Berkla v. Corel Corp.

66 F. Supp. 2d 1129, 1999 U.S. Dist. LEXIS 14471, 1999 WL 742593

• Court: District Court, E.D. California
• Decided: September 9, 1999
• Docket: Civ. S-98-1159 GGH
• Status: Published

Opinion

ORDER

HOLLOWS, Chief United States Magistrate Judge.

INTRODUCTION AND SUMMARY

All art is but imitation of nature. 1

Taking Seneca’s admonition to heart, plaintiff, Dennis Berkla, .has established electronic databases (Garden Hose) containing virtual leaves, blades of grass, flowers, pine cones and the like which constitute a software “nozzle,” which is then digitally “sprayed” to create computerized, realistic pictures of plants, trees, forests, gardens on the computer screen. After supplying some of these databases to defendant Corel for possible inclusion in upcoming Corel software revisions, and after having had Corel reject Berkla’s submissions. Berkla became aware that Corel-DRAW 8, with its application program PHOTO-PAINT 8, contained database images similar to Berkla’s. Corel’s application program performed essentially the same tasks as set forth above. Berkla thereafter sued Corel (Second Amended Complaint) for:

1. Copyright Violation

a. the direct uploading of 79 Berkla files into Corel’s systems with subsequent dissemination to beta testers;

b. the individual components, e.g., leaves, grasses, branches, trunks, pine cones and so forth, of the Corel release were substantially similar to those in Garden Hose 1.5, and consisted of two main groupings:

1. those Corel components which have equivalents in Berkla files;

2. those Corel components which do not have equivalents, but nonetheless are similar enough in form to be a copyright violation;

c. the arrangement of the individual components in the database;

d. “tubular text” infringements.'

2. Breach of Non-Disclosure Agreement (state claim)

3. Breach of Confidence (state claim)

4. Unfair Competition (state claim)

Corel has moved for summary judgment on all claims.

For the reasons set forth herein, Corel is awarded summary judgment on the copyright claims with the exception of that claim pertaining to the dissemination of 79 Berkla files to Corel’s beta testers. Co-rel’s motion as it pertains to state claims is granted in part and denied in part.

COPYRIGHT INFRINGEMENT

Facts

The parties have submitted voluminous filings of undisputed, quasi-disputed, and disputed facts. The court sees no reason to set forth all of them. For the purposes of this summary judgment motion, the court sets forth the important undisputed facts, and also, where necessary and appropriate, the court has adopted the non-moving party’s (Berkla’s) version of disputed facts. Where conflicting inferences are possible, inferences have been drawn in favor of the non-moving party.

A. Facts Related to Nature Images

Plaintiff Berkla (“Berkla”) is a trained artist who owns and operates a company, DigArts Software. Partly at issue in this case are software databases (nozzles) created by Berkla which contain electronic images of plant, tree, flower and bush components (“Garden Hose”). 2 Berkla created the databases by making manual impressions (“drawing” or “painting”) on a computer tablet which the computer translated into colored digital pictures of component leaves, flower clusters, pine cones, etc. which are arranged by particular type in various computer files (database). Each database contained numerous perspectives of the same natural subject. Berkla exercised judgment concerning how many individual component leaves, for example, (including their various perspectives and shadings) were necessary to give the ultimate computer user sufficient choice such that the rendering of a bush, for example, would not have an unrealistic look because of over-duplicative uniform size, perspective and shading of the bush components. Sufficient component leaves were therefore placed in the database for that leaf structure. The court’s description will make more sense if viewed in conjunction with an exhibit supplied by Berkla (attached to this opinion in the court’s Appendix 3 ), which shows an array of poinsettia leaves. 4 The databases, by themselves, are not functional; the databases are combined with a software “image sprayer” or “image hose” to create varying pictures of natural objects and settings. The ultimate creations by the computer operator are not at issue here; the purpose of the software is to allow the computer operator to create realistic singular plants, bushes or flowers, or for practical purposes, a near infinité variety of natural looking gardens and landscapes.

Berkla did not create the functional aspect of his product; rather he designed his databases to be initially used with the Fractal Design Corp. image hose. Reaching an agreement with Fractal, Berkla was permitted to offer his software as an “add-on” to the Fractal program. Berkla received a royalty on every Garden Hose product sold to a Fractal customer.

It is important to the resolution of this case to detail what Berkla was attempting to create in the overall sense: a software tool that would be used to render “photo-realistic” landscapes. Berkla Declaration at 4. “For Berkla’s plant nozzle databases to be capable of creating photo-realistic effects when used with a sprayer tool, the images must have the following design characteristics ... (a) the plants’ leaves flowers or leaves and flowers must be shown as they grow or appear in nature, often from an elevated, frontal or semi profile perspective; (b) depicting naturalistic shadows to define the spatial relationships that commonly exist between the particular plant[’]s leaves, flowers or leaves and flowers; (c) rendering assorted arrays of the plant[’]s leaves, flowers or leaves and flowers so that they characterize the component structures of the plant.” Undisputed Fact 4. 5 “These designs are essential if the images are to apply as coherent media capable of rendering complex, photo-realistic paintings of plants, foliage or nature. As a result, Garden Hose nozzles have the ability to function as coordinated, coherent paint media capable of rendering complex, photo-realistic illustrations of nature.” Berkla Declaration at 16. See also Undisputed Facts 4 and 5.

Seeking a much greater volume of sales than he had experienced with Fractal, Berkla, who had recently initiated a slight business association with Corel as an independent “solutions partner,” referenced his product and its upcoming evolution to Doug Chomyn, the Corel Photo Paint product manager, on March 26, 1997. 6 On or about April 11, 1997, an agreement was reached whereby Berkla sent his work product via CD (the soon-to-be-released Garden Hose 1.5) to Chomyn for purposes of licensing evaluation and possible inclusion in Corel’s upcoming improvement software programs — Corel Draw 8 which included Photo Paint 8. Berkla attempted to protect his work with a non-disclosure agreement; the precise facts pertinent thereto are discussed at length below in the section related to alleged breach of that agreement. Corel uploaded the CD into a format compatible with its programs, although it is undisputed that Berkla’s CD could have been viewed without the format change. Nevertheless, Berkla was placed on notice that his CD would be uploaded such that it was usable in Corel format, Exhibit BB, and the record is silent as to any Berkla contemporaneous objection.

Berkla had evidently achieved his goal of creating an image database with very realistic plant life images. Personnel at Corel were impressed. “Nice stuff. Good Quality” (Pennington Deposition at 68); “very, very good,” (Gray Depo at 68-64); “great deal of imagination, care and artistic integrity” (Derry Depo. at 70). Despite these accolades, Corel did not agree to buy Berkla’s product or arrange for any marketing of Garden Hose with Corel products, and Berkla was informed of this decision on or about May 1, 1997. The record is silent concerning Corel’s precise reasons why it chose not to do business with Berk-la, or at least the parties do not direct the court’s attention to any decision making process. This silence is somewhat mystifying in light of the discovery that has taken place in this case, and the obviousness of that question’s relevance to the facts of this case. Nevertheless, the facts and inferences from those facts in favor of the non-moving party, Berkla, demonstrate that Corel utilized Berkla’s image databases in several ways other than evaluation for possible purchase or add-on development. The precise manner in which Corel allegedly utilized Berkla’s images are better explained with an historical account of Corel’s own efforts in image sprayer technology.

Corel had developed an image sprayer tool in connection with Photo.Paint 7. In function, this sprayer worked much like the Fractal sprayer, and Corel had the ability to spray plant and animal images prior to any contact with Berkla. Corel created its images differently than did Berkla — Corel used a “photo cropping” methodology, i.e., it did not hand draw/ paint its images. At the very least, an issue of fact exists concerning whether the Photo Paint 7 images were inferior to those created by Berkla. The court uses the word “inferior” advisedly, as it understands that it is not in the business of judging artistic value. Nevertheless, if realism in plant imagery was the goal of both Berkla and Corel, the court can fairly say from a review of the respective images that Berkla’s greater use of perspective and more variety in structure gave his images a more realistic look than Corel’s Photo Paint 7 images. One of Corel’s nature image designers (Pennington) offered his opinion in deposition testimony that he was not satisfied with the quality of Corel images — “It didn’t look right to me. It didn’t look good.” Pennington Deposition at 32.

Inferior or not, Corel’s Photo Paint 7 images certainly possessed an attractiveness of their own. The images available were relatively few, but these images and the sprayer tool had caught the attention of Corel’s CEO who exclaimed that he desired “hundreds and hundreds of these images.” The statement was made ostensibly in connection with Corel’s upcoming “8” evolution, and apparently was made in the general time frame of Berkla’s contacts with Corel. Berkla asserts, and an inference can be drawn, that Corel software designers were thereupon pressed to create hundreds and hundreds of images in a short time period so that they would be ready for Photo Paint 8. Berkla further asserts that Corel, being unable to independently create realistic images, used Berkla’s submitted “evaluation” CD as a model on which to emulate the quality of Berkla’s images. There are facts which support this assertion. Pennington Depo. at 66-67. 7 Berkla has produced facts which show that Corel disseminated Berk-la images to its software creators who had little, if anything, to do with licensing evaluation of Berkla’s databases.

