The Registered Design was applied for on 7 D ecember 1994, and that is its date of registration. It is important to note that the application was made, and the Registered Design was registered, under t he R egistered Des igns Act 19 49 as the n am ended, i n p articular by th e Copyright Act 1956 and the Copyright, Designs and Patents Act 1988 ("the pre-implementation 1949 Act"). This was well before the adoption of European Parliament and Council Directive 98/71/EC of 13 October 1998 on the legal protection of designs ("the Designs Directive"), let alone its implementation.

The Registered Design is regist ered in respect of a "vac uum cleaner". It incl udes a statement of novelty pursuant to rule 15(1) of the Registered Designs Rules 1989 in the following terms:

As is well known, Dyson was the first to introduce two-stage cyclone dust-separation technology for vacuum cleaners as an alternative to the use of porous bags. The first machine to be marketed by Dyson in the United Kingdom using this technology was an upright cleaner known as the DC01 (DC standing for Dual Cyclone). This was launched in 1993. It was very successful. The second machine was a cylinder cleaner called the DC02. This was launched in 1995. It is common ground that in general terms the Registered Design protects the design of the DC02, but there are c ertain differences between the Registered Design and the design of the DC02, particularly so far as the rear view is concerned.

The DC02 is regarded as a classic design. It received numerous accolades and awards as a result of its novel and striking design. It has been exhibited at a number of museums including the Victoria and Albert Museum and the Design Museum in London. Its impact has extended beyond the field of vacuum cleaners.

The Designs Directive includes the following recitals:

Whereas the assessment as to whether a design has individual character should be based on whether the overall impression produced on an inform ed user viewing the design clearly differs from that produced on him by the existing design corpus, taking into consideration the nature of the product to which the design is applied or in which it is incorporated, and in particular the industrial sector to which it belongs and the degree of freedom of the designer in developing the design;

Where as tec hnological in novation s hould n ot be ham pered b y granti ng de sign protection to features dictated solely by a technical function; whereas it is understood that this does not entail that a design must have an aesthetic quality; whereas, likewise, the interoperability of products of different makes should not be hindered by extending protection to the design of mechanical fittings; whereas features of a design which are excluded from protection for these reasons should not be taken into consideration for the purpose of assessing whether other features of the design fulfil the requirements for protection; ".

It includes the following operative provisions:

'design' means the appearance of the whole or a part of a prod uct resulting from the fe atures of, in particular, the lines, contours, colours, shape, texture and/or materials of the product itself and/or its ornamentation;

'product' means any industrial or handicraft item, including inter alia parts intended to be assembled into a complex product, packaging, get-up, graphic symbols and typographic typefaces, but excluding computer programs;

A design shall be considered to have individual character if the overall impression it produces on the informed user differs fro m the overall impression produced on such a user by any design which has been made available to the public before the date of filing of the application for registration or, if priority is claimed, the date of priority.

In assessing individual character, the degree of freedom of the designer in developing the design shall be taken into consideration.

The scope of the protection conferred by a design right shall include any design which does not produce on the informed user a different overall impression.

In assessing the scope of protection, the degree of freedom of th e designer in developing his design shall be taken into consideration. "

Section 1(1) of the pre-implementation 1949 Act provided:

"Shape" and "configuration" have long been regarded in UK designs law as cl ose to synonymous, and as denoting the form in which in an article is fashioned in three dimensions, while " pattern" an d " ornament" hav e generally b een regar ded as denoting tw o-dimensional surface decoration. In Mackie Designs Inc v Behringer Specialised Studio Equipment (UK) Ltd

There was no requirement for a statement of novelty in the 1949 Act itself either as originally enacted or as amended down to 1994. The requirement was contained in rule 14(2) of the Designs Rules 1949, which was replaced by rule 15(1) of the 1989 Rules. Both rules were made under the general rule-making power conferred by section 36 of the 1949 Act. Rule 15(1) provided:

The effect of a statement of novelty was stated by Slade LJ giving the judgment of the Court of Appeal in Sommer Allibert (UK) Ltd v Flair Plastics Ltd[1987] RPC 599 at 619-620 to be as follows:

After quoting this dictum, the learned editor of Russell-Clarke and Howe on Industrial Designs (7^th ed) comments at §3-285:

It is common ground that the effect of the transitional provisions in the 2001 Regulations is that the issue of infringement of the Registered Design must be determined by reference to the post-implementation 1949 Act, even though its validity, were it in issue, would have to be determined by reference to the pre- implementation 1949 Act. As the editor of Russell-Clarke and Howe notes a t § 3-07, th e re quirement th at i nfringement be governed by t he post-implementation law appears to be required by the Designs Directive by implication from Articles 11 (8) and 1 2(2); but as he c omments at §3 -03 the application of d ifferent laws to infringement and validity is capable of having startling and unfortunate consequences.

