B. Scott Marshall v. Gipson Steel, Inc.

IN THE SUPREME COURT OF MISSISSIPPI NO. 2000-CA-01901-SCT B. SCOTT MARSHALL, INDIVIDUALLY and d/b/a S.A.M. ESTIMATING SOFTWARE and MARSHALL ENTERPRISES, INC. A MISSISSIPPI CORPORATION v. GIPSON STEEL, INC., A MISSISSIPPI CORPORATION

DATE OF JUDGMENT: 10/16/2000 TRIAL JUDGE: HON. SARAH P. SPRINGER COURT FROM WHICH APPEALED: LAUDERDALE COUNTY CHANCERY COURT ATTORNEYS FOR APPELLANT: W. WHITAKER RAYNER JOHN G. COMPTON ATTORNEY FOR APPELLEE: GREGORY MALTA NATURE OF THE CASE: CIVIL - OTHER DISPOSITION: REVERSED AND RENDERED- 01/31/2002 MOTION FOR REHEARING FILED: MANDATE ISSUED: 2/21/2002

BEFORE McRAE, P.J., COBB AND DIAZ, JJ.

COBB, JUSTICE, FOR THE COURT:

¶1. Gipson Steel, Inc. ("GSI"), a Meridian corporation, filed this action seeking a permanent injunction against B. Scott Marshall, both individually and d/b/a S.A.M. Estimating Software and d/b/a Marshall Enterprises, Inc.(1) GSI alleges that Scott, a former employee, and his brother Alan seek to market software which uses proprietary estimating data and procedures that are trade secrets of GSI. On April 23, 1999, the Lauderdale County Chancery Court issued a temporary restraining order which forbade the marketing, sale or distribution of the Marshall software. The chancellor then heard oral arguments on the temporary restraining order before concluding that the order should not be dissolved pending trial. At the conclusion of a bench trial, the chancellor concluded that the Marshall software violated the Mississippi Uniform Trade Secrets Act (MUTSA), Miss. Code Ann. §§ 75-26-1 to -19 (2000), and permanently enjoined Marshall "from marketing or otherwise disseminating the software programs which contain the trade secrets of [GSI]." Marshall now appeals, raising the following issues:

I. DID THE CHANCELLOR BELOW ERR AS A MATTER OF LAW IN ISSUING TEMPORARY, PRELIMINARY, AND PERMANENT INJUNCTIVE RELIEF BY FINDING THE EXISTENCE OF PROTECTABLE TRADE SECRETES OF GIPSON STEEL UNDER MISS. CODE ANN. § 75-26-3, NAMELY IN FINDING

A. ARTICULATED TRADE SECRETS RELATING TO THE STEEL ESTIMATION PROCESS;

B. WHICH GAVE GIPSON STEEL ECONOMIC VALUE;

C. FROM NOT BEING GENERALLY KNOWN IN THE INDUSTRY; AND

D. WHICH WERE THE SUBJECT OF REASONABLE EFFORTS TO PROTECT THEIR SECRECY?

II. ALTERNATIVELY, WERE THE COURT'S RULINGS AND FINDINGS IDENTIFIED IN ISSUE I ABOVE NOT SUPPORTED BY SUBSTANTIAL EVIDENCE, MANIFESTLY WRONG, CLEARLY ERRONEOUS, OR THE RESULT OF APPLICATION OF A WRONG LEGAL STANDARD?

III. WAS THE CHANCELLOR'S RULING WHICH EQUATED CUSTOM IN THE INDUSTRY AS A LEGALLY SUFFICIENT GROUND FOR FINDING REASONABLE STEPS TO PROTECT TRADE SECRETS AN ERROR OF LAW?

IV. WAS THE COURT'S FINDING THAT THE SOFTWARE PROGRAM SOUGHT TO BE MARKETED BY THE APPELLANTS CONTAINED TRADE SECRETS OF GIPSON STEEL NOT SUPPORTED BY SUBSTANTIAL EVIDENCE, MANIFESTLY WRONG, CLEARLY ERRONEOUS, OR THE RESULT OF APPLICATION OF A WRONG LEGAL STANDARD?

¶2. Ultimately, these four issues can be condensed to the following two questions:

I. DID THE CHANCELLOR ERR IN CONCLUDING THAT GSI COULD ARTICULATE A TRADE SECRECY INTEREST IN THE "GIPSON PROCESS" FOR ESTIMATING STEEL COSTS?

II. DID THE CHANCELLOR ERR IN FINDING THAT THE MARSHALL SOFTWARE INCORPORATED ELEMENTS OF THE "GIPSON PROCESS" WHICH COULD BE SUBJECT TO TRADE SECRECY LAWS?

¶3. Finding the first issue dispositive, we reverse and render in favor of Marshall.

