In 1986 Granger Associates (now owned by the defendant, DSC Communications Corporation ("DSC"), a Texas-based corporation) designed specifications for a telecommunications switching device known as the ECS-3 System. The ECS-3 is a switching device that was intended to be used to route telephone calls in long-distance communications. A crucial component of the ECS-3 is the voltage controlled crystal oscillator, also known as the "VCXO".

The VCXO is a particular type of oscillator, which is a device designed to produce electrical or mechanical vibrations. A VCXO has two defining characteristics. First, it is a crystal oscillator, which means that the principal factor that determines the frequency of the vibrations is the resonance of a piezoelectric crystal. Second, it is voltage controlled, which means that the frequency can be adjusted by varying the voltage of an electrical current that is applied to a metal overlay (or "spot") on the surface of the crystal.

The VCXO itself is a relatively common component. But the ECS-3 required a VCXO with a higher frequency than had been previously available. Each ECS-3 required over 1,000 VCXOs, which served as timing devices. The VCXOs essentially allowed the various components of the ECS-3 to work together, as well as to coordinate communications between the ECS-3 and the rest of the telecommunications system.

Granger attempted to find a company that could design and manufacture the VCXOs it needed for the ECS-3. Originally a California corporation called Monitor Products agreed to supply the VCXOs, but, while it developed a working prototype, it was unable to achieve production quantities within the agreed-to price range, and therefore withdrew from the project.

After Monitor withdrew, Granger asked one of its subsidiaries, Advanced Filter Devices ("AFD") to begin work on a VCXO for use in the ECS-3. But Robert Crawford, the President of AFD, thought the project impracticable in light of the cost and space restrictions and therefore declined to work on the project. Granger then went out looking for other suppliers, and in February 1987 Sokol (a relatively small Wisconsin concern) submitted a proposal to supply the VCXOs for the ECS-3. In March 1987 Granger issued a purchase order to Sokol for sample VCXOs, and in May 1987 Sokol and Granger entered into a confidential information agreement. The agreement, in substance, prohibited Granger from using Sokol's confidential information for any purpose other than that for which it was received.

In the meantime, despite Crawford's protestations, Granger ordered AFD to try to develop a suitable VCXO. By February 1988, Crawford finalized a VCXO design. But while Crawford and AFD were working on their own VCXO, Sokol was supplying Granger with these devices. DSC contends that these VCXOs were failing at an unacceptable rate and that Granger engineers examined them in order to help Sokol produce a better product. To that end, one of Granger's design engineers, Robert Lee (who was working with Crawford to develop AFD's VCXO), "reverse engineered" a Sokol VCXO and made schematic drawings of the Sokol VCXO circuit. In November 1988 Granger told Sokol to cease delivery of the VCXOs. Thereafter, Granger used VCXOs produced by AFD, and in May 1988 AT & T also began shipping VCXOs to Granger.

Sokol sued DSC, claiming that its trade secret had been misappropriated in violation of Wisconsin's version of the Uniform Trade Secrets Act (the parties agree that Wisconsin's substantive law governs this diversity action). Wis.Stat. Sec. 134.90. While there was no direct evidence that anyone at DSC used Sokol's confidential information in the making of its own VCXO, the jury apparently inferred from the fact that DSC had access to Sokol's confidential information and from the similarity between the two devices that DSC misappropriated Sokol's trade secret and that the AFD VCXO was derived from that trade secret. The jury therefore awarded Sokol $2,492,000 in damages. DSC appeals, asserting primarily that the suit is barred by the statute of limitations and that the jury was instructed erroneously. Sokol cross-appeals, claiming that the trial court erred in refusing to submit to the jury whether DSC's conduct was willful or malicious.