Corel compounded the potential for improper use of Berkla’s images, or appearance of impropriety, by actually releasing 79 exact Berkla duplicates in its “8” pre-release version to its Corel Draw/Photo Paint beta testers. 8 Somehow, the database imagery that Corel had uploaded into its files found its way into the beta testers version. Corel deleted these exact images before any of its Draw 8/Photo Paint 8 software was released to the public, but at least at the summary judgment stage, and in the absence of facts concerning how Berkla’s images found their way into pre-release software, an inference can be drawn from this fact alone that Corel was utilizing Berkla’s CD submission for purposes other than licensing evaluation. There is no direct evidence that any of the beta testers kept copies of the Berkla images despite their agreement to destroy all beta test versions sent to them.

Corel’s Photo-Paint 8 images approached, if not equaled, Berkla’s images in photo-realistic quality. However, the court has performed database by database comparisons of seventeen allegedly infringing Corel products along with the corresponding Berkla databases, and does not find any comparison “virtually identical”— the legal standard to be employed in this case which is discussed at length below. The court will now discuss one exemplar database from those in question, the citrus leaf database, because it sufficiently illustrates the types of differences in the remaining databases as well. 9

Attached to this order are color copies of the citrus leaf data base. Court’s Appendix, Exhibits 57A (Corel) and 57B (Berkla). While there are many similarities that follow from the idea of painting a realistic citrus leaf, e.g., general shape and angles of the leaf, general coloration, shadows cast by the -leaf components, which might be expected if the parties are attempting to depict a realistic citrus leaf— some of the expression of the common idea varies between Berkla and Corel databases. The data base array itself is different — Berkla’s contains 28 different perspectives of citrus leaf groupings; Corel contains only 12. Thus, there is no virtual copying (or even substantial similarity in the data base array). The individual leaf groupings themselves exhibit disparity. Berkla’s are darker with more vivid light area highlights; Corel’s are lighter in general. The Corel leaf has more detailed vein structure. The computerized drop down shadowing (a well established computer technique) is different in that Berk-la’s shadowing is sharper, while Corel’s is more blurred. Corel uses different size leaves to a larger degree than does Berkla. The court has no difficulty in finding that the leaf components themselves are not virtually identical — or even close to the plain meaning of those words.

B. Tubular Text (Neon) Claim

The other main component of Berkla’s copyright infringement claim involves his tubular text process. Utilizing ordinary geometric shapes in a software extrusion process, Berkla discovered that he could make abstract, colorful, 3D linear designs. The extruded tubular images are created by using particular brush settings on the sprayer tool. Corel’s Photo Paint 8 (and 7) also contained this capacity. Again, the precise detail of tool settings and so forth is not necessary to resolution of the claim herein, and again, the varieties of finished product by the computer user is not sub ject to a copyright claim herein. The court has included in its appendix Berkla Declaration Exhibit L which shows one of Berkla’s geometric shapes and finished designs as well as Corel’s (made to look similar by Berkla). Berkla claims a copyright in the geometric shapes (spheres and balls) utilized in the extrusion process.

Photo Paint 7, issued prior to any Corel contact with Berkla, contained the ability to create tubular extrusions from balls and spheres, although it is uncertain that Corel had discovered this potential prior to the issuance of Draw8/Photo Paint 8. Berkla has not submitted any direct evidence that anyone involved in designing the balls and spheres files at Corel had any access to Berkla’s tubular text components. The CD submitted by Berkla with plant life imagery had no balls or spheres or other geometric shapes on it. Nor did Berkla and Chomyn ever discuss tubular text. Berkla contends that since at one time in 1997, he had posted one example of a tubular text (tubular neon) extrusion on his website (Undisputed Fact 65), and Chomyn at one time had been directed by Berkla to visit his website, it is possible that this one Corel employee had access to Berkla’s copyrighted work. It is undisputed, however, that the web page contained no programming instructions related to the extrusion process, or instructions on how to reverse engineer the finished extrusion. Berkla further weakly contends that because it is clear many employees of Co-rel had access to his CD leafiflower/plant images even without their conscious knowledge of that fact, perhaps, Corel employees had access to his tubular text even without their conscious knowledge. Perhaps, a Corel employee downloaded the one tubular text example (finished extrusion product), although there is no record support for this surmise.

C. Berkla’s Registration of Copyrights

Berkla first registered his Garden Hose 1.0, 1.5 and 2.0 effective November 17, 1997. Thereafter he submitted a supplement to Form VA for his Garden Hose images on April 19, 1998. Tubular Text was registered effective April 28,1997, and a supplement to this registration was filed on April 28, 1998. Berkla asserts in response that the Tubular Text supplement is not the subject of a copyright registration.

Discussion

The critical issue for the plant life or nature copyright claims centers about the correct standard for determining whether Corel database images, or their array, infringe Berkla’s databases. Berkla takes the position that because there is some evidence of direct copying (the 79 exact duplicates given to the beta testers) there is no need to evaluate the similarity of the infringed and infringing products. In any event, if incorrect on this initial premise, Berkla believes the comparison standard to be one of substantial similarity. Corel counter-asserts that copyrights for realistic, artistic “nature” images are weak copyrights on account of the inherent blending of the ideas and expressions involved in recreating nature in art, i.e., there are very limited ways of expressing the idea of a realistic leaf. Only a virtually identical copying of protected expression will suffice for liability, and in the absence of such, there can be no claim for copyright infringement, regardless of the strength of other evidence of copying or modeling. The parties’ positions repeat themselves for the tubular text issues, but also the court must determine whether there is sufficient evidence of access to Berkla’s tubular text images for purposes of summary judgment.

However, before discussing the critical substantive issues, it is necessary to digress to establish precisely what database imagery is at issue, and whether Berkla has forfeited his right to contest Corel’s alleged infringement because of his failure to produce to the court all exemplars of the precise databases in issue.

A. The Imagery at Issue

Understandably, Corel propounded interrogatories to Berkla seeking to ascer tain the precise identity of those database images that were infringed and infringing. Berkla amended his responses six times, although not all amendments related to the interrogatories at issue. Berkla finally responded as follows.

Interrogatory No. 1:

Separately for each PRE-RELEASE CD and FINAL RELEASE version of CorelDRAW 8 and PHOTO-PAINT 8, identify all images that YOU contend are substantially similar to YOUR copyrighted images [emphasis added].

Response to Interrogatory No. 1:

... Subject to and without waiving the foregoing general and specific objections, Berkla responds as follows: Co-rel’s images that are similar in form and appearance to Berkla’s images, that were created after Corel’s receipt of Berkla’s CD of proprietary images in April 1997, and that thus either infringe Berkla’s images and/or proprietary techniques, or misappropriate, or use or disclose them in an unauthorized manner are identified in the attached Exhibit A. The images identified in Exhibit A are similar to Berkla’s images in that all of them contain: (1) substantially similar use of diverse photo-realistic images of leaves, flowers or leaves and flowers as they grow or appear in nature, often from an elevated, frontal or semi-profile perspective, (2) substantially similar depictions of naturalistic shadows that define the spatial relationships that commonly exist between the particular plant’s leaves, flowers, or leaves and flowers, and (3) substantially similar database arrays of the plant’s leaves, flowers or leaves and flowers so that they characterize the component structures of the plant. Some of them also contain flowers, leaves or flowers and leaves of substantially similar size, shape or specie. Exhibit A contains a complete listing of the CorelDRAW 8 and PHOTO-PAINT 8 which plaintiff believes are substantially similar to his copyrighted images.

Interrogatory No. 2:

Separately for each PRE-RE LEASED [sic] CD and FINAL RELEASE version of CorelDRAW 8 and PHOTO-PAINT 8, identify all images that YOU contend are virtually identical to YOUR copyrighted images, [emphasis added]

Response to Interrogatory No. 2:

... Subject to and without waiving the foregoing general and specific objections, Berkla responds as follows: Co-rel’s images that are identical in form and appearance to Berkla’s images contained on the CD Berkla provided Corel on or about April 11, 1997, and that thus either infringe Berkla’s images, or misappropriate, or use or disclose them in an unauthorized manner are identified in the attached Exhibit B.

Because Berkla does not know what defendants mean by “virtually identical” and how that phrase differs from “substantially similar,” Berkla has included on Exhibit A all images that are not identical to Berkla’s copyrighted images, but which have a degree of similarity equal or greater to “substantial similarity.”

