Opinion
No. 4459.
November 23, 1933.
Williams, Bradbury, McCaleb Hinkle, of Chicago, Ill., and Arthur M. Smith, of Detroit, Mich. (Lynn A. Williams, Curtis B. Camp, Clifford C. Bradbury, and C.W. Ooms, all of Chicago, Ill., of counsel), for plaintiff.
Fish, Richardson Neave, of New York City, and Stevenson, Butzel, Eaman Long, of Detroit, Mich. (Merrell E. Clark, William R. Ballard, M.R. McKenney, and Henry R. Ashton, all of New York City, of counsel), for defendants.
In Equity. Suit by Kellogg Switchboard Supply Company against Michigan Bell Telephone Company and another.
Decree for plaintiff in part and for defendants in part.
The amended bill of complaint in this case asserted infringement of six patents. Four of these, the patents to Winston, No. 1,197,166, to Currier, No. 1,316,477, to Sparks, No. 1,358,182, and to Shull, No. 1,566,622, were withdrawn prior to the opening of the trial in court. This leaves but two patents to be here considered, namely, No. 1,428,762, to Currier, and No. 1,438,170, to Currier.
Patent No. 1,428,762 relates to those features which for brevity have been referred to respectively as "noninterfering answering," "instantaneous disconnect," and "multiple line lamp recall." Some of the claims relied upon relate to the first of these features, some to the second, and others to a combination of the three. The second patent, No. 1,438,170, concerns what is known as "flashing recall."
Both of these patents have to do with the telephone equipment used in central office switchboards for interconnecting subscribers. The defendants' switchboard, which is alleged to infringe the "noninterfering answering" claims of patent No. 1,428,762, is the so-called No. 11 machine ringing board. Some of these same boards are equipped with the "instantaneous disconnect" and "multiple line lamp recall" features and therefore are alleged to infringe, in addition, the claims directed to these features of that patent. Others of these No. 11 boards are equipped with the "flashing recall" and are alleged to infringe some of the claims of patent No. 1,438,170, as are also certain of defendants' so-called No. 1 switchboards.
Patent 1,428,762.
This Currier patent, No. 1,428,762, relates to improvements in telephone exchange systems. The case is not free from difficulty. The patent had a long and stormy course through the Patent Office, and it is in an art which bristles with difficulties in comprehension and exposition. The application for the Currier patent was involved in several contested interference proceedings, including one between Currier and Johnson, whose application was owned by the Western Electric Company, the manufacturing subsidiary of the American Telephone and Telegraph Company. All of the principal prior art now relied upon by the defendants was considered by the tribunals of the Patent Office in connection with inter partes motions in these interference cases, and the claims of the patent were finally allowed after having been subjected to the test of various motions to dissolve and other adversary proceedings.
While the inter partes and ex parte proceedings in the Patent Office led to various amendments and substitutions of claims, the plaintiff stands upon the claims as finally allowed. There is no file wrapper estoppel which can be invoked by the defendants, despite the complication of the long and involved Patent Office proceedings. It seems clear in the last analysis that the patent claims as finally issued are entitled to be read and applied upon their own merits, and without too much regard for the pro and con arguments of the patent solicitors for Currier and his adversaries in the Patent Office. Byers Machine Co. et al. v. Keystone Driller Co., 44 F.2d 283, 284 (C.C.A. 6, 1930); Frey et al. v. Marvel Auto Supply Co., 236 F. 916 (C.C.A. 6, 1916); Bundy Mfg. Co. v. Detroit Time-Register Co., 94 F. 524, 542 (C.C.A. 6, 1899); A.G. Spalding Bros. v. John Wanamaker, 256 F. 530, 533 (C.C.A. 2, 1919); Auto Pneumatic Action Co. v. Kindler Collins et al., 247 F. 323, 328 (C.C.A. 2, 1917); Traitel Marble Co. v. U.T. Hungerford Brass Copper Co., 34 F.2d 670, 671 (D.C.S.D.N.Y., 1929); General Electric Co. v. P.R. Mallory Co., Inc., 298 F. 579, 585 (C.C.A. 2, 1924); Hoe Co. v. Goss Printing Press Co., 30 F.2d 271, 275 (C.C.A. 2, 1929).
When telephones first came into use, the number of subscribers was small, and one operator sufficed to make all of the connections between calling subscribers and called subscribers, each of whose telephone lines had but one terminal (or spring jack) in the switchboard.
But the number of telephone subscribers increased and the telephone exchange switchboard grew so large that all of the necessary interconnections could not be taken care of by one operator or one position, or indeed by the use of only a single spring jack terminal for each line. The art was therefore confronted with the problem of distributing the subscribers' incoming calls in such a way that the work of answering them and of making the necessary connections with called lines could be divided as equitably as possible among the several operators.
Before the date of Currier's application for patent it had been proposed to attain this result by automatic distributing switches which fed the calls evenly to the various operators. It had also been attained to a substantial degree in the defendants' No. 1 switchboard by the use of an "intermediate distributing frame" and by the use of multiple appearances of the line lamps. In this switchboard each subscriber's line had enough multiple spring jacks distributed along the face of the board so that any operator could reach one of them to complete a connection to any line called for by a subscriber. In addition a smaller selected group of these same lines terminated before each operator in "answering" jacks, each with a line lamp to indicate incoming calls; these being located below the jacks of the main multiple. The jacks in the main multiple had no line lamps. Except under special circumstances the calls were answered in the answering jacks and "completed" in the main multiple. Each operator could reach and serve any answering jack in her own position and those in the positions immediately to her right and left. The intermediate distributing frame was a long narrow rack with all of the subscribers' lines connected to terminals on one side and all of the answering jacks to terminals on the other side. By means of jumper wires between the two sides, any line could be connected to any answering jack. The management endeavored so to make these connections that the group of lines connected to the answering jacks before each operator would give her a fair average load of calls to be answered, due regard being given to the probable calling rate on the several lines selected for each group. Instead of having a single answering jack for each line, these boards frequently had two and sometimes three answering jacks for each line connected in multiple and spaced along the board so that each was available to three different operators. An incoming call might thus be answered by any one of three, six or nine operators (depending upon the number of multiple answering jacks used), the least busy operator being likely to take a waiting call. Thus a natural distribution among the team of three, six, or nine operators resulted, which supplemented the distribution effected by the intermediate distributing frame, taking care of some of the variation in the expected calling rate of the lines assigned to the several operators. Also, when there was no idle operator at the answering jacks of a line, the supervisor might call out the number of the line and it could be answered in the main multiple by any idle operator.