I. Statute of limitations.

Under Wisconsin law, an action claiming misappropriation of a trade secret "shall be commenced within 3 years after the misappropriation of a trade secret is discovered or should have been discovered by the exercise of reasonable diligence." Wis.Stat. Sec. 893.51(2). Sokol brought this action on November 15, 1991; so the claim is within the statutory period only if the misappropriation occurred (or should reasonably have been discovered) after November 15, 1988. DSC contends that, if there was a misappropriation, it took place in July 1988, and therefore that Sokol's claim is barred by the statute of limitations.

This analysis hinges on the definition of a "misappropriation." Under Wisconsin law, a misappropriation is defined as either (1) acquiring a trade secret by improper means, Wis.Stat. Sec. 134.90(2)(a), or (2) "disclosing or using ... a trade secret of another," id. at Sec. 134.90(2)(b). Granger acquired Sokol's trade secret by lawful means. Ascertaining the date of the misappropriation therefore requires us to determine the date on which the trade secret was "disclosed or used" (in this case "used").

The questions therefore are (1) what is the date of the misuse and (2) when did Sokol know (or should it reasonably have known) of it. Pinpointing the date of "misuse" is difficult because there was no direct evidence that DSC ever used Sokol's confidential information in developing its own VCXO. Instead, the jury inferred the misuse from the fact that DSC had access to this information and from the similarity between Sokol's and DSC's products. But the difficulty in defining the date of misuse turns out not to matter, because regardless of when DSC began to misuse Sokol's trade secret, the statute of limitations does not begin to run until Sokol knew or should have learned of it. Because it is not alleged that Sokol failed to exercise reasonable diligence in determining whether DSC was misappropriating its trade secret, the question ultimately becomes what did Sokol know and when did it know it.

In this connection, DSC argues that Sokol's cause of action accrued no later than July 1988 when Sokol allegedly believed that AFD was using Sokol's trade secret in making its own VCXOs. Sokol, however, insists that it did not know that DSC was misusing its trade secret until DSC sold an ECS-3 system (containing the AFD VCXO that resembled its own) to a third party.

While it is clear that Sokol was concerned all along that DSC might be misappropriating its trade secret, the district court found that the statute should not be construed to require Sokol to bring suit based only on this abstract concern. On this point, the district court relied on the analysis in Intermedics, Inc. v. Ventritex, Inc., 775 F. Supp. 1258, 1266 (N.D.Cal.1991). There, the court observed that it "cannot apply statute of limitations law in a way that pressures litigants to file suits based merely on suspicions and fears. At least in the commercial setting at issue here, suspicion and fear are not sufficient predicates for launching a lawsuit, and to file an action with no other basis would offend norms articulated in rules like Federal Rule of Civil Procedure 11."

DSC responds by suggesting that Sokol had a stronger suspicion that DSC was misappropriating its trade secret than the plaintiff had in Intermedics. While this is perhaps true, we do not find the difference to be decisive. In July 1988, AFD sent Sokol twelve samples of its VCXO boards. While Sokol knew at that time that the AFD board was similar to its own, it did not yet know the use to which these boards were going to be put. Seeing these boards surely led Sokol to be concerned that its trade secrets were going to be misappropriated. But, while these concerns and suspicions were perhaps stronger than those in Intermedics (where the plaintiff's initial concerns that the defendant had misappropriated its trade secret were assuaged by the report of an independent auditor), these apprehensions were still just concerns and suspicions rather than knowledge of misuse. They therefore do not start the clock of the statute of limitations. Sokol did not know that DSC was misusing (and therefore misappropriating) its trade secret until DSC sold an ECS-3 system (containing a VCXO that resembled Sokol's) to a third party. That being the case, the filing of the complaint, on November 15, 1991, was within the three-year statutory limit.