Interrogatory No. 3:

Identify each of YOUR works of authorship which you contend was infringed by COREL, [emphasis added]

Response to Interrogatory No. 3:

... Subject to and without waiving the foregoing general and specific objections, Berkla responds as follows: Berk-la’s original images that have been infringed, misappropriated, and used or disclosed in an unauthorized manner by Corel are identified in the attached Exhibit C, by Berkla’s directory and file names and by the Bates-stamp number of the Corel CD on which the corresponding offending Corel images appear. Berkla’s proprietary techniques that Co-rel has also infringed, misappropriated, and used or disclosed in an unauthorized manner are described in Berkla’s copyright registration papers already produced.

Interrogatory No. 4:

For each work of authorship (or element thereof) that YOU contend is infringed by CorelDRAW 8, PHOTO-PAINT 8 or any other COREL software product at issue in this litigation, identify the element of the work infringed and each and every screen shot from, image contained in, or element of the COREL software at issue in this litigation which YOU contend infringed that work (or element thereof).

Response to Interrogatory No. 4:

... Subject to and without waiving the foregoing general and specific objections, Berkla responds as follows: An illustration of the extent and nature of Corel’s infringement of Berkla’s images and proprietary techniques can be seen in a visual review of Berkla’s images identified in the attached Exhibit C, and a visual review of images on the Corel CDs identified by Bates-stamp numbers in Exhibit C. A listing showing how Berkla’s and Corel’s images correspond is attached as Exhibit D. The images identified in Exhibit D include all those that are responsive to this interrogatory. However, Exhibit D does not include those images of Corel’s that infringe, misappropriate, and use or disclose Berkla’s proprietary techniques for which there are no Berkla images with directly corresponding plant or other objects depicted as the subject matter.

Corel contends that Berkla’s interrogatory response number 4 identifying, in Exhibit D, seventeen infringing Corel database images limits potential liability to only those seventeen databases. 10 Berkla argues that there need not be one-to-one correlation between infringed and infringing databases. That is, it is possible that one Corel infringing database can infringe upon several Berkla databases because although the subject matter of the databases may be different, the copied artistic expressions may nevertheless be similar. Therefore, the court should review not only the Corel seventeen, but all the Berk-la databases which Berkla contends were infringed.

The court finds Berkla’s position to be theoretically correct, but practically flawed in this case. First, if Berkla has a duty in opposition to summary judgment to produce all exemplars of infringed and infringing databases (see below), and has not done so, Berkla will be in no position to contend that the absent databases are in issue. Second, if the correct test is “virtually identical,” which the court finds below is appropriate, Berkla will be hardpressed to ever demonstrate that a database with different subject matter from a Corel database was identically copied in terms of its protectable expression. Berkla conceded as much at hearing.

Thus, although all listings in the interrogatories are initially at issue, the court’s ruling as to the seventeen for which comparison exemplars have been provided will govern the case in any event.

B. Summary Judgment Standards

General summary judgment standards, which do apply as well to copyright cases, are well known, and are expressly in issue here because of Corel’s claims that Berkla had a duty to produce for the court’s review exemplars of each and every alleged infringed product along with the infringing Corel product.

Summary judgment is appropriate when' it is demonstrated that the standard set forth in Fed.R.Civ.P. 56(c) is met. “The judgment sought shall be rendered forthwith if ... there is no genuine issue as to any material fact, and ... the moving party is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c).

Under summary judgment practice, the moving party

[A]lways bears the initial responsibility of informing the district court of the basis for its motion, and identifying those portions of “the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any,” which it believes demonstrate the absence of a genuine issue of material fact.

Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 2553, 91 L.Ed.2d 265 (1986).

“[W]here the nonmoving party will bear the burden of proof at trial on a dispositive issue, a summary judgment motion may properly be made in reliance solely on the ‘pleadings, depositions, answers to interrogatories, and admissions on file.’” Id. The moving party who does not bear the burden of proof at trial is not required to produce evidence which negates the opponent’s claim. Celotex, 477 U.S. at 323, 106 S.Ct. at 2552; Lujan v. National Wildlife Fed’n, 497 U.S. 871, 885, 110 S.Ct. 3177, 3187, 111 L.Ed.2d 695 (1990). Accordingly, as to issues on which the non-moving party will bear the burden of proof at trial, the moving party need only point to materials on file which demonstrate the absence of a genuine issue of material fact. Celotex, 477 U.S. at 323-24, 106 S.Ct. at 2553.

If the moving party meets its initial responsibility, the burden then shifts to the opposing party to establish that a genuine issue as to any material fact actually does exist. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986). The nonmoving party who bears the burden of proof at trial “must establish each element of his claim with significant probative evidence tending to support the complaint.” Barnett v. Centoni, 31 F.3d 813, 815 (9th Cir.1994) (internal quotations omitted).

The evidence of the opposing party is to be believed, Anderson v. Liberty Lobby, 477 U.S. 242, 255, 106 S.Ct. at 2513, 91 L.Ed.2d 202, and all reasonable inferences that may be drawn from the facts placed before the court must be drawn in favor of the opposing party, Matsushita, 475 U.S. at 587, 106 S.Ct. at 1356 (1986). Nevertheless, inferences are not drawn out of the air, and it is the opposing party’s obligation to produce a factual predicate from which the inference may be drawn. Richards v. Nielsen Freight Lines, 602 F.Supp. 1224, 1244-45 (E.D.Cal.1985), aff'd, 810 F.2d 898, 902 (9th Cir.1987). Conclusory statements are insufficient to defeat a properly supported motion for summary judgment. Scott v. Rosenberg,, 702 F.2d 1263, 1271-72 (9th Cir.1983).

These general summary judgment standards apply in copyright cases. However, due to the objective impreciseness inherent in the term “substantial similarity,” and to a lesser degree in “virtually identical,” the court should be very careful in determining non-appropriation of protected expression as a matter of law. Frybarger v. Int. Business Machines Corp., 812 F.2d 525, 528 (9th Cir.1987). Nevertheless, when entry of summary judgment is proper, a court should not be unduly hesitant to enter it. Id.

Whether derived as interpretation of the best evidence rule, or simply a requirement of summary judgment standards, courts have required the non-moving party with the burden of proof in copyright cases to produce the alleged infringed and infringing products for comparison purposes at the summary judgment stage. Brown Bag Software v. Symantec Corp., 960 F.2d 1465, 1476 (9th Cir.1992); Seiler v. Lucasfilm, Ltd., 808 F.2d 1316, 1318-1320 (9th Cir.1986); Gemisys Corp. v. Phoenix American, 186 F.R.D. 551, 563-64 (N.D.Cal.1999).

The contents of Seiler’s work are at issue. There can be no proof of “substantial similarity” and thus of copyright infringement unless Seiler’s works are juxtaposed with Lucas’ and their contents compared.

Seiler, 808 F.2d at 1319.

This is not to say that the original electronic databases or printouts need be produced; it is sufficient if adequate reproductions are available to the court for comparison purposes. Data East USA Inc. v. Epyx, Inc., 862 F.2d 204, 207 (9th Cir.1988); Segrets, Inc. v. Gillman Knitwear Co., 42 F.Supp.2d 58, 80 (D.Mass.1998).

Although Berkla has reproduced various exemplars for illustrative purposes, Berkla has submitted no comprehensive item-by-item reproduction. Corel is thus partly correct in its urging that Berkla has failed to submit probative evidence with which to resist summary judgment. However, Co-rel produced the reproduction comparisons with respect to the seventeen databases which it contends are at issue. Corel’s Appendix, Exhibits 48-65, and 88. With stated exceptions in Appendix B to this opinion, this reproduction, in the record, will suffice for the court’s review of these items. Moreover, the court need not juxtapose the 79 exactly copied databases sent to the beta testers with Berkla’s original databases. By definition, the databases are virtually identical. To the extent that Berkla has not presented his other infringed databases for review, Corel is entitled to summary judgment. This includes all of the tubular text databases in that no juxtaposition is possible in the absence of designated reproductions of the allegedly infringed Berkla databases along with Corel’s infringing databases. 11

C. Copyright Infringement Standards and Application

1. The Plant Life Databases

Although Corel maintains that Berkla holds a “weak” copyright, Corel does not contest the copyrightability of Berkla’s databases, nor Berkla’s ownership of that copyright in its primary argument. 12 Therefore, Berkla meets the first well established condition to recover on a copyright claim. The second required element for a copyright infringement claim is that the defendant: “cop[ied][the] constituent elements of the work that are original.” Feist Publications v. Rural Tel. Service, 499 U.S. 340, 361, 111 S.Ct. 1282, 1296, 113 L.Ed.2d 358 (1991). In the context of this case, copying of Berkla’s pro-tectable artistic expression would satisfy the second requirement. Further, there is no doubt that “copying” would include using the copyright protected expression as a model or a template for one’s own work even if done from memory as opposed to actual or electronic tracing. 4 Nimmer on Copyright, § 13.01[B]. 13 Just as undisputed is the proposition that ideas, techniques, or processes are not protectable under copyright law — only the original expression of the idea. Sega Enterprises v. Accolade, Inc., 977 F.2d 1510, 1523 (9th Cir.1992).

