Abalance must always be struck be tween keeping secret a technical ad vance that cannot be patented and telling the world about it. As their paper in this issue of Interfaces shows, Epstein et al., the winners of the 1998 Franz Edelman com petition, have been particularly open in making their methods available to the world, including their competitors. Many of those present at the Edelman contest presentation commented on this openness, and it made me ponder some more on an issue I have dealt with in various ways for over 30 years. A hard-nosed manager can easily take the position that since its stockholders have paid for a technical advance the com pany should keep it secret to get maxi mum advantage from it. He or she may legitimately worry that making the ad vance public will help the company's com petitors become tougher competitors. This is a selfish position and one that is eco nomically inefficient (all of the theorems of classical economics notwithstanding). Sometimes, guarding company secrets may be appropriate; a manager is, after all, an agent of stockholders, who may well be selfish. However, often it is short sighted even from the point of view of the stockholders. In the 1960s and 1970s, Shell Develop ment Company, where I worked, had a rule of thumb for clearing technical pa pers. While commercial considerations led it to make exceptions in both directions, Shell's basic rule was that a paper would be cleared after it had been available inter