There was, however, no distribution generally between the teams of operators, except the relatively permanent one effected by the intermediate distributing frame. This involved soldered connections and it was not feasible to be continually changing them. Large variations in the calling rates on groups of answering jacks, due to seasonal or other causes, were taken care of by a rearrangement of the jumper wires. This switchboard contained no means to prevent two operators from connecting simultaneously to the same line by plugging into separate answering jacks to answer a call thereon. The chances of confusion from this cause were reduced by preferentially zoning the multiple answering jacks and by establishing rules as to the withdrawal by one operator when two got in together.
In addition to these No. 1 switchboards having two or possibly three appearances of separate answering jacks and lamps, there was in actual commercial use also, a switchboard at the Hollenden Hotel, in which it was attempted to avoid double answering by operators. That board had no separate answering jacks, the line lamps being associated with the jacks of the main multiple, and it had series jack contacts to prevent more than one operator from taking the call. It was, however, so arranged that one girl was always in the master chair, and she could always take a line away from any of the other girls who had seized that line and attempted to answer its call. In actual practice the interval of time within which one operator would take an answered call away from another operator would be very short, unless the second operator plugged in deliberately in such a way as to deprive the first operator of an effective response to the subscriber's call. If, in the Hollenden switchboard, all of the operators simply attempted to plug in upon seeing a subscriber's lamp signal light, it would be only a short interval of time by which one would precede the other. Yet it remains a fact that, no matter how many others plugged into a multiple jack for answering the subscriber's call, this girl who held the master position would take the call away from the others. Whenever the operator who sat in the master chair did plug in to a multiple jack of a calling line, she would be the one that would get the call, and she would be the one who alone could complete the call. To that extent, the Hollenden Hotel switchboard did not provide perfect distribution. Even if the Hollenden Hotel switchboard had an advantage over the No. 1 switchboard such as I have previously described, it had also the disadvantage to which I have last alluded. The disadvantage ran in a series such that the operator who sat at the position next to the master position could take the call away from the third operator, and the operator who sat in the master position could take a call away from either or both of the others. That did not tend to perfect distribution.
There are many other patents which come on the horizon as we pass down through the years. In so far as I am aware, no switchboards have been built and put into operation in accordance with these other patents; at least the record does not show it to have been done, and the natural presumption to draw is that, in a lawsuit tried as well and so thoroughly covered as this one has been, probably they have not been made and used, or that fact would have been shown. There are many patents that have aimed at this problem that I have been discussing.
For example, the French patent No. 320,686, of January 5, 1904, attempted to simplify this matter of distribution and avoid the necessity of changing the jumper wires in order to give an operator more work or less work. In this French patent it was proposed to provide all of the multiple spring jacks with line lamp signal sockets so that the line lamp could be removed from the socket associated with a spring jack located before one operator, and inserted into the lamp socket associated with a spring jack located before some other operator. The effect of this was to decrease the number of calls which would come before one operator and increase the number coming before some other operator. This scheme provided an easy method of shifting the line lamps from position to position in such a way that any particular operator would have more line lamps or less line lamps located before her; in other words, more subscribers or fewer subscribers to take care of. The French patent succeeded in eliminating the intermediate distributing frame and jumper wires above described, but it made no effort at all in the direction of what has been called "non-interfering answering."
Coming now to the switchboard which the plaintiff, the Kellogg Company, has produced: This Kellogg switchboard includes some things not covered by the Currier patent in suit, but which adapted themselves very readily and nicely to the invention of the Currier patent. These switchboards manufactured and sold by the Kellogg Company in a substantial commercial way did include the invention of the Currier patent in suit, and, while the switchboards produced and used by the defendants do not make use of every little detail shown by Currier, I find that they are substantially the same.
The Kellogg Company's "feature" switchboard, which embodies the invention of the Currier patent in suit, has at each operator's position, or within the reach of each operator, one, but only one, spring jack for each subscriber's line, and this one spring jack is used both as a calling jack and an answering jack; and at every one of its multiple appearances throughout the switchboard each spring jack has associated with it a line lamp signal. (Sometimes in large installations, half of the lines will have spring jacks equipped with line lamp signals at half of the switchboard positions, and the other half of the lines will have spring jacks equipped with line lamp signals throughout the other half of the switchboard, etc.) These spring jacks and their associated line lamps are multipled throughout the switchboard so that every operator at every position of the switchboard has within her reach one, but only one, spring jack representing each telephone line in the exchange; and each of these spring jacks is accompanied by a line lamp. There are no separate jacks or lamps for answering. There is but one spring jack field for each operator, and, no matter how many times this field may be duplicated or repeated throughout the length of the switchboard, there is in each field one and only one spring jack representing each line, and each spring jack is associated with a line lamp signal. (I sometimes speak of these positions or separate spring jack fields as being served by one operator. It is convenient to think and speak of them in that way, although it is customary, as previously indicated, to assign three operators to each section of the switchboard as represented by one complete spring jack field.) The result is that every time a subscriber calls, his particular spring jack is lighted up within reach of each and every position throughout the length of the switchboard.