Another axiom of copyright law is that copying is generally proven by showing defendant’s access to the protected work and substantial similarity between the protected and the infringing work. Sid & Marty Krofft Television v. McDonald’s Corp., 562 F.2d 1157, 1162 (9th Cir.1977). The parties, of course, do not dispute Co-rel’s access to the Berkla nature databases. But at this point, the agreement of the parties ceases. Berkla believes that with his evidence of direct copying (the 79 databases supplied to the beta testers) and the arguably strong inferential showing of Co-rel’s use of Berkla’s work as a model to improve the [expressive] quality of its own databases, he need not show evidence of substantial similarity (or virtually identical expression). In essence, Berkla believes that an attempt at copying, even so poorly or shrewdly performed such that the compared works are not substantially similar, would still result in copyright infringement liability. At the very least, citing Shaw v. Lindheim, 919 F.2d 1353, 1361 (9th Cir.1990), Berkla opines that his burden of demonstrating substantial similarity is greatly reduced.

Berkla is incorrect in the former assertion, and the latter assertion, although correct is not applicable to the facts herein. In the first instance, Berkla does not have evidence of direct copying for any of the allegedly infringing databases at issue in this section (the commercially released Co-rel nature databases). Aside from the per se potential liability for having uploaded and disseminated the 79 Berkla databases on a one time basis to its beta testers, which is discussed in subsection D below, the uploading does not prove that Corel’s commercially released nature databases were directly copied from any Berkla database. As detailed above, the databases which were uploaded never made their way into any commercially released Corel product. The uploading adds to the inferential evidence of copying/modeling the Berkla databases, but does not constitute direct evidence of copying of the Berkla databases vis-a-vis the released Corel databases.

Moreover, even if Berkla did have evidence of direct copying or modeling of the databases at issue, he still remains unexcused from demonstrating the requisite similarity. “ ‘Not all copying, however is copyright infringement’.... As Feist illustrates, copying as a factual matter is insufficient, if improper appropriation is lacking.” Nimmer, supra, at 13-9, 13-10. “[C]opying as a legal proposition must still be established; hence, substantial similarity remains an indispensable element of plaintiffs proof, even in cases (such as Feist) in which defendant does not contest factual copying.” Id. at 13-13. Thus, Berkla remains required to show substantial similarity, or as the court will explain below, virtually identical copying of Berk-la’s protected expression. 14

The difficult issues in this summary judgment motion are to separate what is protectable expression from the idea, and to determine what shapes, forms, colors and so forth follow so ineluctably from the idea as to be unprotectable.

No one in this action contends that the idea of using realistic plant, flower and other nature components for construction of a digital garden is protectable under copyright law. However, the expression of the idea — the actual rendering of plant components, follows very closely from the unprotected idea. That is, there are very few ways of depicting a realistic citrus leaf or forget-me-not or clover. The plant components themselves are relatively unchangeable, e.g., aside from the rare exception, clovers have three petals, and will not have shapes that significantly vary. Their color, although potentially multi-hued, is relatively constant within the specific species. Unlike higher forms of life, the plants and flowers, as seen in nature, do not have a multitude of common expressions. The relative size of the particular plant component at a given time of development is constant in nature. Even in areas where Berkla exercised judgment based on his artistic skill and knowledge in creation of the imagery, that judgment was based, for the most part, not on expressive decision, but rather on the essentials in creating realistic plant components. For example, in paragraphs 25 and 85 of his declaration, Berkla explained why he chose specific “axis points” for the plant components located near the stems of leaves and flowers (the growth point): “This ‘growth point’ design is essential if the images are to apply as coherent media capable of rendering complex, photo-realistic paintings of plants, foliage or nature.” Berkla Declaration at para. 85 (emphasis added). In short, when one is realistically depicting nature in isolation, the potential for differing artistic expression is limited. 15

Of course, if one departs from the goal of creating realistic plant components, and one desires to exaggerate size or color and so forth, the departure from nature is pure expression. But that is not what this case is about. It is undisputed that all involved set about to create photo-realistic nature components for use in creating photo-realistic nature structures and settings. It is also obvious that realistic components can be combined to make a very expressive whole. A field of withered corn comprised of realistic withered corn plant components could be an artist’s way of expressing futility, desperation, bareness and the like. Again, this case does not speak to the ultimate use of realistic plant components as part of a whole expressive purpose.

Also, the court does not find that expression is completely lacking in the databases at issue. As Corel’s counsel correctly responded at hearing, the particularized color highlights, the dimensionality, the sharpness of the shading, the decision of how many different perspectives to include in database array are all expressive to some degree of the artist’s varied judgments about what might look more realistic, or “better.” However, even these expressions are designed to render the plant component more realistic, although the court finds that in these specific areas there may be more than one completely uniform method in which to demonstrate the reality. As observed above, the parsing of the protected expression from the unprotected ideas is the very difficult conceptual issue here. Not all aspects of Berkla’s and Corel’s imagery fit neatly into an “idea” or “expression” category. Indeed, aspects of the work in question contain a little expression along with the essential elements of the idea, e.g., shading is an essential component of the idea of creating a realistic leaf, but the type of shading employed contains expression. A more appropriate way of viewing the problem is along a spectrum where the ends are clearly defined, but the vast majority of aspects of a work will fall between the ends.

The court’s comments and Corel’s position are fully supported by established law. “No copyright expression may be afforded ... to elements of expression that necessarily follow from the idea....” Aliotti v. Dakin, 831 F.2d 898, 901 (9th Cir.1987); see also Pasillas v. McDonald’s Corp., 927 F.2d 440, 443 (9th Cir.1991); Rachel v. Banana Republic Inc., 831 F.2d 1503, 1507 (9th Cir.1987); Atari, Inc. v. North American, Etc., 672 F.2d 607, 616 (7th Cir.1982) (similarity of expression which results from the fact that the common idea is only capable of expression in a stereotyped form will preclude a finding of actionable similarity). In a case involving creation of artificial roses, the court’s stated principle directly applies here: “Plaintiffs’ difficulty in this regard is that any two devices purporting to represent a natural prototype or archetype are likely to be similar, quite apart from any copying. Thus, a copyright on a work which bears practically a photographic likeness to the natural article, as here, is likely to prove a relatively weak copyright.” First American Artificial Flowers v. Joseph Markovits Inc., 342 F.Supp. 178, 186 (S.D.N.Y.1972). And finally:

A painter like Monet when dwelling upon impressions created by light on the facade of the Rouen Cathedral is apt to create a work which can make infringement attempts difficult. On the other hand, an artist who produces a rendition with photograph-like clarity and accuracy may be hard pressed to prove unlawful copying by another who uses the same subject matter and the same technique. [FN1 omitted] A copyright in that circumstance may be termed “weak,” ... since the expression and the subject matter converge. Sid & Marty Krofft Television Productions, Inc. v. McDonald’s Corp., 562 F.2d 1157 (9th Cir.1977).

Franklin Mint Corp. v. National Wildlife Art Exchange, Inc., 575 F.2d 62, 65 (3rd Cir.1978).

Weak copyrights such as Berkla has here, where the protectable expressions are few and the unprotectable similarities many, remain entitled to protection, but not pursuant to the traditional substantial similarity test. Only if the defendant has virtually identically copied the plaintiffs expressive work will infringement be found. “When the range of protectable and unauthorized expression is narrow, the appropriate standard for illicit copying is virtual identity.” Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1439 (9th Cir.1994) (applying the virtual identity standard); Rachel v. Banana Republic, 831 F.2d at 1507 (“ ‘indispensable expression’ of ideas may be protected only against virtually identical copying”); Frybarger v. IBM, 812 F.2d 525, 530 (9th Cir.1987); Nimmer, supra at § 13.03[B][2][b] at 13-64 (if the original aspect of a work lies in its literal expression “then only a very close similarity, verging on the identical, will suffice to constitute an infringing copy”).