II. Laches.

DSC next argues that even if the claim was brought within the statutory limit, it is nonetheless barred by the equitable doctrine of laches. In this context, the laches doctrine essentially requires that the plaintiff mitigate damages. DSC alleges that Sokol, by sitting on its hands even though it allegedly knew that DSC was misappropriating its trade secret, in effect ran up the damages caused by the misuse. But this question essentially collapses into the statute of limitations analysis. Just as with the statute of limitations, the question is when did Sokol know that it had a cause of action? DSC contends that Sokol knew all of the relevant facts in July 1988, and that it should not be allowed to recover damages that it suffered by virtue of its delay in bringing suit. But the district court rejected the statute of limitations defense because Sokol was not required to bring suit when it merely suspected that DSC was misappropriating its trade secret. Because we agree with the district court's resolution of that issue, DSC's laches defense must also fail.

III. Jury instruction.

DSC claims that the trial court erred in refusing to give its proffered jury instruction on the question whether it misappropriated Sokol's trade secret. According to Wisconsin case law, a defendant is liable for misappropriation of a trade secret if "substantially the same device used by the defendant is derived from the Plaintiff's secret in breach of a relationship of trust and confidence." M. Bryce & Associates, Inc. v. Gladstone, 107 Wis. 2d 241, 319 N.W.2d 907 (App.1982). The trial court offered exactly this instruction, which was expressly approved in M. Bryce.

But in a subsequent diversity case applying Wisconsin law, this court, citing to M. Bryce, offered a slightly different formulation. According to our decision in American Can Co. v. Mansukhani, 742 F.2d 314, 328-29 (7th Cir.1984), a defendant is liable for misappropriating a trade secret if its "product or process is substantially derived from the [plaintiff's] trade secret." DSC proffered a jury instruction along these lines, which the trial court declined to give.

The parties therefore engage in a syntactical debate, centering on the difference between saying "substantially the same device is derived" (as the trial court and Wisconsin decision say) and "the device is substantially derived" (as American Can suggests). The question then is whether under Wisconsin trade secret law "substantially" modifies "the same device" or "derived."

Here, DSC argues that while it is generally permissible to use "substantially" to modify "the same device," American Can carves out a special class of cases (into which this case falls) in which "substantially" must modify "derived." "American Can held that when the alleged trade secret is a particular combination of well-known and readily-available components (as here), a misappropriation finding requires that the defendant's device be substantially derived from the plaintiff's secrets." Reply Br. at 26.

We do not read American Can to stand for such a proposition. In American Can we were simply interpreting and applying Wisconsin law, and we cited M. Bryce for the proposition that a defendant misappropriates the plaintiff's trade secret if her product "is substantially derived from the trade secret." American Can, 742 F.2d at 329. While, as this case demonstrates, skilled attorneys can articulate a nuanced difference between these formulations, American Can intended no such distinction, and the trial court did not err in offering the instruction taken directly from M. Bryce. In any event, it "is well established that a trial court has broad discretion when instructing a jury. If an appellate court can determine that the overall meaning communicated by the instruction as a whole was a correct statement of the law, and the instruction comported with the facts of the case at hand, no grounds for reversal exist." White v. Leeder, 149 Wis. 2d 948, 440 N.W.2d 557, 559-60 (1989). Because the trial court's instructions did exactly that, they provide no grounds for reversal.

IV. Sufficiency of the evidence.

DSC next contends that no reasonable jury could conclude that it misappropriated Sokol's trade secret on the evidence presented. It therefore contends that it is entitled to a judgment as a matter of law. In the alternative, it contends that the jury's decision was against the weight of the evidence, and moves for a new trial.

A. Judgment as a matter of law.

The rule in the Seventh Circuit is that in diversity cases, a district court should look to state law to determine the standard to be applied to a motion for judgment as a matter of law.1 See Bilski v. Scientific Atlanta, 964 F.2d 697, 699 n. 2 (7th Cir.1992); Fort Howard Paper Co. v. Standard Havens, Inc., 901 F.2d 1373, 1382 (7th Cir.1990). But see Gudgel v. Southern Shippers, Inc., 387 F.2d 723 (7th Cir.1967) (finding it "well settled" that federal law controls); 9 Charles Alan Wright & Arthur Miller, Federal Practice and Procedure, Sec. 2525 (1971) (noting this tension in this court's precedents and arguing persuasively for the federal test).