There is no provision for a statement of novelty in the post-implementation 1949 Act or in the Registered Designs Rules 2006, although rule 6 of the 2006 Rules does provide for optional partial disclaimers to be made by an applicant for registration.

In Case T- 9/07 Grupo Promer Mon Graphic SA v Office for Harmonisation in the Internal Market [2010] ECR II-0000 the General Court of the European Union held at [62]:

In Case T-1 53/08 Shenzhen Taiden v Office for Harmonisation in the Internal Market [2010] ECR II-0000 the General Court held:

The qualifier 'informed' suggests in addition that, without being a designer or a technical expert, the user knows the various designs which exist in the sector concerned, possesses a certain degree of knowledge with regard to the features which those designs normally include, and, as a result of his interest in the products concerned, shows a relatively high degree of attention when he uses them.

However, contrary to what the applicant claims, that factor does not imply that t he i nformed user is able to disti nguish, beyond the experience gained by usi ng the product concerned, the aspects of t he appearance of the product which are dictated by the product's technical function from those which are arbitrary. "

Recital 13 of the Designs Directive (corresponding to recital 14 of the Community Designs Regulation) makes it clear that the overall impression produced on the informed user depends on "the existing design corpus", taking into consideration the nature of the product to which the design is applied, and the industrial sector to which it belongs.

In Grupo Promer the Community design was reg istered for "prom otional items for games". The General Court held at [56]:

Article 7( 1) of t he D esigns D irective e xcludes fr om prot ection " features of appearance... which are solely dictated by its technical function". The parties were divided as to the correct interpretation of this provision.

Counsel for Dyson relied upon the decision of the Court of Appeal in Landor & Hawa International Ltd v Azure Ltd[2006] EWCA Civ 1285, [2007] FSR 9. That case concerned a claim for infringement of unregistered Community design right under the Community Designs Regulation (and a parallel claim for infri ngement of domestic unregistered design right) in certain features of a design for a suitcase. The defendant contended, inter alia, that those features were excluded from protection by Article 8(1) of the Community Designs Regulation, which corresponds to Article 7(1) of the Designs Directive.

At first i nstance His H onour Judge Fysh QC sitting in the Patents County Court rejected this contention for reasons he expressed as follows:

In the Court of Appeal the defendant attacked this conclusion on the grounds that (i) the judge had been wrong to construe Article 8(1) so narrowly and (ii) the features in question were purely functional.

So far as the construction of Article 8(1) was concerned, the defendant argued that the judge ought to have held that "dictated solely by function" meant "attributable to function", so that if more than one shape could perform the same function then all such shapes were excluded from protection. In support of this argument the defendant relied upon the decision of the Court of Justice of the European Communities (as it then was) in Case C-299/99 Koninklijke Philips Electronics NV v Remington Consumer Products Ltd [2002] ECR I-5475 at [78]-[79] interpreting

Neuberger LJ (as he then was), with whom May and Wilson LJJ agreed, held that neither of these decisions could safely be reli ed on i n order to interpret Article 8(1) and noted the comment of Advocate General Colomer in his opinion in the former case at [34] that the wording used in Article 7(1) of the Designs Directive was narrower than that in Article 3(l)(e) of the Trade Marks Directive. Neuberger LJ also said at [38]:

Counsel for Dyson submitted, and I acc ept, that it is clear t hat the Court of Appeal endorsed the judge's construction of Article 8(1). Counsel for Vax, however, drew attention to what Neuberger LJ went on to say (emphasis added):

For his part, counsel for Vax relied on the decision of the Third Board of Appeal of OHIM in Case R 690/2007-3 Lindner Recyclingtech GmBH v Franssons Verkstader AB[2010] ECDR 1. In that case the Board, whose rapporteur was David Keeling, considered the interpretation of Article 8(1) of the Community Designs Regulation in a passage which merits quotation in full despite its length:

The multiplicity-of-forms theory has been adopted by courts in the United Kingdom (see th e ju dgment o f Ju ly 28 , 2006 of th e Court of Ap peal in Landor & Hawa International Ltd v Azure Designs Ltd[2006] EWCA Civ 1285; [2006] E. C. D. R. 31) and Spain (Juzgado de lo Mercantil PTO Numero Uno de Alicante, Auto No. 267/07 Silverlit Toys Manufactory Ltd v Ditro Ocio 2000 SL Unreported, November 20, 2007.