The analysis now turns to the application of the virtual identity test. Does one attempt to first parse the completely unprotectable elements of the work and then apply the test to the remainder, or does one simply apply the test to the whole object? The answer, taken from the Apple Computer court, is both. 16 The Apple Computer application of the virtual identity standard is essentially the embodiment of the Ninth Circuit Sid and Marty Krofft extrinsic/intrinsic analysis, albeit with a stricter similarity analysis-that of virtual identity. The extrinsic part of the standard permits dissection of the elements of the work to separate unprotected elements and compare similarities of protected elements, as well as consideration of the manner in which the work was put together. This is in essence what the court has done already to determine the weakness of Berkla’s copyright in the first place, and hence the applicability of the virtually identical standard. Apple Computer, 35 F.3d at 1443. The intrinsic or subjective application looks at the reaction of a reasonable observer — would that observer believe that the works at issue are virtually identical when viewing them as a whole. In the factual discussion, and the attached Appendix B, the court has found that the close similarities between Berkla’s and Co-rel’s work lie in those elemental aspects of the natural objects that are stereotypical and hence unprotectable — the basic leaf or petal structure, typical, general coloring for the species, the fact of perspective and dimensionality (shading), the axis of the image about the growth points. Those elemental aspects involving some original expression such as highlighting on the leaves, emphasis on particular structure (e.g., the vein system), the precise type and form of shading, the varied perspectives, the precise illusion of dimensionality (shading) are dissimilar. When viewed in isolation, the expressive elements are dissimilar. When placed together in a component cluster, or in a database array (several clusters), the court cannot find that any reasonable juror would believe that Corel copied any database or database array to the extent that any are virtually identical. 17

In arriving at the above conclusion, one must apply a definition of virtual identity — something that is not precisely provided in the case law. Clearly, the plain meaning of the words will not allow any significant dissimilarities. But, the standard will be satisfied at some point short of photographic reproduction. However, it is difficult to pin down a definition more precise than the narrow spectrum just given, and ultimately, the decision is made on subjective observation. Therefore, the court relies on its own observations. However, in so doing at the summary judgment stage, the question is not simply what the undersigned believes, but whether reasonable minds could differ on the outcome. “A ‘genuine issue’ exists when the plaintiff provides indicia of a ‘sufficient disagreement’ concerning the [virtual identity] of the two works ‘to require submission to a jury.’ ” Brown Bag Software v. Symantec Corp., 960 F.2d at 1472. The court, exercising great hesitancy, concludes not. 18

Berkla argues vigorously that the “virtual identity” test cannot apply in cross-media situations because the use of different media will always preclude virtually identical copying by nature of the different look and feel of the media itself. The argument is not without force in the abstract; however it fails for two reasons in the present case. The premise that a cross-media environment characterizes the Berkla/Corel imagery is faulty. The media here is digital art. The fact that Berk-la hand crafts his realistic images on an electronic pad, while Corel uses electronically scanned-in photographic images, does not change the nature of the media. There are simply no significant differences in the way the imagery is presented to the observer-images on a computer screen. The situation here is much like that of a watercolor and canvas media. It does not matter whether the artist uses a brush, a feather, a finger, or a spray gun to place the watercolor on the canvas — -the media is the same — watercolor on canvas.

Moreover, as it does in same-media cases, the court when adjudicating a cross-media copyright infringement case will simply have to parse, to the best of its ability, the protected expression from the unprotected essential idea (and what “expression” necessarily flows from that idea), and make its best judgment concerning whether the potential infringing expression has sufficient dissimilarities.

For the reasons set forth above, the court finds that Corel did not infringe any of Berkla’s nature databases. 19

2. Tubular Text

As previously indicated, summary judgment should be awarded on the tubular text claims on account of Berkla’s non-submission of comparative reproductions of the art components at issue. However, a more substantive reason exists for the entry of summary judgment-there are no material issues of fact concerning access by Corel to the protectable expression, if any, at issue which would preclude summary judgment in Corel’s favor.

Access to the protected expression in a copyright case is, of course, part of the “copying” equation. Because there is often no direct proof of access, as there is often no direct proof of copying itself, the Ninth Circuit permits access to be shown by circumstantial evidence. Baxter v. MCA 812 F.2d 421, 423 (9th Cir.1987). A striking similarity between the infringed and infringing work may be sufficient to demonstrate access. Id. As Nimmer well puts the issue, it is the defendant’s “reasonable opportunity” for access that is at issue, not actual access. 4 Nimmer, at § 13.02[A]. However, “reasonable opportunity does not encompass any bare possibility in the sense that anything is possible. Access may not be inferred through mere speculation or conjecture.” Id. at 13-19.

The only circumstantial fact relating to access is that the Paint 8 product manager Chomyn may have seen one finished extrusion, including some consolidated geometric shapes, on Berkla’s website because he had been directed to the website. However, even if this access by a Corel person, who was not involved in the creation of Corel geometric shapes for its Paint 7 and 8 programs, was sufficient for circumstantial access, dispositive to the issue here is that the extrusion is not the product for which protection can be given. The extrusion is simply the manipulation of geometric shape databases with the computer tool. Just as the computer user for the natural objects can create a near infinite variety of settings using the databases and image sprayer, so too the computer operator can make a near infinite variety of tubular texts using the geometric shapes and brush settings. And while the finished product can be the subject of copyright protection, such is not what is at issue here—Berkla claims again that Corel stole his tubular text component databases and process (something which is not subject to copyright protection in any event). There is no evidence whatsoever that Co-rel ever had access to Berkla’s geometric shapes, his computer code, or his process. There is no evidence that one could have reverse engineered Berkla’s components by investigating a final extrusion. Finally, it-is undisputed that Corel had created its geometric shapes for Paint 7 before any contact with Berkla.

Berkla is reduced to arguing that because Berkla’s nature images CD received moderate dissemination within Corel “it is similarly impossible [sic] for Corel employees to have access to Tubular Text images created by Berkla without knowing that they were Berkla’s.” Plaintiffs response to Undisputed Fact 66. Berkla essentially replicates this argument concerning Co-rel’s demonstrated lack of access to Berk-la’s brush file and programming language, i.e., anything is possible. Response to Undisputed Facts 72, 73.

Although access is often easily found in copyright cases, there are limits. Berkla has not overcome those minimal limitations, and Corel is entitled to summary judgment. 20

D. The 79 Directly Copied Images

With respect to the 79 Berkla Garden Hose (natural images) files directly uploaded to Corel’s computer equipment, and later sent to its 19 beta testers, Corel concedes a technical copyright violation. 21 Corel briefly argues that since Berkla did not suffer monetary harm, and Corel did not acquire monetary gain, this is a no harm — no foul situation. Moreover, at present, there is no firm evidence that the beta testers did not destroy Corel’s pre-release Draw/Paint 8 as they had promised to do.

Berkla has eschewed statutory damages for the conceded infringement, quite possibly on account that such would not be awardable pursuant to 17 U.S.C. § 412 because of Berkla’s belated copyright registration. Opposition at 53. Berkla does not attempt to proffer any proof of actual monetary harm caused by the one time pre-commercial release distribution (later retracted before commercial distribution), but suggests that the distribution may be compensable in royalty payments. The court finds that a possibility of damages exists, de minimis as they may be:

Some courts have developed an alternative method for calculating actual damages to protect a plaintiff in situations where an infringement results in neither a profit to the infringer nor a quantifiable loss to the owner. In such cases, courts apply the “value of use” theory, which is essentially an implied license theory. See Steven Greenberg Photography v. Matt Garrett’s of Brockton, 816 F.Supp. 46, 49 (D.Mass.1992); see also 4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 14-02[A] (1995). The value of use is equal to what a willing buyer would have been reasonably required to pay to a willing seller for the plaintiffs work or, in other words, what the cost of a license for use of the work would have been. Sid & Marty Krofft Television Productions, Inc. v. McDonald’s Corp., 562 F.2d 1157, 1174 (9th Cir.1977).

Quinn v. City of Detroit, 23 F.Supp.2d 741, 750-51 (E.D.Mich.1998).

The amount of a reasonable royalty for the 79 images infringement was not fairly posited by Berkla’s summary judgment motion; therefore, a trial must be had on this issue.

BREACH OF THE NON-DISCLOSURE AGREEMENT

Whether ultimately protectable by copyright or not, Berkla considered his nature database images to be proprietary, and he was wary of simply giving them to Corel for evaluation purposes. Therefore, in the Spring of 1997, he suggested several alternatives to Chomyn to protect his images, and finally settled upon a non-disclosure agreement (NDA). Berkla did not tailor this NDA to his particular database images, but rather used a boilerplate NDA which had universal intellectual property application. The pertinent parts of the NDA (dated April 11, 1997) read as follows (italics added):

Confidential Information Recipient [Co-rel] has agreed to provide ceHain services or products to the Company [Berk-la’s DigArts company]. Recipient may have access to ... certain confidential, secret or proprietary information or materials (“Confidential Information”).... Included within the meaning of Confidential Information are matters of a technical nature (such as inventions ... computer programs, software ... secret processes ...) matters of a business nature (such as information about costs, profits....) and any other information of a similar nature not available to the public. Information shall not be deemed to be “Confidential Information” to the extent that it was (a) in the public domain at the time of the Company’s communication thereof to Recipient or subsequently enters the public domain without breach of any confidentiality obligation to the Company, or (b) already in the Recipient’s possession free of any obligation of confidence at the time of the Company’s communication thereof to Recipient.