In Wisconsin, a motion challenging the sufficiency of the evidence is to be denied unless "the court is satisfied that, considering all credible evidence and reasonable inferences therefrom in the light most favorable to the party against whom the motion is made, there is no credible evidence to sustain" the verdict. Wis.Stat. Sec. 805.14(1).

But even though a district court is to apply the state law standard in deciding the motion, the standard of review that we apply in reviewing the district court's determination is a federal standard. That is, we review the district court's decision de novo, and we do so even if a state appellate court would be more deferential. While our precedents have perhaps failed to make this entirely clear, announcing that "we 'apply the same standard as the court below,' " see Commercial Credit Equipment Corp. v. Stamps, 920 F.2d 1361, 1365 (7th Cir.1990) (brackets omitted) (quoting F.W. Hempel & Co. v. Metal World, Inc., 721 F.2d 610, 613 (7th Cir.1983)), this is simply to say that we review the district court's decision de novo.

DSC advances a number of reasons why, in its view, there is no credible evidence to sustain the verdict. First, it contends that there was no evidence that its VCXO was derived from Sokol's trade secret. While it admits that there was evidence that its VCXO was similar to Granger's, it maintains that similarity alone does not suffice to prove misappropriation. Before a jury could reasonably find that it misappropriated the trade secret, there would need to be some evidence that DSC had access to Sokol's VCXO design and that it used this confidential information.

We agree that a reasonable jury could not conclude that a trade secret was misappropriated if there were no evidence before the jury that the defendant had access to that trade secret. But the simple fact that Sokol shipped its product to DSC would likely be sufficient evidence of access to sustain the jury's verdict. But there is more, for Alex Dadnam, a former DSC engineer, testified in a deposition (that was read to the jury) that he "took the caps off [of the Sokol VCXOs] to see what was there." R.O.A. 106:135-36. While there was no direct evidence that Crawford received any information from Dadnam, this is not surprising, since any plaintiff would be hard pressed to present direct proof of the flow of information inside the defendant's company. Here, the engineer who designed DSC's VCXO had access to this information. From the evidence introduced at trial, a reasonable jury could certainly have inferred that Crawford had access to Sokol's trade secrets.

DSC also points to the absence of direct evidence that it used Sokol's trade secret, even if it did have access to it. But here again, once the jury concluded that (1) DSC had access to Sokol's trade secrets, and (2) DSC's product was similar to Sokol's, it was entirely reasonable for it to infer that DSC used Sokol's trade secret. See Droeger v. Welsh Sporting Goods Corp., 541 F.2d 790 (9th Cir.1976) (suggesting that where a trade secret is disclosed to the defendant, and it then manufactures a closely similar device, the burden shifts to the defendant to prove that it did not use the trade secret); cf. Composite Marine Propellers v. Van Der Woude, 962 F.2d 1263, 1267-68 (7th Cir.1992) (per curiam) (finding that the devices were too dissimilar to allow an inference that the trade secret was misappropriated). While the jury here was perhaps not compelled to draw an inference of misappropriation from the circumstantial evidence presented, we cannot say (especially in light of the fact that any direct evidence on this point would also be firmly in the defendant's control) that it was unreasonable to do so.

B. Motion for a new trial.

While it is less than clear why a motion for a new trial should be treated differently than a motion for judgment as a matter of law for purposes of the Erie doctrine, it is well-settled that the federal standard governs such motions. That standard is that a district court is permitted to grant such a motion where the jury's verdict is against the weight of the evidence. See Fed.R.Civ.P. 59; Byrd v. Blue Ridge Cooperative, 356 U.S. 525, 540, 78 S. Ct. 893, 902, 2 L. Ed. 2d 953 (1958) ("The trial judge in the federal system has ... discretion to grant a new trial if the verdict appears to him to be against the weight of the evidence.").