FACTS

A. The Parties.

¶4. GSI is a Meridian steel fabricator, i.e. a contractor which fabricates steel components for buildings and other construction projects. Scott Marshall is a former estimator for GSI whose job was to estimate how much GSI could bid on a project and still make a profit. Scott is currently an independent contractor. The dispute at the core of this case involves a computer program written by Scott's brother, Alan, for use in steel fabrication estimation. The original version of this program ("the Gipson program") was written at Scott's request on behalf of GSI, and Alan has apparently never been paid for it.(2) Scott and Alan seek to market a modified version of the program ("the Marshall program") to other steel fabricators. GSI, however, claims that both the Gipson program and the Marshall program rely on proprietary information which GSI uses in bid estimation. GSI further alleges that it has a protectable trade secret in this proprietary information. Originally, bids were estimated by hand in a sequence of events that GSI refers to as "the Gipson Process." It is this Gipson Process which the Gipson program essentially replaced. An overview of the bid estimation process as it is used both industry-wide and by GSI in particular is instructive.

B. The Bid Estimation Process.

¶5. In the steel fabrication industry, estimators review plans for proposed construction projects and calculate how much it would cost the firm to perform the project, a crucial first step in the process of submitting a bid. According to Robert Ryan, a court-appointed expert on steel fabrication industry practices, the estimation process contains a number of steps which are essentially the same for all steel fabricators. First, blueprints are studied and the steel parts to be used are "taken off", which means that an estimator reviews blueprints and calculates the total amount of steel which will be required for the job, assigning a weight to each steel piece based on the category of steel member into which it falls. Apparently, the cost of steel fluctuates depending on whether it is made into a wide flange beam, a pipe, a plate, etc. All weights and sizes, however, are set by the American Institute of Steel Construction (AISC) and are constant throughout the steel fabrication industry. Assuming the take off was done accurately, the quantity of steel derived from the blueprints should be the same for any estimator who prepares a bid. The final results of the take off are typically recorded on some type of recap sheet.

¶6. Next, the estimator applies a cost per pound to the total quantity of steel derived from the blueprints. This per-pound cost may either be derived from recent steel purchases, or the estimator may contact the actual source of the steel itself to learn what the charges would currently be for the quantity of steel needed. While there may be minor fluctuations in the cost of steel (such as when a fabricator having some left over from a prior job, thereby decreasing his steel needs), the cost of steel is generally constant across the industry. The total cost of the steel required is then added to the recap sheet.

¶7. Next, the estimator adds the costs associated with miscellaneous items such as joists (parallel beams used to support a floor or ceiling), decks, ladders and stairs. These items are generally acquired from independent suppliers, and the prices are also basically constant across the industry. Estimators must also calculate how much paint is required and add that to the recap sheet. The next factor to consider is the cost of bolts and fasteners, which allows for some flexibility. Some companies calculate a specific price based on the number of bolts and fasteners that will actually be needed for a job, while other companies will simply charge an overhead of so many dollars per ton of steel to cover bolts. All estimators, however, consider bolts and fasteners as part of the estimation process.

¶8. Estimators must also include the total cost of freight to the job site, which will also be "pretty close" for everyone in the industry. Typically, steel fabricators do not bid on jobs a significant distance from their place of business precisely because local fabricators would have a substantial advantage over out-of-town fabricators due to lower freight costs.

¶9. According to the court-appointed expert, only two things are actually "estimated" during the estimation process. First, the estimator must determine what the labor costs are expected to be. This will fluctuate from job to job depending on its complexity, as a low, flat building such as a warehouse is much easier to make than something more complicated like a church with lots of steeples and ornate windows. Second, the estimator must assess the costs of detailing work. Detailing is poorly defined by the record, but apparently consists of reviewing blueprints to determine how the steel pieces should be cut and fit together. As with labor, detailing costs will fluctuate according to the relative complexity of the job. Essentially everything else consists of plugging set measurements and costs into a formula.

¶10. The final factor added to the recap sheet by the estimator is the amount of profit the company plans to make in excess of its costs. The court-appointed expert testified that the only information he would not want

a competitor to see (of the information on the recap sheet, which is not included in the final bid) would be: (1) the number of hours per ton of steel he expected the job to require, (2) the cost of the hours per ton in labor, (3) how much he charged for detailing, and (4) the percentage of profit. The expert did concede, however, that it was possible to estimate the number of hours that a job would require with some accuracy and that most steel companies have a profit margin of between 3% and 5%.