Non-Disclosure. Recipient agrees that neither Recipient or any of its employees or agents shall disclose or use (except for the purpose of providing the services or products that Recipient has agreed ivith the Company to provide ....) Recipient shall not communicate or disclose Confidential Information to any third party. Internal access shall be limited on a “need to know’’ basis for the purpose of providing the services ... that the Recipient has agreed with the Company that it will provide. Recipient shall neither use Confidential information nor circulate it within its own organization except for the foregoing purposes ... Recipients (sic) will maintain a list of recipient personnel permitted access to the Confidential Information....

Corel Exhibit 6 (Lenburgs Declaration)

The NDA was to be interpreted in accordance with California law.

The parties do not dispute that the “services” Corel was supposed to render involved evaluation of Berkla databases for possible sale or license to Corel, or at the very least, as a possible add-on to Corel products. Berkla was told as early as May 1, 1997, that Corel was not interested in procuring his product for inclusion with Corel products.

It is undisputed that Berkla transmitted the CD with his nature database (Garden Hose 1.5) images on or about mid-April 1997 to Doug Chomyn. There is an issue of fact concerning whether Chomyn was authorized to execute the NDA; however, for purposes of summary judgment, the court infers that he was so authorized. There is also an issue of fact concerning the “newness” of the images sent to Corel. It is possible that some of the database images had been included in Berkla’s release of GardenHose 1.0 in which case the images would have been “available to the public” at the time Berkla sent the CD to Corel. It is also fairly clear that the CD contained new images that would not be released for public consumption until June (or July) 1997 with Garden Hose 1.5. Co-rel also relates several events of limited or partial disclosure of Garden Hose 1.5 information to individuals or companies prior to its commercial release, but the extent and relevance of these disclosures to the “public” are disputed.

Corel uploaded the CD when received, in part, to evaluate its compatibility with Corel Draw and Paint programs. However, the facts show that the Berkla databases were transmitted to approximately twelve Corel employees, some of whom did not even know they were viewing Berkla databases. The parties do not direct the court to any facts concerning precisely when and why each of the twelve individuals (aside from Chomyn and Gray) received access to the Berkla CD. However, the facts do support an inference that some of the Corel employees who received access to the CD were not involved in evaluating it, at least for purposes germane to Berkla’s desires, and it appears that it was quite possible that the employees viewed (and modeled) the Berkla images prior to June 1997 for purposes of developing Corel’s own software databases. For the reasons discussed below, this information is important to resolution of the NDA claim.

Finally, as observed earlier, 79 images from Berkla’s CD were included in the new Corel software which was sent to its beta testers after Garden Hose 1.5 was released. The beta testers were to report flaws and possible improvement in the new software prior to public release. By virtue of contractual obligations between Corel and its testers, all beta versions of the new Corel software were to be destroyed. However, there are some indications that not all beta versions were destroyed, i.e., a Corel newsletter indicated that several versions may have gone “astray.” It is further possible that Corel may have some Berkla images to this day on various “backup” files.

Corel does not attempt to dispute factual issues contesting whether it actually over-disseminated the Berkla CD, or whether it used the CD for development of its own software. Neither does it contest the fact that Berkla believed his CD to be covered by the terms of the NDA. Rather it argues that no breach of the NDA could have ultimately occurred because the confidential information given to it by Berkla either was not confidential by the terms of the NDA itself, or it lost such status as a matter of law prior to the release of Corel Draw/Paint 8. The dispositive issues for the NDA claim center about the meaning of “not available to the public” and “public domain,” and the timing of the “de-classifi-cation” of Berkla’s database images. Co-rel also raises an alternative copyright preemption defense.

There is not a scintilla of parol evidence that either Berkla or Chomyn considered or negotiated the language “not available to the public” or “public domain,” which in the context of the NDA, and read in isolation, could possibly be construed to have differentiated meanings. Therefore, the court must give those terms their expected meaning under California law, and attempt to harmonize any potential inconsistencies. The court finds that information that was “not available to the public” in the ordinary meaning of that phrase was information not confidential in the first place, or which lost its confidentiality by virtue of dissemination to the general public. In this construction, “not available to the public” equates with “public domain.”

Berkla argues that the latter term “public domain” has a fixed meaning in copyright law, and information given to Corel was confidential until in the public domain in a copyright sense (uncopyrightable). However, such an argument ignores the fact that the agreement explicitly was intended to encompass many items that were not copyrightable, e.g., business cost information, items for which the copyright sense of public domain would not be a logical fit. There is no evidence suggesting that the NDA’s public availability qualification was supposed to be interpreted with variously defined limitations depending on the precise nature of the information or the nature of the later filed claim. Moreover, the general phrase “and any other information of a similar nature not available to the public” is clearly intended to be a plain meaning, general qualifier of what may be considered confidential and what will not be. Even the section that sets forth what information will not be deemed to be confidential, and which includes the term “public domain,” also provides that otherwise confidential information that was in the Recipient’s possession at the time of its further transmission by the Company will also be deemed not to be confidential (regardless of whether that information was in the “public domain,” i.e., copyrightable). Borrowing the statutory canon of construction, noscitur a sociis (a word is known by the company it keeps), 22 the term “public domain” should be interpreted in the same sense as the other modifiers of confidential information — not available to the public.

Moreover, the court’s interpretation dovetails with the common law meaning of public domain under California law, as well as the Ninth Circuit interpreting California law in claims related to breach of an NDA. Under common law copyright principles, public domain meant simply that copyrightable expression had been made publicly available. Klekas v. EMI, 150 Cal.App.3d 1102, 1109, 198 Cal.Rptr. 296, 300 (1984) (unpublished work retained common law copyright until it was publicly released, and thereafter, anyone could copy it unless the expressionist had received a statutory copyright). And in Entertainment Research Group v. Genesis Creative Group, 122 F.3d 1211 (9th Cir.1997), the Ninth Circuit, in adjudicating a breach of confidence tort claim, found that publicly disclosed information could not qualify as “confidential” as a matter of law.

Likewise, in the instant case, we affirm the district court’s grant of summary judgment to Genesis and Aerostar since because ERG’s costumes were already on the market before Genesis became involved, any design or manufacturing “information” that ERG may have disclosed to Genesis was not “confidential” as a matter of law.... Thus, even though the Disclosure Agreement states that the information conveyed was “confidential,” as a matter of law, this information was actually not “confidential.”

Entertainment Research Group, 122 F.3d at 1227. 23

However, the fact that Berkla’s database nature images became publicly available no later than the commercial release of Garden Hose 1.5 (June or July 1997), does not mean that Corel was free to breach the NDA prior to that time. Indeed, the most significant potential breach of the NDA accrued when Corel allegedly over-disseminated the Berkla CD to Corel persons not actually involved in its licensing evaluation, and when Corel personnel allegedly utilized the CD images in improving Corel’s product (drawing all inferences at this time in favor of Berkla). If Berkla’s version of the facts is finally demonstrated, these actions on the part of Corel would constitute a flagrant breach of the NDA with potential, significant damages. 24 If on the other hand, the breach occurred after Berkla’s database images became publicly available, there was no breach of the NDA in using Berkla’s images in that the information was no longer “confidential information” as defined by the NDA. 25 All of these facts, and Corel’s further assertions that there were sufficient limited disclosures to the public prior to June 1997, or that Garden Hose 1.5 was simply a rehash of the publicly disseminated Garden Hose 1.0, are disputed, and must await trial.

Corel further asserts that the NDA is preempted by copyright law because the NDA claim, along with the other state claims, are simply redundant to the copyright claim. “A state law cause of action is preempted by the Copyright Act if two elements are present. First, the rights that a plaintiff asserts under state law must be ‘rights that are equivalent’ to those protected by the Copyright Act. 17 U.S.C. § 301(a); 1 Nimmer, § 1.01[B] at 1-11. Second, the work involved must fall within the ‘subject matter’ of the Copyright Act as set forth in 17 U.S.C. §§ 102 and 103. Id.” Kodadek v. MTV Networks, Inc., 152 F.3d 1209, 1212 (9th Cir.1998). “[C]laims are not preempted by the federal copyright statute so long as they ‘contain elements, such as the invasion of personal rights ... that are different in kind from copyright infringement.’ Wendt I, 1995 WL 115571 at * 1 (quoting Waits v. Frito-Lay, Inc., 978 F.2d 1093, 1100 (9th Cir.1992) (citing H.R.Rep. No. 1476, 94th Cong., 2d Sess. 132 (1976))).” Wendt v. Host Int., 125 F.3d 806, 810 (9th Cir.1997).