Moreover, appellate review of the judge's decision is extremely limited. This, too, is a federal standard. "Under federal law, '[t]he district court's denial of a motion for a new trial should be overturned ... only where the circumstances reveal a clear abuse of discretion.' " Blumenfeld v. Stuppi, 921 F.2d 116, 118 (7th Cir.1990) (quoting Cook v. Hoppin, 783 F.2d 684, 688 (7th Cir.1986)); Bilski, 964 F.2d at 700 n. 3.2 Thus, the question is ultimately "whether the district court abused its discretion in concluding that the jury did not abuse its discretion." Galard v. Johnson, 504 F.2d 1198, 1202 (7th Cir.1974). On the basis of the evidence surveyed above, it is clear that the trial court's decision not to grant a new trial was firmly within the scope of its discretion, and we are thus disinclined to second-guess this determination.

V. Damages.

DSC also argues that the damage award should be reversed.3 It first argues that Sokol's damages should be limited to the "head start" that AFD received by virtue of its misappropriation. It is true that, where a misappropriation of a trade secret only gives a competitor a "head start" in developing a product, damages should be limited to the injury suffered in that "head start" period, see Minuteman Inc. v. Alexander, 147 Wis. 2d 842, 434 N.W.2d 773, 779 (1989). But the jury here was instructed that it could award lost profits "only if you determine that the wrongful act of the defendant caused the loss," and that it could "award plaintiff the cost that defendant would have incurred in acquiring the trade secrets through its own experiments or lawful means to the extent that you have not previously taken that cost into account in determining plaintiff's damages." R.O.A. 199. The point of the "head start" period is that, once the defendant has discovered, or would have discovered, the trade secret without the misappropriation, any lost profits from that time forward are not caused by the defendant's wrongful act. Because we presume that jurors understand their instructions, see Gacy v. Welborn, 994 F.2d 305, 313-14 (7th Cir.1993), we can assume that they did not award damages for lost profits after the head start period. Here, in any event, the jury may well have concluded that DSC would never have developed a workable VCXO had it not misappropriated Sokol's trade secret.

For proof of damages, Sokol relied on the testimony of Gordon Meicher. Meicher looked to the costs of material, labor and overhead, and determined that it cost Sokol approximately $11.50 to produce each VCXO. To determine the price at which Sokol would have sold the VCXOs, Meicher looked at the market price. To that end, he observed that the price at which Sokol sold the VCXOs to Granger fluctuated greatly. Similarly, AT & T sold VCXOs to DSC at a price that ranged from $26 to $46 each. Meicher therefore estimated that the devices would sell for $35 each. Thus, he concluded that for every VCXO that Sokol would have sold to DSC but did not because of the misappropriation of its trade secret, Sokol lost approximately $23.50 in profits.

In determining the number of units that Sokol would have sold, for the period prior to trial, Meicher looked at the units that AFD (DSC's subsidiary) sold to DSC. For future sales, Meicher looked to the average number of units sold in the past, and estimated that those sales would decline over an eight-year period as other technologies advanced. For both the past and future sales, Meicher determined the present value of the lost profits by assuming a 6.5 percent interest rate. Employing this calculus, he estimated the lost profits to be $5,138,396.

The jury apparently accepted some, but not all, of Meicher's testimony, returning a verdict for $2,492,000. As for the attack on the sufficiency of the evidence, and applying the Wisconsin standard, there is, in our view, sufficient credible evidence to sustain the verdict. Wis.Stat. Sec. 805.14(1). DSC objects to a few particular elements of Meicher's damage estimate that it says are unsupported by the evidence. First, it claims that Meicher included over $400,000 in research and development costs as an element of damages. But, because the initial profit calculation included a recovery for the cost of research and development, the initial profit estimate of more than $5.5 million was reduced by $400,000. It is not clear to us why the research and development costs were included in the first place, but, in any event, as noted, they dropped out.