C. The Two Programs.

¶11. The original Gipson Program was created for GSI by Alan Marshall in 1998. GSI had apparently sought a software program for some time which would reduce the amount of time necessary to tabulate the various components of the estimation process, and presumably eliminate human error as well. GSI rejected several programs which would have required it to change its estimation procedures before Scott suggested Alan's services. The centerpiece of what GSI refers to as the "Gipson process" for estimation is a specific recap sheet which the company has used since its founding. The recap sheet itself is based on one which Hoot Gipson, GSI's founder, used when he was previously employed at a different and now defunct company, Bates Steel. The final product generated by Marshall software is similar (but not identical) to the Gipson recap sheet, and GSI's chief complaint about the Marshall software appears to be that it would allow competitors to essentially reconstruct the Gipson recap sheet and thereby determine how much GSI charges for labor and detailing. GSI also claims a proprietary interest in the "definition codes" that both the Gipson software and the Marshall software use to route information to specific locations on the recap sheet. During oral arguments, GSI clarified that definition codes are simply computer codes which determine into which column of the recap sheet a specific piece of information will be placed. Moreover, as Marshall points out, the term was used at trial interchangeably with the "description codes" which identify each steel component and in which GSI has conceded that it has no proprietary interest.

STANDARD OF REVIEW

¶12. This Court will not interfere with a chancellor's findings of fact unless they are "manifestly wrong, clearly erroneous or an erroneous legal standard was applied." Bell v. Parker, 563 So.2d 594, 596-97 (Miss. 1990). However, we review the chancellor's interpretation and application of the law de novo. In re Carney, 758 So.2d 1017, 1019 (Miss. 2000).

ANALYSIS

I. DID THE CHANCELLOR ERR IN CONCLUDING THAT GSI COULD ARTICULATE A TRADE SECRECY INTEREST IN THE "GIPSON PROCESS" FOR ESTIMATING STEEL COSTS?

¶13. The MUTSA defines a trade secret as follows:

"Trade secret" means information, including a formula, pattern, compilation, program, device, method, technique or process,(3) that:

(i) Derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use, and

(ii) Is the subject of efforts that are reasonable under the circumstances to maintain its secrecy. Miss. Code Ann. § 75-26-3(d)(2000).

¶14. Although the MUTA does not define "proper means," § 75-26-3 (a) does define 'Improper means' to include "theft, bribery, misrepresentation, breach or inducement of a breach of a duty to maintain secrecy, or espionage through electronic or other means." The MUTSA is identical to the Uniform Trade Secrets Act (UTSA) as promulgated by the National Conference of Commissioners on Uniform State Laws. Compare Miss. Code Ann. § § 75-26-1 to -19, with Unif. Trade Secrets Act, 14 U.L.A. 437 (1990).

¶15. Fred's Stores of Mississippi, Inc. v. M & H Drugs, Inc., 725 So.2d 902 (Miss. 1998) is the only case in which we have interpreted the MUTSA. There, the trade secret at issue was a customer database which a drug store employee took with him when he left and went to work for a competing drug store, and the dispute was limited to whether the list had been reasonably protected by the plaintiff. Fred's, 725 So.2d at 910. We answered in the affirmative.

¶16. In the case sub judice, Marshall argues that GSI has failed to prove any element of a trade secret. We now consider the disputed software under both prongs of § 75-26-3(d).

A. The first prong.

¶17. In order to satisfy the first prong, the Gipson process must derive "independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use." Miss. Code Ann. § 75-26-3(d)(i) . In assessing whether GSI satisfied this prong, the chancellor stated the following:

It is clear from a review of the facts in this case that the Gipson Steel bid estimating system is a process that has independent economic value to Gipson. It is the means by which Gipson prepares its bids on steel fabricating jobs, and other steel fabricators in competition with Gipson would obtain economic value from that process if it were disclosed, as it would enable them to undercut Gipson's bid on the project and get jobs which might otherwise be awarded to Gipson.

While the chancellor did an admirable job of encapsulating the facts and interpreting the law in an area with sparse precedent, her assessment of the first prong has one glaring omission: the chancellor apparently did not consider whether the secret was "readily ascertainable by proper means."

¶18. The comment to the UTSA specifically lists "[d]iscovery by 'reverse engineering', that is, by starting with the known product and working backward to find the method by which it was developed" as a proper means of ascertaining a trade secret. Unif. Trade Secrets Act § 1 cmt., 14 U.L.A. at 438. This result is consistent with similar cases in other jurisdictions, although many of them are pre-UTSA cases. A California appellate court concluded that methods of pricing and bidding used by an exterminator could not constitute a trade secret. Fortna v. Martin, 323 P.2d 146, 149 (Cal. Ct. App. 1958)("At most, the evidence discloses that Fortna utilized a system which considered his costs, but any successful business man must do the same."). See also Lawler Mfg. Co. v. Bradley Corp., 2000 WL 1456336, at *5 (S.D. Ind. Apr. 26, 2000)("Whether information is readily ascertainable depends upon the degree of time, effort and expense required to duplicate or acquire the information by proper means. . . . The party asserting a trade secret has the burden to show that replicating the information would require substantial investment of time, expense and effort."); McLeod v. Meyer, 372 S.W.2d 220, 223 (Ark. 1963)("The knowledge of how to bid for a job . . . would hardly seem to come under the classification of 'secret formulas' "); Ovation Plumbing,

Inc. v. Furton, 33 P.3d 1221, 1224 (Colo. Ct. App. 2001)(distinguishing earlier cases which held bid estimation procedures were not trade secrets from case where defendant actually took information on specific bids for purpose of bidding against plaintiff).

