Harrison's eyes burned. He would have to quit pretty soon and dictate the report. There wasn't any use in trying to go beyond a certain point. You got so damned tired you couldn't think straight. You might as well go to bed and rest. Bob Mills had gone long before.
He poured over the blueprint again, striving to concentrate. Why in hell had he not given up altogether? What possible contribution could an engineer make toward the solution of such a problem?
Construct this.
You simply made the thing according to a simple blueprint. You tried out what you got, found out what it was good for, found out then what was keeping it from doing that. You fixed it.
Well, the sword had been constructed. Fantastic effort had been directed into producing a perfect model of the print. Every minute convolution had been followed to an incredible point of perfection. Harrison was willing to bet there was less than a ten thousandths error—even in the handle, where the curves seemed to be more artistic than mechanical.
It is defective.
What was defective about it? Nobody had actually tried the ancient weapon, it was true. You didn't go around chopping people's heads off. But experts on such things had examined the twelve-pound blade and had pronounced it "well balanced"—whatever that meant. It would crack a skull, sever arteries, kill or maim.
Correct....
What was there to correct? Could you make it maim or kill better? Could you sharpen it so that it would go through thick clothing or fur? Yes. Could you make it a bit heavier so that it might slice a metal shield? Yes, perhaps. All of these things had been half-heartedly suggested. But nobody had yet proposed any kind of qualitative change or been able to suggest any kind of change that would meet the next admonition of the alien:
Correct that which renders it not useful.