The court finds only part of Berkla’s NDA claim preempted. To the extent that Berkla claims that release per se of Corel’s images to the public in Corel Draw/Paint 8 was a violation of the NDA, the court cannot find that the claim is anything but an infringement claim dressed up as breach of contract. However, as noted before, the NDA has a much broader reach than merely prohibited release of protected images; it also precluded over-dissemination and use of Berkla’s images for any reason other than evaluation. The extra element, aside from infringement elements, is Corel’s promise itself not to perform such activities.

Computer programming codes generally qualify as copyrightable subject matter. Computer Assoc. v. Altai Inc., 982 F.2d 693, 702 (2d Cir.1992). Protection from breach of contract, however, is not equivalent to copyright protection because a contract claim requires an ‘extra element,’ id. at 716, that renders the claim qualitatively different from a claim for copyright infringement: a promise by the defendant. See 1 Nimmer on Copyright § 1.01[B][l][a], Tort-like copyright infringement claims, unlike breach of contract claims, do not require a promise by the defendant to refrain from using protected subject matter.

Architectronics v. Control Systems, Inc., 935 F.Supp. 425, 438 (S.D.N.Y.1996).

Again, to the extent that the Berkla databases were not publicly disseminated at the time Corel allegedly over-distributed the Berkla databases, and assertedly utilized the databases in improving its product, a breach of the NDA took place. This part of the NDA claim is not preempted. 26

BREACH OF CONFIDENCE AND UNFAIR COMPETITION CLAIMS

Plaintiffs breach of confidence and unfair competition claims do not add much to the liability equation in this case, although there may be different remedies in tort as opposed to breach of contract. Indeed, the unfair competition claim merely borrows its unfair practices from other areas of state law. However, Corel attacks these claims only insofar as Berkla’s databases were not publicly disclosed and copyright law would preempt them — the same areas of attack directed towards the NDA. As was the case with the NDA claim, the breach of confidence claims survive in part and are invalid in part.

Corel’s citation to Entertainment Research Group, supra, is on point to the discussion for these tort claims. To the extent that Berkla publicly disclosed his work by commercial sale of Garden Hose 1.5, there can be no breach of confidence for Corel’s disclosure of the work, nor can there be unfair competition.

The same preemption result obtained for the NDA also obtains for these tort claims.

‘An actionable breach of confidence will arise when an idea, whether or not pro-tectable, is offered to another in confidence, and is voluntarily received by the offeree in confidence with the understanding that it is not to be disclosed to others, and it is not to be used by the offeree for purposes beyond the limits of the confidence without the offeror’s permission. In order to prevent the unwarranted creation or extension of a monopoly and restraint on progress in art, a confidential relationship will not be created from the mere submission of an idea to another. There must exist evidence of the communication of the confidentiality of the submission or evidence from which a confidential relationship can be inferred.’

It is thus established that information need not be protectable under copyright law to be the subject of a breach of confidence action. Nevertheless, appellate decisions have uniformly required that an idea must be confidential and novel to warrant protection.

Tele-Count Engineers, Inc. v. Pacific Tel. and Tel., 168 Cal.App.3d 455, 462, 214 Cal.Rptr. 276, 279 (1985) (citations omitted).

The breach of confidence claim is not predicated on the existence of an express contract, but rather the law supplies a duty not to disclose confidential information where the parties had an understanding that the information was confidential, and that the receiving party would maintain that confidentiality. Id. at 465, 214 Cal.Rptr. at 282. 27

Similar to the NDA claim, the extra element present for a breach of confidence tort claim is the understanding of the parties that information would not be used by Corel but for the purpose of a product sale or licensing evaluation. To the extent that Corel did utilize Berkla’s databases for its own unilateral business purposes, and to the extent the information that it used was not publicly available, Corel would have breached confidence with Berkla. Again, the facts demonstrate that the parties had such a “non-use” understanding, and the facts of use inimical to that understanding must be fleshed out at trial.

Finally, the summary judgment result for the unfair competition claim, Cal. Bus. And Prof.Code § 17200 et. seq. is the same as the breach of confidence. The scope of the unfair competition law is broad, and encompasses violations found in other areas of California law, if such violations would be considered harmful to competition. See Cel-Tech Communications v. Los Angeles Cellular Tel. Co., 20 Cal.4th 163, 180, 83 Cal.Rptr.2d 548, 560, 973 P.2d 527 (1999). This claim will survive here to the same extent as the NDA and breach of confidence claims.

CONCLUSION

Summary judgment is awarded to Corel on Berkla’s copyright claims with the exception of the conceded dissemination of 79 exact duplicate Berkla files to Corel’s beta testers. Summary judgment is granted to Corel on every state law claim to the extent that the actions giving rise to such claims post-dated the commercial release of Garden Hose 1.5. Corel’s motion is denied on the state claims in all other respects.

The court will issue a separate order unsealing the moving and opposing papers in this case. The court will also issue a separate order on Coréis’ motion to sever/bifurcate issues in this case for trial.

APPENDIX A

SAMPLED NOZZLE IMAGES FROM DRAW 8 & GARDENHOSE

[[Image here]]

FLWR007.CPT (Draw 8)

[[Image here]]

[[Image here]]

[[Image here]]

APPENDIX B

The following comparisons reflect the court’s personal observations of the dissimilarities between Berkla’s and Corel’s database images. The court finds significant dissimilarities not only after dissecting elements of each nature component, but also considering the whole of the nature component and database array. These images utilize reproductions of the seventeen Corel databases that were alleged to be infringing, and the Exhibits contain a reproduction of the corresponding Berkla database.

1. Exhibit 48 Plum Blossom

Berkla’s blossoms have a more detailed center, and are a markedly different color than those of Corel’s. Berkla’s leaves are generally larger and darker, and also are of a different color than Corel’s. Corel’s shadowing is generally darker and ■ less expansive than Berkla’s. Corel’s blossoms are more “flattened” than Berkla’s. Berk-la’s database array contains 18 perspectives while Corel’s contains 12.

2. Exhibit 49 Poinsettia

Corel’s poinsettia contains yellow flowerings, and Berkla’s do not. Berkla’s poinsettia is much more dimensional, and the shadowing is very crisp and larger than Corel’s more subdued shadowing. Corel’s leaf structure is more detailed and less regular in shape than Berkla’s. Berkla’s database array contains many more perspectives than Corel’s.

3. Exhibit 50 Forget-me-not

Berkla’s flowers are a slightly different hue than Corel’s. The major dissimilarities exhibit themselves in the shadowing-Berkla’s is more separated from the flower than Corel’s, and is generally larger, but more subdued than Corel’s. Corel has included some plant leaves and a stem on one of its flowers, and its flower center is more developed. Berkla has included more perspectives than Corel.

4. Exhibit 51 Stalks of Grain

These grains are entirely dissimilar. Co-rel has developed the grain head in much more intricate detail.

5. Exhibit 52 Grasses

There are very few similarities concerning the grasses. Corel’s are more “haylike” with.more developed stalks; the grasses are of different colors. Berkla has included 10 perspectives and Corel 8 in the respective database arrays.

6. Exhibit 53 Grasses

Again, there is little similarity. Corel’s images are less wispy, and one could say-spinach like. The coloration of the grasses is different with Corel’s being a much darker green. Berkla’s grass stalks are less uniform.

7. Exhibit 54 Fall Leaves

The exemplars in this exhibit are more similar than previous exemplars. However, Corel’s leaves are less varied in color, and the shapes are more discernible. Co-rel has also used shadowing for each leaf, while Berkla’s shadowing is more subtle. Berkla’s images have more tree “debris” as well while Corel pictures only leaves.

8. Exhibit 55 — There was no Exhibit 55 in the submittal to the court. See Exhibit 83.

9. Exhibit 56 Blackberry Leaves

The leaves are not even of the same shape. Berkla’s leaves are more varied in hue with some leaves being very dark. Corel has a more detailed leaf structure, and uses a lighter green for the leaves than does Berkla. The shadowing is quite different in that Corel uses very dark shadowing surrounded by blueish shadowing; Berkla’s shadowing is almost entirely blueish, Berkla’s array contains many more perspectives than Corel’s.

10. Exhibit 57 Citrus Leaves

This comparison has been performed in the text of the opinion.

11. Exhibit 58 Poinsettia Leaves

Berkla’s leaves are much more developed, almost having a rippling texture to them. Berkla has various color gradations on his leaves, while Corel’s are generally constant in color. Berkla’s shadowing is much larger and more distinctive. Berkla has made leaf clusters while Corel has imaged individual leaves. Berkla has included more perspectives than Corel.

12. Exhibit 59 Pine Cones

Berkla’s cones appear darker and more aged; Corel’s are browner, and have more detail. The pine cones depicted appear to be from different species of trees — Corel’s are apparently taken from Douglas Fir, while Berkla’s are taken from an unknown type of pine tree. Corel’s pine cones have more varied perspectives.