In accordance with the noninterfering answering feature of the Currier patent, the arrangement is such that the first operator to answer gets the subscriber and gets his line and completes his call, and the first operator that plugs in, not only gets the line and the subscriber, but she extinguishes her line lamp and all of the other line lamps so that there is no longer any incentive for the other operators to answer, unless they are already on their way; and, if they are on their way and plug in, the one first to plug in gets the line, answers the subscriber, and completes the call, and all the other operators are shut out. There is no master girl, no master position, no inferior position. All the operators are of equal rank and standing. The arrangement is not as in the defendants' No. 1 board, where both operators are connected with the calling subscriber. It is not as in the Hollenden Hotel switchboard, where the master girl gets the call in preference to all others. In the plaintiff's Currier board, the one first to plug in gets the line, answers the subscriber, and completes the call, and all the others are excluded. If other operators are on their way and do plug in, they will not get the line, they will not get the calling subscriber, they will not interrupt anybody, but they will all be barred. They withdraw their plugs and go to some other work.
This arrangement is such that all the pending work in the way of answering calling subscribers is at all times visible at every position of the switchboard, and all the girls get work so long as there is work to be done. The operators have been referred to as "racing" to answer calls. I am not much impressed with the idea of their racing to answer calls, but I am impressed with the proposition that all of the operators always have before them the total amount of unfinished pending work — calls to be answered — so that they can all work intelligently and understandingly at the job of answering any and all pending calls.
This arrangement may not fit into the needs of some particular exchange in some particular locality, but it is a thing worth while and valuable to the art. This arrangement which I have described will take care of a large number of lines in a large exchange. The operators can all work at the switchboard, and they can all see all of the work ahead of them. They all know whether they are accomplishing results, or whether they are getting behind. They know when they are caught up with the work to be done. They know whether they ought to work harder. The management also can tell whether they are taking care of the work to be done, and whether to lop off one or more operators for reasons of economy. The management can repeatedly reduce the number of operators at the switchboard to a point where those who remain at work will have enough work to keep them busy. No other adjustment is necessary. With a long row of girls working on such a switchboard, the management can drop off one or more or add one or more operators, and still have this uniform division of opportunity and work. This adjustment of personnel can be accomplished without any mechanical change of any kind. If it should so happen that one girl was new and inexperienced, and therefore slow in her work, whereas another was experienced and rapid, the experienced girl can work rapidly while the inexperienced girl works slowly. Operators of all sorts of capabilities can be mixed in together and each will have an opportunity to do her best, whatever that best may be. There is an advantage in having each position of the switchboard the equivalent of every other position, all of equal rank, so that any operator, no matter whether she sits at the first, second, or third, or what-not position, has the same opportunity as any and every other operator to answer and complete every incoming call.
There may be an advantage in the idea that the operators race or compete with one another in answering calls, but I do not attach much weight to that. It is an incentive in the right direction. It is claimed by the defendant that it is a disadvantage, but I do not agree with that. This switchboard does accomplish even distribution which adjusts itself to changing conditions.
Then, in addition to this matter of even and automatic distribution of incoming calls which I have discussed, there is the so-called "instantaneous disconnect." This is an arrangement in and of this switchboard such that, when a party has finished talking and hangs up his receiver, he does not have to wait any length of time at all before he can take his receiver off the switch hook and again light his line lamp signal. This matter of handling the "recalls" of a subscriber who may be making a series of calls at one sitting has been a good deal of a problem in this art. In other cases, I have heard much about various methods devised to take care of such recalls. In the old days the telephone subscriber had to wiggle his switch hook up and down — "jiggle" it — in order to flash the supervisory lamp on the operator's key shelf to attract her attention. The inventions involved devices such that, when a subscriber hung up his receiver and then took it down, the supervisory lamp would flash automatically. The arrangement of this Currier patent, No. 1,428,762, is such that, when the subscriber hangs up his receiver, the connected cord circuit is split in such a way that electrically it is just the same as if the answering plug of the cord circuit had been pulled out of the spring jack terminals of the telephone line. In other words, the subscriber's line is all ready for a "recall" by the lighting of the line lamp signals associated with the spring jacks, wherever they appear in the switchboard, just as soon as the subscriber hangs up his receiver at the termination of the previous conversation; so that instantly and immediately, upon the subsequent removal of his receiver from the hook, the subscriber's line lamps are lighted at the switchboard just as quickly and in the same way as they would light if he were making an original call. The only difference is that an answering plug of the cord circuit of the operator who had previously been the winner, and who had completed the previous call, may still be in a spring jack of the recalling line. If this operator, who has handled the previous connection, sees one of her plugs in a spring jack whose associated line lamp is lighted, she will have an incentive to take care of this recall. She has the means of knowing that the glowing lamp of a spring jack in which one of her plugs is inserted represents a recall.
The way the operator who handled the previous connection answers a recall is simply to pull her calling plug out of the spring jack of the line which was last called, and this act connects the operator's telephone so that she can answer the recalling subscriber without having to insert or reinsert an answering plug. The other operators also have an opportunity to answer this recall. They also see the recalling subscriber's line lamps light. This recall signal appears before them just the same as though it represented an original call. All of the subscribers' spring jacks bear a lighted lamp and any of the other operators can insert an answering cord plug. If any one of them does so before the operator that had answered the previous call has pulled down her calling plug, then the new operator would get this recall and complete the connection, to the exclusion, not only of other new operators, but also of the operator whose answering plug still remained in the spring jack of the recalling subscriber.
It is a well-arranged switchboard.