¶19. As Marshall points out, both GSI's expert witness and the court-appointed expert witness stated that they would be able to determine GSI's labor rates "by hand" if provided with GSI's bids on previous jobs whose steel requirements were known to them. Furthermore, the record is devoid of any evidence by GSI disputing the claim that anyone could figure out GSI's supposedly secret information through simple math. Finally, the description codes in which GSI claims proprietary interest appear to perform no other purpose than to direct non-secret information to a specific location on GSI's recap sheet, and we decline to hold that the mere arrangement of non-proprietary information can be a trade secret. We therefore find that the chancellor misapplied the law (or, if she did consider reverse engineering, her findings in that regard are not supported by substantial evidence).

B. The second prong.

¶20. The second prong considers whether the information was subject to reasonable attempts to preserve its secrecy. In Fred's, we held that storing a customer list in an unlocked file cabinet in the pharmacy area of a drugstore and protecting the computer version of the same list with a password were sufficient to satisfy this prong. 725 So.2d at 911.

¶21. In the case sub judice, GSI testified that the "elements of the process" including the cost recap sheet are never released and never leave the estimating shop, an area to which only specified employees have access. Bid packets are kept in filing cabinets which are in a controlled area but which are not kept locked. Older files are kept either in a monitored area or in a locked metal building on the GSI grounds. GSI also testified that only three employees have access to the computer terminals in which the program is stored. GSI conceded that Marshall was never bound to any sort of non-disclosure agreement, but argues that Marshall was made aware of the sensitive nature of the bidding process and that knowledge placed an obligation on him to not reveal the knowledge in a manner that would amount to breach of confidence or bad faith.

¶22. Although the security measures taken by GSI are at least comparable to those employed by Fred's, the GSI facility arguably is inherently less secure than a pharmacy. While we have some concern about upholding the existence of a trade secret where an employer has not bound his employees with any type of non-disclosure agreement and instead merely "trusts them to be honest," such an agreement is not per se required under a § 75-26-3(d) analysis. On the whole, we cannot say that the chancellor's findings are not supported by substantial evidence with regard to this prong, although it is a close call. In any case, the point is moot, as GSI clearly does not satisfy the first prong.

II. DID THE CHANCELLOR ERR IN FINDING THAT THE MARSHALL SOFTWARE INCORPORATED ELEMENTS OF THE "GIPSON PROCESS" WHICH COULD BE SUBJECT TO TRADE SECRECY LAWS?

¶23. Marshall argues that GSI's claim must still fail because GSI did not prove that any of its trade secrets were incorporated into the Marshall software. Specifically, Marshall points to the fact that no expert who testified was actually allowed to see both programs in operation. Furthermore, Alan Marshall was not permitted to describe differences between the programs because the chancellor concluded that doing so fell

under the rubric of expert testimony for which GSI was not given notice. The evidence supporting the finding that GSI's trade secrets are actually incorporated into the software is both sparse and speculative, although some evidence does exist in support of the chancellor's findings. However, because we reverse and render on Issue I, we need not address the merits of Issue II.

CONCLUSION

¶24. Based on the foregoing reasoning, we reverse and render in favor of Marshall, as the alleged trade secret at issue is information which could be readily ascertainable through reverse engineering. The judgment of the Lauderdale County Chancery Court is reversed, and judgment is rendered here finally dismissing GSI's complaint and this action with prejudice.

¶25. REVERSED AND RENDERED.

PITTMAN, C.J., McRAE AND SMITH, P.JJ., WALLER, DIAZ AND CARLSON, JJ., CONCUR. EASLEY, J., DISSENTS WITHOUT SEPARATE WRITTEN OPINION. GRAVES, J., NOT PARTICIPATING.

1. The defendants are collectively referred to as "Marshall." When B. Scott Marshall is referred to in his individual capacity, he will be called "Scott."

2. There are also indications of a copyright dispute over ownership of the program itself. However, that issue was not considered by the chancery court and is not before this Court today.

3. Terms such as "information" and "process" are not defined by the Act itself. For the purposes of this opinion, they are interpreted as broadly as possible, making potentially any business practice a trade secret if the two prongs of §75-26-3(d)(i)-(ii) are met. The injunction identifies GSI's system for estimating bids as a "process," which may be defined as "a systemic series of actions directed to some end." Random House Webster's College Dictionary 1075 (2000). We agree that the "Gipson process" meets this definition.