DSC also argues that Meicher's estimate assumed a 200 percent profit margin. It contends that this is against the weight of the evidence, because evidence was introduced at trial that in November 1987 Sokol asked DSC for a price increase sufficient to allow it to make a profit, but agreed that once it attained a 20 percent profit rate on the VCXOs, all further cost savings would be passed on to DSC. But this argument only concerns one factor among several. That factor does not determine whether a jury could reasonably believe on the evidence presented that Sokol would be able to produce VCXOs at $11.50 each and sell them for $35. That being the case, the alleged profit ceiling provides an inadequate basis for reversing the jury's finding. The district court was surely within its discretion in concluding that the jury's finding was not against the great weight of the evidence (and therefore refusing to grant a new trial). See Tony Spychalla Farms v. Hopkins Agricultural Chemical Co., 151 Wis. 2d 431, 444 N.W.2d 743, 748-49 (App.1989).

It does appear, however, that one technical error was made. As noted, in Meicher's estimate he adjusted the profit figure to present value, both for future and past sales. With respect to the past sales, however, adjusting to present value is equivalent to awarding prejudgment interest, which is forbidden in this context under Wisconsin law. See Tony Spychalla Farms, 151 Wis. 2d 431, 444 N.W.2d at 750 (App.1989); D'Huyvetter v. A.O. Smith Harvestore, 164 Wis. 2d 306, 475 N.W.2d 587, 594 (App.1991).

But DSC did not object to the introduction of any of this evidence either before or during trial. Rather, it appears that DSC first raised this issue in its post verdict motion. R.O.A. 213:36. That is too late. A motion for a new trial or for judgment as a matter of law "cannot be used to raise arguments which could, and should, have been made before the judgment issued." Federal Deposit Insurance Co. v. Meyer, 781 F.2d 1260, 1268 (7th Cir.1986).4

But ignoring, for a moment, the waiver, the trial court's suggestion that this was a harmless error is problematic. The district court's response to the argument that some "costs were improperly incorporated" is that they must have been taken into account by the jury, "since it awarded a damage amount of less than one-half of that testified to and requested by plaintiff's expert." Memorandum and Order (March 30, 1993) at 8-9.

In this circuit the harmlessness standard is treated as procedural and therefore in a diversity case is governed by federal law and Fed.R.Civ.P. 61. Simmons v. Pinkerton's, Inc., 762 F.2d 591, 597 n. 2 (7th Cir.1985).5 Here, the error involved allowing evidence of the interest that Sokol would have earned on its VCXO income to reach the jury. In so far as waiver and harmless error analysis are related, the error can be seen as harmless because it was waived. See 11 Charles Alan Wright & Arthur Miller, Federal Practice and Procedure, Sec. 2885 (1973) (an "error will be treated as harmless if the evidence was not objected to"). But, on the merits, simply because it was possible that the jury did not count a particular erroneous measure of damages (and this is not a case where the court's instruction cured the error, see Purex Corp. v. St. Louis Nat. Stockyards Co., 374 F.2d 998 (7th Cir.1967)), we cannot conclude that the error therefore did "not affect the substantial rights of the parties." Fed.R.Civ.P. 61. Cf. 11 Wright & Miller, Sec. 2885 ("if the evidence is insufficient to support the verdict without the erroneously admitted evidence, the error must be held prejudicial").6 But because DSC did not make a timely objection to the introduction of this evidence, its claim is now waived.

VI. Willful and malicious.

Sokol cross-appeals on the grounds that the trial court erred in not allowing the jury to determine whether the misappropriation was willful and malicious under the Wisconsin statute, Wis.Stat. Sec. 134.90(4)(b). But the statute says that if a misappropriation is willful and malicious, "the court may award punitive damages" (emphasis supplied). The district court, in refusing a willfulness instruction, said that there is "nothing in the record to demonstrate that the act was malicious." R.O.A. 211:181. It is therefore clear that the court would not have awarded punitive damages, and thus that failing to give the instruction (even if it was error) was not prejudicial. The district court's judgment is therefore affirmed.