Many of these elements, individually and alone, can be found scattered throughout the old art in either what was done in practice or shown in patents, but not all of them can be found in any one place. Here they are gathered together. All of the claims of the patent in suit are combination claims. During the trial and arguments, some of them were referred to particularly as "combination" claims because they combined the two big principles that I have discussed. But in a way all of the claims in suit describe combinations of things which were separately old in the art. I have seen fit to dub the various relays and appliances as "the brick and mortar" of the art, and in this case "the brick and mortar" was found in the prior art. But when the brick and mortar were put together in this Currier patent (and I am not now talking about any particular claim), there was produced a switchboard which was new in the art, and it produced for the art a new, useful, and valuable result, particularly in the way of distributing work between operators. Here you can have your row of operators, your row of positions, and without mechanical change you can increase the volume of the work either as a whole because it is seasonal, or you can increase it in spots when the business of one subscriber greatly increases while the business of his next door neighbor may fall off and become very little. It is a switchboard which automatically adjusts itself from day to day and from hour to hour; and all that it is necessary to do is to add additional operators to the switchboard or to remove some operators from the switchboard and the whole job of distribution or redistribution is complete.
This principle of distribution has been involved in some of the prior art switchboards, but here it is coupled up with this matter of noninterfering answering which is a useful thing to have. It fills a useful place and serves a useful purpose to have this kind of a selfadjusting distributing arrangement, and a desirable part of it, at least for certain uses, is this noninterfering feature. It is valuable to have this noninterfering feature wherein the positions are equal as distinguished from the arrangement in which there is a master position as in the Hollenden switchboard.
Of course, multipling the spring jacks was old; but Currier multiples his jacks and also multiples his lights and he has this complete noninterference. With the noninterfering feature the arrangement is satisfactory in every way. In some of his claims Currier adds the instantaneous disconnect feature so that, for the purposes of handling recalls, the switchboard is just the same as it is for handling original calls, with no need on the part of the operators of watching for supervisory lights or in any other place for recall signals. All of the calling and recalling signals are up in the face of the switchboard.
There may be instances where a telephone manager would prefer for some reason to use a supervisory light for recall signals, but it is valuable in this art to have the Currier arrangement wherein all of the calling and recalling signals are located and combined in one upright of the switchboard.
We have so many of these separate features which can be found in the prior art that it would require a long discussion to take them up one at a time and to attempt to cover them in detail. For instance, there is the McQuarrie and Loveridge patent, No. 1,217,472. It is entitled to much thought here. When we liken the automatic finder switch to the operator and the potential of the two points that come together and the way they operate there, we find very many things in McQuarrie and Loveridge that are very similar to what is in Currier. On the other hand, the means of distribution in the two systems are very far apart. If you look at the matter from the standpoint of the operators, they surely are very far apart, because in the McQuarrie and Loveridge system the girl's work is parceled out to her. She is handed by the machine the things to be done. She does not know how much work is pending. McQuarrie and Loveridge were inventing in the mechanical art and Currier in the art of switchboards operated by human hands. As I have said in previous decisions, we had in this great art as it developed the automatic and we had the manual systems. We have had some automatic exchanges operated almost entirely without the aid of human operators. You probably can never make the whole world automatic or the United States wholly automatic. There are going to continue to be a large number of these switchboards operated by girls. The situation does not mean that there is no room for invention in these switchboards that are operated by the girls. The telephone art is going to be taking things from one system to the other, and usually the more difficult step is from the manual over to the automatic.
Yet that is not always true. In this matter of distributing calls between operators, which is the thing that Currier's invention had to do with, we have the human element — the fact that some girls can do more work than others, for example. These facts inject into the matter of distribution an element that is not present with the automatic or mechanical systems. Currier provided a system that takes care of that human element along with the many other elements.
In McQuarrie and Loveridge, the machines, like all mechanical things, move necessarily in fixed grooves. If no subscriber were talking, the McQuarrie and Loveridge board would be entirely idle. If that were the situation, a given subscriber upon projecting a call into this entirely idle place would always get through one particular machine to one particular operator; he would always get the same girl. The same machine would always pick up and handle his call in preference to other electrical switches; the one that was second would always be second, and the one that was third would always be third, because it is mechanical. I point this out, not because it is a disadvantage, but to show that the McQuarrie and Loveridge patent is for a different scheme and a different arrangement than Currier's arrangement for these operators — these girls — where, regardless of their ability, regardless of their number, regardless of the load, the first to plug in is the one who seizes the line and answers the call and completes the call, and all others are barred out. They all have before them all of the work, and they see it all of the time, and the human elements stimulate them to greater activity when necessary. There are people watching them work, the supervisor can see all of the conditions right there before her all of the time.
As a distributing means, McQuarrie and Loveridge was entirely different from Currier. It undoubtedly had a place in the art. It ought to be taken into account, but it meets a different need.
In many respects what has been said of the McQuarrie and Loveridge patent may be said also of the prior patents to McBerty, No. 780,679, and to Fay and Molina, No. 1,073,748. They are all automatic switch systems and none is more pertinent nor indeed so much urged as McQuarrie and Loveridge.
The prior patent to Dyson, No. 1,208,718, does not require detailed discussion. Dyson did not have call distribution in mind at all. He was concerned with an automatic busy test located at the calling end of a cord circuit. An automatic busy test and ringing system such as that of the Dyson patent is designed and intended to permit an operator to plug directly into the jack of a called subscriber's line without waiting to make the usual busy test (by applying the tip of the plug to the sleeve of the jack) to determine whether the called line is busy or idle. If the called line proves to be idle, the called subscriber's bell is automatically rung to summon him to the telephone; but, if the called line is busy, the the ringing current will not be applied, but a busy signal will be automatically transmitted back to the calling subscriber to notify him of the busy condition of the called line. Regardless of the mechanism by which Dyson accomplished his results, it is clear that his patent did not relate to the distribution of incoming calls or to the noninterference of operators in answering them.