13. Exhibit 60 Stones

It is difficult to tell whether the differences here stem from the quality of reproduction. However, from what is seen, Corel’s stones are more varied in shape-design and coloration. Corel’s stones have patterns within them, and .Berkla’s are fairly non-descript.

14. Exhibit 61 Trees

Berkla’s perspectives are more numerous and varied than Corel’s. Berkla has worked significant separation (spaces) into some of his trees, while Corel’s are much more broccoli-like in thickness. Berkla’s tree coloration is slightly darker than Co-rel’s.

15. Exhibit 62, 68 and 64 Tree Trunks, Limbs or Branches

It is not possible to effectively compare the tree trunks in that Berkla’s are depicted by black and white photocopies off a computer display, while Corel has a color plate image. As set forth in the summary judgment standards analysis, it was Berkla’s responsibility to provide the correct comparisons.

16. Exhibit 65 Seeds or Seed Pods

Berkla’s seeds are more detailed, and utilize shading much more than Corel’s. Berkla’s seeds have light highlights (shiny spots) as well.

17. Exhibit 88 (possibly in place of 55) Walnut Leaves

Berkla’s leaves are much more developed in texture, and individual leaves themselves appear to undulate. Corel displays individual leaves, while Berkla displays leaf clusters. Berkla’s coloring is more varied, and Corel’s has a rather constant lime green coloration. Corel has developed the leaf vein structure to a greater degree.

1 . Lucius Annaeus Seneca, Epistles, 65, 3, translated by Gummere, et al. (Loeb Classical Library) as set forth in Bartlett’s Familiar Quotations (16th ed.) at 102.

2 . There is also one database of rocks and pebbles apparently at issue as well. For the sake of efficiency, the court will often refer to all nature images, including rocks, with a generic "plant life’’ or “nature” characterization.

3 . The parties have submitted visual reproductions of various databases at issue. These databases are now part of the public record. Use of these reproductions for litigation purposes are considered fair use of copyrightable images. Religions Tech. Center v. Wollersheim, 971 F.2d 364, 367 (9th Cir.1992).

4 . Berkla’s rendition (Garden Hose 2) and his compilation rendition appears on the right. Corel’s images are "Draw 8” and the compilation rendition appears on the left.

5 . Berkla does not dispute these facts, but has supplemented the agreed upon basic facts. The supplementation is not necessary for purposes of this general factual background.

6 .The precise details concerning the initiation of Berkla’s business relationship with Corel are not important to the result here. In summary, Berkla, just prior to his attempts to have Corel review his Garden Hose images, had evidently been approved for "premier status” in Corel Solution Partners. This arrangement gave Berkla some non-public benefits with Corel products as well as enhanced visibility to Corel customers of certain Berkla products or applications. Corel hoped to receive in turn some third-party technical input on Corel products.

7 . Q. Now she said, “Let's try to get this quality.” Did you do anything to try to get that quality?

A. "Yeah. I grabbed better quality images off the gallery and tried cropping them better.”

8 . Beta testers assist software manufacturers in working out glitches or suggesting improvements in the software before it is released to the public.

9 . Appendix A to this opinion contains a breakdown of each of Berkla’s databases and Co-rel's allegedly infringing databases.

10 . The seventeen Corel databases are described by title in Undisputed Fact 19, and a comparison is made between each of these databases and Berkla’s corresponding databases in Undisputed Facts 22-45 inclusive. Actual reproductions of the databases at issue are found in Corel’s Appendix, Exhibits 48-65, and 83 inclusive.

11 .Corel’s exhibits do contain various and sundry pictures of its, balls and spheres (some of which are so reduced in size that they are of no comparative use), Corel's Appendix, Exhibit 34, and Exhibit 86 contains various pictures of Berlda’s finished tubular text images and a few, difficult to discern, geometric shapes. The parties do not give the court any • designation or means by which to compare the protectable elements of Berkla's geometric shapes with those of Corel’s. Since Berk-la bears the burden of production, he will bear the result of non-production.

12 . Corel argues as an alternative that because Berkla infringed Fractal’s copyright, his works are not original thus invalidating Berk-la’s copyrights. This issue is not amenable to summary judgment because of factual issues, and the court will not discuss it further.

13 . All Nimmer citations by the undersigned are to that work authored by Melville Nimmer and'David Nimmer.

14 . Berkla’s cited case of Norse v. Henry Holt & Co., 991 F.2d 563, 566 (9th Cir.1993) is distinguishable from the present situation in that the protected expression had been actually copied, i.e., there was more than virtual identity when the printed and protected words had been precisely appropriated. Moreover, Norse did not involve a situation where the expression of an idea followed necessarily from the idea itself.

15 . To be sure, there are poorer ways of depicting nature, but if the goal is to depict realism in nature, Berkla's methodology is one of the best, and according to Berkla, the only, way to do it. Although it takes skill and judgment to recognize this particular artistic axiom regarding growth points, this skill relates to a functional aspect of the imagery and not an expressive one. Berkla cannot attempt to monopolize an ''essential” design feature by virtue of copyright law.

16 . "Having found that the similarities in Windows 2.03 and 3.0 consist only of unprotecta-ble or licensed elements, and that the similarities between protectable elements in Apple’s works and NewWave are de minimis, the district court did not err by concluding that, to the extent there is creative expression left in how the works are put together, as a whole they can receive only limited protection. When the range of protectable and unauthorized expression is narrow, the appropriate standard for illicit copying is virtual identity. For these reasons, the GUIs in Windows 2.03, 3.0 and NewWave cannot be compared for substantial similarity with the Macintosh interface as a whole. Instead, as the district court held, the works must be compared for virtual identity.” Apple Computer, 35 F.3d at 1439 (footnotes omitted).

17 . Indeed, even Berkla cannot opine that any of the databases are virtually identical. The best he can say on this issue is that the databases are substantially similar. Undisputed Fact and Response 20.

18 . Judges are often called upon to determine hypothetical applications of facts in terms of whether any reasonable juror could have come to a different conclusion. For example, in habeas corpus actions involving an issue of juror misconduct, the court must determine in the abstract whether extrinsic juror misconduct would have affected the vote of any juror (and the court is not permitted to directly ask the former juror(s) about this determination). The situation here is analogous.

19 .Corel, in the alternative, claims that Berk-la's databases are not original in the copyright sense because he infringed Fractal's database images. Such a claim is insufficiently supported, even given the slight burden to Corel on summary judgment, and in any event, is rife with issues of fact not appropriately decided on this motion for summary judgment.

20 . Corel also makes arguments about the non-copyrightability and non-infringement of the Tubular Text database images. The court need not address these arguments as it has found no comparison possible between Berkla and Corel geometric shapes on the present record.

21 . Because Corel was permitted to upload the 79 images for licensing evaluation, the copy right violation occurred when the uploaded images were sent to the beta testers.

22 . Gustafson v. Alloyd Co., 513 U.S. 561, 575, 115 S.Ct. 1061, 1069, 131 L.Ed.2d 1 (1995).

23 .The Entertainment Research Group case reads with some difficulty in that the Ninth Circuit initially referenced the NDA in that case, as if the contract between the parties would be controlling, but thereafter adjudicated a tort claim. This court does not read Entertainment Research Group as precluding a valid contract prohibiting the disclosure of information that had already been publicly disseminated. If, in this case, Berlda had specifically referenced his CD as non-disclosa-ble, and Corel had promised not to disclose that CD regardless of whether Berkla would disclose it in other places, this court sees no reason why such a contractual promise could not be enforced. However, the NDA at bar did not specifically reference the CD, and it affirmatively provided that publicly available information was not confidential.

24 . The court does not find that the early, alleged breach was excused by the later commercial release of Garden Hose 1.5, nor does it find that damages would be cut off at that time. The court need not discuss here the measure of damages that would flow from Corel's pre-Garden Hose 1.5 release breach if liability were ultimately to be found.

25 . It is therefore highly likely that the dissemination of the Berkla images to the beta testers came at a time when the Berkla CD was no longer deemed confidential. In any event, to use the vernacular, any breach of the NDA occasioned by the simple release of Berkla images to to beta testers (along with the other Corel images and programs), is small potatoes compared to the potential breach described in the text.

26 . Berkla also contends that the uploading of his CD itself was a violation of the NDA. Not so-nothing in the NDA precluded Corel's evaluation of Berkla’s databases in terms of compatibility with Corel systems, and Berkla was placed on express notice by Chomyn that he desired to upload the databases for this purpose. The court finds as a matter of law that the uploading itself was not a violation of the NDA.'

27 . It may be that the breach of confidence claim cannot co-exist with a breach of contract claim encompassing the same obligations, but Corel does not attack the tort claims on this basis.