N INE of 14 companies that manufacture pacemakers in the United States now provide models that are “programmable”; most, if not all other companies, including those from overseas, plan to introduce their own versions in the near future. The term “programmability” when referring to noninvasive control of a pacemaker, was first used when the Cordis Omnicor series was introduced in 1973.’ Even earlier, however, there were pacemakers in which the output could be changed, most notably the Medtronic Model 5870 in which either the output or rate could be altered by rotation of a needle that was inserted through the skin into a self-sealing nipple-like receptacle connected to a potentiometer. Earlier still, a permanent change in pacemaker rate or output had been achieved by means of a special external magnet placed over the implanted pulse generator (General Electric, Vitatron). This magnet method of single step change in a pacemaker function is still commonly used for other purposes. A more recent magnet application, and one that offers a truly noninvasive rate and output adjustment, was provided in another Medtronic pacemaker, model 5963, operated by rotation of an external bar magnet over the pacemaker. The magnet, coupled magnetically to another bar magnet within the pacemaker, rotated the potentiometer. Noninvasiveness is now a primary criterion of the definition of pacemaker programmability, but it is not the only criterion. A more satisfactory definition must include permanency (until otherwise altered); it may include, in some interpretations, taking external commands to transmit information or to take a definitive, but transient action, such as the discharge of a burst of rapid impulses. (Actions such as this have come to be called “programmed stimulation.“) Still more troublesome for the lexicographer is the definition of effects produced by the application of a bar magnet to most sensing pacemakers; this noninvasive manuever will produce a semipermanent change in pacemaker function, usually conversion to fixed rate (VOO). There have been several attempts to define programmability, such as: “A deliberate alteration of one or more pacemaker parameters to predetermined values.“3 or, “a non-invasive, stable, reversible change in an operating parameter of an implantable pulse generator.“4 In neither case were all the requirements of an adequate definition met. Nor does our own definition fill the bill: “The capacity to alter, noninvasively and semipermanently, one or several aspects of the function of an implanted pacemaker.” Perhaps the best definition to date, from an anonymous member of a lecture audience is: “Controllable adjustment of a pacemaker function that persists after the program is activated.” Doubtless, through common usage, we shall all come to understand what is meant by programmability: a noninvasive alteration of pacemaker functions, and permanency until otherwise “reprogrammed.” The definition should not include such features as pacemaker “interrogation,” by which , through noninvasive means, information will be transmitted from the pacemaker to an external receiver. Nor should it include “automatic” pacemakers that soon will be with us. (These devices will be touched upon below in a discussion of future pacemakers.) We are therefore left with four new terms that may be a part of a future pacemaker glossary: temporary pacemaker intervention (magnet application), pacemaker programming, pacemaker interrogation, automatic pacemakers.