There is none the less a major flaw in the m ultiplicity-of-forms theory. If it is accepted that a feature of a product's appearance is not 'solely dictated by its function' simply because an alternative product configuration could achieve the same function, art. 8(l) CDR will apply only in highly exceptional circumstances and its very purpose will be in danger of being frustrated. That purpose, as was noted above, is to prevent design l aw fro m being used to achieve monopolies over t echnical solu tions, t he assumption being that s uch m onopolies are only j ustified if t he more restricti ve conditions imposed by patent law (and in some countries by the law of utility models) are complied with. If a technical solution can be achieved by two alternative methods, neither solution is, according to the multiplicity-of-forms theory, solely dictated by the function of the product in question. This would mean that both solutions could be the subject of a design registration, possibly held by the same person, which would have the consequence that no one else would be able to manufacture a competing product capable of perform ing th e sam e technical f unction (see W . Corn ish a nd D . Llew elyn, Intellectual Property: Patents, Copyright Trade Marks and Allied Rights, 5th edn,

T he princi pal alt ernative, discussed by academic authors, to the multiplicity-of-forms theory has its origin in English case law. The case of Amp Inc v Utilux Pty Ltd [1971] F. S. R. 572 concerned the interpretation of a provision of the Registered Designs Act 1949 which denied protection to the features of a desi gn that were solely dictated by a product's technical function. The House of Lords held that a product's configuration was solely dictated by its technical function if every feature of the design was determined by technical considerations. The striking similarity between s.1(3) of the 1949 Act and art. 8(1) CDR does not of course mean that the approach of the House of Lords in Amp Inc v Utilux Pty Ltd must necessarily be adopted in relation to the Community provision. Indeed, as was noted above in [29], the multiplicity - of-forms theory has now been adopted by the English Court of Appeal in Landor & Hawa International Ltd v Azure Designs Ltd. Thus the Court of Appeal must have thought th at the a pproach ta ken in Amp Inc v Utilux Pty Ltd w as no l onger valid, following harmonisation, in spite of the similar wording of the Community provisions and the 1949 Act. The approach taken in Amp Inc v Utilux Pty Ltd would, however, have the advantage of allowing the purpose of art. 8(1) CDR to be achieved. No-one would be able t o s hut out c ompetitors by regis tering as Community d esigns th e handful of possible configurations that would allow the technical function to be realised. This may explain why the French courts, which formerly espous ed the m ultiplicity-of-forms theory, began to abandon that theory at the beginning of the 21st century in favour of an interpretation which closely resembles the Amp Inc v Utilux Pty Ltd approach (see the judgments cit ed by D . Co hen, Le droit des des sins et modéles, 2n d ed n, (Paris:

In addition t o b eing supported by a tel eological in terpretation, the approach discussed in the previous paragraph is also supported by the wording of art. 8(1) CDR. That provision denies protection to features of a product's appearance that are, 'solely dictated by its technical function'. Those words do not, on their natural meaning, imply that the feature in q uestion must be the only means by which t he product's technical function can be achieved. On the contrary, they imply that t he need to achieve the product's technical function was the only relevant factor when the feature in question was selected.

G ood design involves t wo fund amental e lements: the product m ust perform its function and it should be pleasant t o look at. In the case of some products, such as pictures and ornaments, their very function is to please the eye. In the case of other products, such as the internal working parts of a machine, the visual appearance is irrelevant. That i s wh y the Co mmunity d esign l egislation deni es p rotection to component parts that are not visible in normal use. In the case of most products the designer will be concerned with both the functional and the aesthetic elements. That applies also to large items of industrial equipment, such as shredders for use in recycling plants. The shredder must, in the first place, perform its function effectively and safely and without creating excessive noise, but it is also desirable that the shredder should be pleasing to the eye and thus enhance the working environment of the people who operate it and see it in use. For that reason there is no objection in principle to granting design protection to industrial products whose overall appearance is determined largely, but not exclusively, by functional considerations.

