necessarily the result of an error in the previous line. A good keyboardist, as he was assembling a line of matrices and spacebands in preparation for casting, would instinctively know if he had stroked a wrong key. Rather than sending the line with the error through the casting process, then reassembling the line from the beginning, casting the correct line, then (hopefully) remembering to pull out the incorrect line and discard it, he would simply reach up, pull out the incorrect matrix from the assembler, and place it to the right of his copy tray.

Obviously, after accumulating a number of these "mats," he would deem it time to return them to circulation. He could do this by getting up from his chair, going to the back of the Linotype, pulling the second elevator shifter out to clear the second elevator, and sliding the mats onto the second elevator bar to be redistributed into the magazine. Or he could simply place the random mats into the assembler, then fill the line with more random mats and space bands, and send the line through the normal casting and distributing process, taking care, of course, to see that this line, once cast, was pulled from the galley.

Having said all that, the length of line, hence the number of matrices in the line, might determine whether the operator followed the procedure elucidated by mjdixon1 above (e.g., single-column newspaper "straight" matter), or the one just suggested (book composition). While all this explaining may seem somewhat esoteric, if not indeed obscure, to those who are far removed from the wonders of hot-metal linecasting, perhaps one can get a glimmer of the complexity not only of the Linotype machine itself, but also the skill—and artistry—demanded of the competent operator. Vive la Linotype!