Another patent of the prior art is that to Andre, No. 816,894. Andre did not have the distribution of calls in mind at all. Currier was not the first to have all of the accumulated work within the view of all of the operators, but Currier included with it the idea of letting the first operator to insert an answering plug in the jack of a calling line complete that call, and complete it without interference by any other operator. Andre did not have any such purpose. His noninterfering arrangement was to prevent eavesdropping. It was the sort of thing that Currier might have taken to use in his new combination; but we must give Currier credit, not only for taking it and using it in a way to do the trick, but for thinking of taking it and doing the trick. Andre was not a distributing device at all, and it was not intended as a noninterfering device. Andre had in mind to keep other operators from listening in to what was going on between two subscribers who were already talking. Andre arranged it so that another girl could not plug in to the jack of a busy line and listen. He was not thinking about doing what Currier did. He did not couple up his eavesdropping preventative with any means which could be used in the distribution of incoming calls.
Whether we look at the old structures separately or collectively, it is plain that Currier has in this patent produced a new switchboard which accomplishes certain new and useful results. Currier has made an invention which is worthy of a patent.
We turn now to the claims in suit to see whether or not Currier has covered in an effective way this thing which was valuable to the art. The claims of the patent do not cover everything on the switchboard which I have described, but that switchboard uses the claims as a very necessary part. All of the things that I have discussed fit into the scheme of Currier.
I wish first to consider briefly claims 2 and 12. These claims cover multiple jacks, multiple lights, and noninterfering answering — which represent the really big thing that Currier did. The two claims use different words, but I do not see very much difference in them. I think it is noninterfering answering, coupled with the multiple jacks and multiple lights that entitled Currier to a patent over the defendants' No. 1 switchboard, the Hollenden Hotel switchboard, and all the other structures of the prior art. I sustain claims 2 and 12 as valid.
Claim 5 leaves out the multiple lights, and I cannot see that it adds anything to the protection afforded by claims 2 and 12. It includes multiple jacks and noninterfering answering, but does not with certainty include signals to tell the operators where to answer. It is my view that without the lights claim 5 presents only an incomplete and fragmentary combination. Without the multiple lights you do not have the whole of the worthwhile thing in Currier. Unless claim 5 be construed as including the multiple lights, the combination lacks the utility which I have ascribed to Currier's invention, and I do not think Currier's claim 5 is entitled to a construction which would include the multiple lights. It has the noninterfering answering, but lacks one of the elements of the combination in which Currier's noninterfering answering functions, and to construe claim 5 as including the multiple lights would tend to make claim 5 a duplicate in substance of claims 2 and 12. So I think the fair thing to do is to disallow claim 5 because it describes an incomplete combination.
Farrington v. Haywood, 35 F.2d 628 at page 630 (C.C.A. 6); Daniel Green Felt Shoe Co. v. Dolgeville Felt Shoe Co., 210 F. 164 at page 165 (C.C.A. 2); Rodman Chemical Co. v. Deeds Commercial Laboratories, 261 F. 189 at page 192 (C.C.A. 7).
I next take up claims 30 and 33, which relate to the "instantaneous disconnect."
The "instantaneous disconnect" is an arrangement for, in effect, disconnecting the operator's cord from the subscriber's line the moment he hangs up his receiver. This puts his line and its signaling equipment at once in normal condition so that he may either make another call or receive a call from someone else.
If he makes another call while the plug is still in the jack, it may be taken by some other operator, or it may be taken by the same operator. If the same operator takes the call, she must reconnect her telephone set to the cord, and, to do this with an "automatic-listening" cord such as shown in the patent, she must withdraw the calling plug. With a key-listening cord she would operate her listening key. In either case the relay which has opened the cord circuit to accomplish the "instantaneous disconnect" must at the same time be released so that the cord conductors are again continuous. The claims to the "instantaneous disconnect" here relied upon are claims 30 and 33, and these include the idea of the reseizure of the call by the same operator.
The "instantaneous disconnect" idea itself is admittedly very old in the art. A patent to Sabin, issued in 1882, discloses this idea, and the plaintiff itself owned a patent to Dean, No. 854,279, covering such an arrangement, which expired in 1924. The Clausen patent, No. 1,088,879, filed in 1911, shows an "instantaneous disconnect" provided for both ends of the cord circuit, instead of for but one end as in the Currier patent, and it is so arranged that the withdrawal of the calling plug reunites the cord conductors to enable the same operator to reseize the line when another call is made by the same subscriber before the plug is withdrawn from his jack. Clausen shows a key-listening cord, so that the operator's telephone set is not reconnected by this withdrawal of the calling plug. Automatic listening was, however, admittedly very old before Currier, and, if the conventional automatic listening circuit were applied to the Clausen cord, the withdrawal of the calling plug would, of course, also reconnect the operator to the cord without the use of the listening key. The disconnect relays in the Clausen patent are under the joint control of the calling and called subscribers, while in Currier the calling subscriber alone controls this relay. Mr. Eaton's testimony shows that it has been controversial among telephone experts whether the better plan is to give the control to a single subscriber or to make it a joint control by both.
The Aitken patent, No. 1,221,404, filed in July, 1913, is another example of an "instantaneous disconnect" and in this case the disconnect relay is under the control of the calling subscriber alone, as in the Currier patent. In the Aitken patent, if the same operator desires to take another call by the same subscriber after the disconnect relay has operated, she does it by throwing a listening key, which connects her telephone to the cord and reunites the cord conductors. This corresponds with the Currier arrangement, except that the manual act for accomplishing these two results in the Currier patent is the withdrawal of the calling plug instead of the throwing of a listening key, but this difference is merely incidental to the fact that the Currier circuit uses the conventional automatic listening equipment.
Briefly, I hold that these claims are void on Clausen patent, No. 1,088,879, and Aitken patent, No. 1,221,404. In this instantaneous disconnect feature alone I cannot see enough to support a valid patent. In the claims as worded by Currier, there is something a little bit different from Clausen and a little bit different from Aitken, but, when I take Clausen and Aitken together, they tell the whole story.
This leaves claims 13, 35, and 39 for consideration, and they involve rather a peculiar situation. I feel, as I have said, very strongly that Currier's total accomplishment was valuable. The results are better because the additional features of claims 13, 35, and 39 are combined with those of claims 2 and 12. The additional feature of having the recall signal come in on a plurality of line lamps is not found in the prior art. These claims 13, 35, and 39 represent a creation more valuable than that of claims 2 and 12 which I have just held to be valid. The invention of claims 13, 35, and 39 should, perhaps, be regarded as a refinement of the invention of claims 2 and 12.
Having reached the conclusion that "instantaneous disconnect" standing alone did not involve enough ingenuity over the prior art to give validity to claims 30 and 33, it may seem inconsistent to sustain claims 13, 35, and 39, which we have been designating as the "combination claims." These claims might perhaps be regarded as narrow claims to the same feature. But in claims 13, 35, and 39 "instantaneous disconnect" becomes a part of a larger combination. It should be said in this connection that this feature fits very beautifully into the whole arrangement. So, despite the prior art which makes me think that standing alone claims 30 and 33 are void, it is my conclusion that the "instantaneous disconnect" fits neatly enough into the whole arrangement to constitute a part of a proper combination. I hold claims 13, 35, and 39 valid as such combination claims. Claims 2 and 12, which do not include the instantaneous disconnect, do not provide as desirable a switchboard as the complete combinations of claims 13, 35, and 39. The instantaneous disconnect fits in there advantageously, and the whole combination is new. Currier is the first man to have brought the combination together, he is the first to have made this kind of a switchboard. Claims 13, 35, and 39 taught the art how to build a better board than is shown in claims 2 and 12. Whether we call it a narrowing of claims 2 and 12, or whether we call it a distinct combination, is not a thing to worry about. In any event, I think that all of the elements of these claims are so related that they fit in together, that they work together, and that they co-operate to produce a common, useful, and new result. Therefore, the elements constitute a proper combination, and I hold claims 13, 35, and 39 to be valid.
I think highly of this patent as a thing helpful in its field, and that field a varying one. It is not an invention which would greatly affect the number of telephones in use; the business of the Michigan Bell Telephone Company would doubtless be about as large without this invention as with it; but Currier produced a helpful thing which went into actual and extensive use. It tends to better telephone service at a lesser cost. I find in it an abundance of novelty and merit to sustain the claim of patentability.
Coming now to the defendants' claim that they have a license under claims 2 and 12. My real problem is to take the license agreement which was entered into on June 2, 1914, and to determine from all four corners of that instrument what the parties meant and had in mind; and not to do it by any technical or arbitrary rule.
There are three reasons that lead me to the conclusion that the license from the Kellogg Company to the Western Electric Company for patents covering "automatic busy tests" does not apply to the Currier patent in suit.
I have a great deal of confidence in Mr. Bailey, one of the defendants' engineers, and I know that in his testimony he called the patent in suit one for an "automatic busy test"; and I am aware that Mr. Camp, when acting as the solicitor in prosecuting the application for this Currier patent, used the term "busy test" in connection with his discussion of a prior art patent to McQuarrie, No. 502,466, which discloses shutters mechanically operated in such a way as to bar the insertion of a second plug into any one of a set of multiple connected trunk jacks when some other plug has already been inserted into some other jack of the set, and that Mr. Camp in referring to what he called the "busy test" of this McQuarrie trunk circuit went on to describe this term by saying "that is for preventing a second link circuit from operatively connecting to a calling line." I cannot read Mr. Camp's communication to the Patent Office in any way which does not carry the implication that he applied the term "busy test" to something comparable to calling lines, rather than to called lines. But it is only by implication that this construction of his statement can be applied, because the mechanical shutters in the McQuarrie patent are associated with a trunk line extending between two telephone exchanges, and not with the spring jacks of a subscriber's line at all, — either a calling or a called subscriber's line.
What we think of when we speak of "an automatic busy test" is at the other end of the cord circuit, i.e., at the calling end or calling plug. That is what the layman thinks about. He does not say that his own line is busy. He says that the man he is calling is busy. During the trial of this case there has been much discussion about busy operators, but in such discussions we have referred to quite a different thing from what we have in mind when we say that a subscriber's line is busy. The natural thing is to think of a busy line as the called line and of the busy test as associated with the calling end of the cord circuit. I will not say that you could not possibly apply that term to the answering end of a cord circuit, but it is not the natural thing, the easy thing, the simple thing, and nine hundred and ninety-nine out of a thousand automatic busy tests at least would be made with the calling end of the cord circuit and with the calling plug and the spring jack of the called line.
In the second place, the "automatic busy test" is coupled up in section 1 of the license agreement with the calling or ringing end of the cord circuit, in the phrase "automatic busy test or ringing system." The automatic busy test and the application of ringing current to the called subscriber's line are hitched together, and the "ringing system" has to do with the party that one wishes to call and with ringing the bell of the called subscriber's telephone. In connection with the ringing, I think of the calling end of the cord circuit and of ringing the called subscriber. Who is it that the operator rings? She does not ring the man that is calling. He gets angry if by mistake she does that. The operator rings the subscriber with whom the calling subscriber wishes to talk. I think that, when the contract refers to "automatic busy test or ringing systems," it means to refer to the calling end of the cord circuit and to the subscriber to be called.
The third thing is that on the whole the conduct of the parties supports that interpretation of the contract. I do not fail to recall the letter that was written by one of the defendants' patent attorneys, but that sounds like a letter which one writes when he is thinking of lawsuits and has hopes, but is not certain. These parties entered into this license agreement, and after that they had this contest in the Patent Office over this very Currier patent in suit. During all the long period of that contest in the Patent Office over this particular patent in suit, there was no claim of a license. The letter was an interdepartmental letter of the Western Electric Company, a subsidiary of the defendant, but no claim of a license was made to the plaintiff company until after this lawsuit was brought.
Now these three considerations, all fitting in together, lead me to the conclusion that this license agreement does not cover or include this particular Currier patent in suit.
It is very clear that claims 2, 12, 13, 35, and 39 are infringed and I so hold.
Patent No. 1,438,170.
The second patent in suit, No. 1,438,170, relates, as I have said, to the so-called "flashing recall," which is an arrangement by which a subscriber can give to an operator an automatically flashing lamp signal to indicate that he desires to communicate with her, either in connection with a call she is handling or for the purpose of making another call, before she has pulled her plug out of the line jack following a call which has been finished.
In telephone systems long in common use, the operator's cord circuit is provided with two "supervisory" lamps, one under the control of the calling subscriber and one under the control of the called subscriber so long as the plugs are in their respective jacks. While the conversation is going on, that is, while both subscribers have their telephones off the hooks, these lights are dark. When either subscriber hangs up, his supervisory lamp is lighted. This indicates to the operator that the plugs should be withdrawn from the jacks. With this arrangement, if either subscriber desired to recall the operator while the plug was still in, he did it by moving the switch hook up and down repeatedly to cause an intermittent flashing on and off of his supervisory lamp; and this has for years been the practice in making recalls.
The subject-matter of this patent is simply a means for automatically causing an intermittent flashing of the lamp for this same purpose. It is started by a single depression and release of the switch hook. The automatic flashing is accomplished by so arranging matters that the first depression of the switch hook (the plug being still in the jack), not only lights the supervisory lamp as usual, but also energizes a relay which locks itself up and makes ready a circuit which will transfer the supervisory lamp from a steady to an interrupted circuit as soon as the switch hook is again released.
In the old art the steadily burning light indicated to the operator that she should remove the plug, and the flashing light resulting from the up and down movement of the switch hook indicated to her that the subscriber wished to speak to her again, but confusion sometimes resulted from the fact that the subscribers would occasionally remove their receivers without vibrating the switch hook and merely wait to be answered as with an original call; or they would vibrate the hook several times and then wait for attention. If the operator had not been looking at the light when it was thus flashed by the operation of the hook, she would not know that the subscriber had hung up his receiver at all, and might assume that he was still talking, since the lamp would again be dark; or the subscribers sometimes flashed their hooks too fast, in which case the lamp would not give any signal to the operator. These are examples of the confusion that was avoided by causing the lamp to flash automatically after a single operation of the switch hook.
There is, in my opinion, clearly room enough for an invention over the old practice just described in the idea of making the lamp thus flash automatically without any thought on the part of the subscriber, so that either by hanging up and immediately taking down the receiver, or by giving a quick flick with his finger to the switch hook, he gives the operator a continuously flashing signal to tell her that he wishes to talk again. It would not in any sense be a revolutionary invention. Doubtless many subscribers would still vibrate their switch hooks to make recalls, not knowing of the automatic equipment; but it was a worthwhile thing.
However, I find that the patent to Smythe, No. 758,116, which antedated Currier by eleven years, discloses a flashing recall circuit which fully anticipates Currier's invention, so far as anything here involved is concerned. The drawing of the Smythe patent shows only the calling end of the cord, and the flashing recall equipment is illustrated as applied to that end of a cord, but I do not find anything in the specification of the patent which indicates that he considered his invention to be limited at all to use only upon that end, and the testimony shows that it is in fact usable upon either end, or both. Recalls are more often made by the calling subscriber than by the called subscriber, and I do not understand that Smythe, by showing his recalling equipment upon the end of the cord connected to the called subscriber, intended to suggest anything to the contrary. I think it is clear that he showed the equipment on this end of the cord because that end presents a more difficult problem. There, an additional type of signal is necessary if all confusion is to be avoided, and by applying his invention where the problem was most complex he covered the entire field. At the answering end of the cord, the subscriber's telephone is already off the switch hook when the operator plugs in to answer, so that there are only three types of signals required, namely, one to show that the subscriber is talking or ready to talk, one to show that he has finished and that the plug should be withdrawn, and one to show that he desires to recall. At the calling end of the cord, which is connected to the called subscriber's line, four types of signals are required, one to show that the subscriber's telephone is still on the hook and that his bell should be rung and three more which correspond to the three just enumerated for the answering end of the cord.
Under the old practice there were two chances for confusion in the called subscriber's signals. When the lamp was burning, the operator would not know (unless she remembered) whether the subscriber had finished the conversation and hung up his receiver or had not yet answered the bell and should be rung again. When the lamp was dark she would not know whether the subscriber was still in conversation or had hung up and again taken down his receiver to make a recall. This last is the same possible confusion which might arise on the answering end of the cord. Smythe provided four distinct signals which were enough to avoid confusion under all circumstances, even at the calling end of the cord. At the answering end, three are enough. In order to get his four signals, he used the steadily burning lamp as the ringing signal, that is, to indicate that the subscriber had not yet answered; the dark lamp for the talking condition; a slowly flashing lamp to indicate the recalling condition; and a rapidly flashing lamp for the disconnect signal. The evidence shows that the very equipment which Smythe provides may be used upon both ends of a cord circuit without change and serve to provide all of the signals necessary to avoid confusion on either end of the cord. In such case the steadily burning lamp, which is the ringing signal on the calling end of the cord, would not come into play at all on the answering end, since the calling subscriber's telephone is already off of the hook before the operator plugs into his line jack.
If one decides, as Currier did, to use the distinctive signals upon the answering end of the cord only, the separate ringing signal used by Smythe becomes unnecessary, and the natural thing to do would be to use the steadily burning light (now unnecessary as a ringing signal) for the disconnect signal just as had been done for years under the old practice. This is what Currier did. I do not find any invention in substituting the old steadily burning light as a disconnect signal for the rapidly flashing light of Smythe, where the signals are to be applied only to the answering end of the cord. Smythe provides a flashing recall signal, just as Currier does, and this is the essence of what Currier claims to have invented. The real ingenuity, such as there is, lies very largely in the conception of the thing to be done and the general conception of how to do it, rather than in the working out of the details. Long before Currier, the telephone development — the brick and mortar of the art, as I have termed it here — was such that engineers were well acquainted with the ways in which relays and their circuits could be used for perfecting the details of such an arrangement.
When Currier's application was filed in the Patent Office there followed an interference with an application filed by Clausen, one of the defendants' engineers, which resulted in the study of the prior art, including Smythe of course, to determine the patentability of the interfering subject-matter. It is not necessary for the court, nor for counsel, to harmonize everything that has been said on either side during the long history of this patent either by counsel or by the Patent Office, and it would be difficult to do, because the viewpoints of different people have been different, and the viewpoints of the same people have perhaps been different at different times. I do not regard the proceedings in the Patent Office as creating an estoppel against either party in the present proceeding nor as binding upon this court. I will say, however, that the discussion of the Law Examiner on that first interference between Currier and Clausen is along a line with which I find myself in agreement. It seems to me that the thing which the Patent Office there held to be unpatentable and took out of the applications of Currier and Clausen constituted the main trunk of the tree we are considering and was, in substance, what we find in the Smythe patent. What was left were really nothing but the shoots which have grown up around it, and I find that these are, all of them, mechanical things and old things.
One of the things which Currier claims to have done that Smythe did not do was to arrange for the operator to reseize the call by pulling down the calling plug. There is nothing in any of the claims here relied upon (33, 34, and 46) which specifies this method of reseizing the call. Claims 33 and 34 indicate that the flashing mechanism for the recall light is conditioned for operation when the plug is inserted in the calling line. This concerns the so-called dual control which has been much discussed in this case. All there is to this, in my opinion, is that Currier took Smythe, and, in place of the listening key which in Smythe conditioned the flashing mechanism when the operator "listened out," used the conventional mechanism for automatically listening the operator in and out, which is responsive to the putting in of the plugs at the opposite ends of the cord. From this substitution it naturally resulted that, when the calling plug was inserted, the operator was listened out and the flashing mechanism conditioned for operation at the same time. The important thing in any recall mechanism is to condition it for operation by the calling subscriber the moment the operator is disconnected from the circuit, because up to that time there is no need of any recalling mechanism, while at any moment thereafter a recall may be necessary. This is just what Smythe did, and it is just what Currier did. It is also just what the defendants do in their alleged infringing boards, except that, in their No. 11 board, they do not condition the mechanism for operation when the plug is inserted in the calling line, as specified in these claims, because they do not then disconnect the operator from the cord circuit — they do so a moment later when the ringing key is depressed. Consequently, with the defendants' No. 11 arrangement, the recall mechanism is conditioned at the time the ringing key is depressed. I am convinced, therefore, that, if the old automatic listening in and listening out of the operator be substituted for the key-listening, in and out, shown by Smythe — a thing easily done by any telephone engineer — we get all that is involved in what the plaintiff has here referred to as reciprocal or joint control; and that, therefore, there is nothing of patentable substance in it.
What I have said about the merits of the claims makes is unnecessary to talk about the infringement, but the situation illustrates how little there is left to hang a valid claim upon after the substance of the Smythe invention is carved out of the Currier disclosure. If the court should try to hold these claims valid over Smythe, they would become so narrow that it would be easy for any one to escape them. For example, I agree with the defendants that in claim 33 the word "when" in the clause "a relay conditioned for operation when the link circuit is connected with the called line" so limits the claim that it would not be infringed by the defendants' system, in which the conditioning is done after that operation by the depression of the ringing key. Another example is the expression in the claims relating to the series arrangement of the interrupter and the signal lamp. Smythe shows a shunt arrangement of these elements, and the defendants use a shunt arrangement. There is evidence that these are equivalents and I believe them to be, but if I were to hold that Currier distinguished from Smythe in this respect, I should also have to hold that the defendants do not infringe. The things by which the claims distinguish from Smythe are merely mechanical and should not be treated as the product of genius.
This opinion shall be construed as a statement of the findings of fact and conclusions of law, as required by Equity Rule 70½ (28 USCA § 723).
A decree will be entered for the plaintiff against the defendants, holding claims 2, 12, 13, 35, and 39 of patent No. 1,428,762 valid and infringed, and reference will be made to William S. Sayres, as master, to compute the profits and damages. The master will be instructed also to report a reasonable royalty, so that it may be adopted as the measure of damages if such adoption should seem to the court to be appropriate. Claim 5 is disallowed, and, as to claims 30 and 33 of patent No. 1,428,762, and claims 33, 34, and 46 of patent No. 1,438,170, they are held to be invalid, and the decree will be for the defendants.
The decree will be without costs to either party.