It is often pointed out that the Community design legislation, unlike the old laws of some Member States, does not lay down any requirement of aest hetic merit, artistic creativity or eye appeal. The absence of such a requirement is expressly mentioned in the 10th recital in the preamble of Regulation 6/2002 and in the 14th recital in the preamble to Directive 98/71. Some authors infer from this that purely functional designs are protectable. That is a false analysis. Community design law is concerned with the visual appearance of products. That is clear from the definition of 'design' in art. 3(a) CDR and from the requirement of visibility in normal use for component parts in art.

The significance of limiting protection to the visual appearance of products is that aesthetic c onsiderations are in pr inciple cap able of being re levant on ly w hen the designer is developing a product's visual appearance. Most of the time the designer will be concerned with both elements of good design: functionality and eye appeal. In some cases functionality will be the dominant preoccupation of the designer. The need to make a pr oduct that works will be uppermost in t he designer's mind and will largely determine the appearance of the product. As long as functionality is not the only relevant factor, the d esign is i n principle el igible for pro tection. I t is only when aesthetic considerations are com pletely irrelev ant th at the fe atures of the design are sol ely dictated by the need to achieve a tec hnical solution. This is no t, it must be stressed, tantamount to introducing a requirement of aesthetic merit into the legislation. It is simply recognition of the obvious fact that when aesthetics are totally irrelevant, in the sense that no one cares whether the product looks good, bad, ugly or pretty, and all that matters is that the product functions well, there is nothing to protect under the law of designs.

It follows from the above that art. 8(1) CDR denies protection to those features of a product's appearance t hat were ch osen exclusively for th e p urpose of d esigning a product that performs its function, as opposed to features that were chosen, at least to some degree, for the purpose of en hancing the product's visual a ppearance. It goes without saying that these matters must be assessed objectively: it is not necessary to determine what ac tually went on in the designer's mind when the design was bei ng developed. The matter must be assessed from the standpoint of a reasonable observer who looks at the design and asks himself whether anything other than purely functional considerations could have been relevant when a specific feature was chosen. "

Article 9(2) of the Designs Directive provides that the degree of freedom of the designer is to be taken into account in assessing the scope of protection conferred by a registered design.

In Grupo Promer the General Court held:

At paragraph 18 of t he contested decision, the Board of Appeal stated t hat al l the ' rappers' or ' tazos' exa mined i n t he prese nt c ase consisted of small, flat or slightly curved discs which may be made of plastic or m etal. A ccordingly, i t c oncluded, at p aragraph 20 of t hat decision, that th e free dom en joyed by the des igner responsible f or designing a product of that kind was severely restricted, because, for that type of product, ' [t]he paradigm... is a small flat or nearly flat disk on which coloured images can be printed [and o]ften the disk [is] curved toward[s] the centre, so that a noise [is] made if a child's finger presses the c entre of t he dis k', an d ' [a] ra pper that does not p ossess these characteristics is unlikely to be accepted in the marketplace'.

In that connection, it must be noted that 'pogs', 'rappers' or 'tazos' are circular in shape and that, on the date of filing of the application for registration of the contested design, in this case on the date of priority claimed for the design, 'pogs', 'rappers' or ' tazos' had those common features which the designer had to take into account, as set out at paragraphs 18 and 20 of the contested decision and reiterated at paragraph 68 above. That finding is not, moreover, contested by the parties.

Therefore, it must be held that the Board of Appeal was correct to find in the contested decision that, on the date of priority claimed for the contested design, the designer's freedom was severely restricted since he had to incorporate those common features in his design for the product in question. Moreover, as the Board of Appeal pointed out at paragraph 20 of the contested decision, the designer's freedom was also limited is so far as th ose items had t o be inexpensive, safe for children and fit to be added to the products which they promote. "

Counsel f or Vax subm itted, a nd I ac cept, t hat t his p assage indicates tha t d esign freedom may be constrained by (i) the technical function of the product or an element thereof, (ii) t he n eed t o incorporate fe atures common to s uch pr oducts an d/or (i ii) ec onomic considerations (e. g. the need for the item to be inexpensive).

In Shenzhen the General Court held:

In order to challenge the validity of that conclusion, the applicant puts forward, first, the fact that many features of a conference unit as well as their configuration are dictated by the technical function of the device and, second, the existence of a general trend favouring small, flat, rectangular devices, often including hinged elements.

With regard to the first factor, it is admittedly true that, in order to fulfil its essential function, a conference unit must have, at the very least, a speaker and a microphone, directed in such a way that the user can hear the sound reproduced by the speaker and his speech can be captured by the microphone. Similarly, buttons which are accessible to t he user are ne cessary, in particular, in or der t o tur n the microphone on a nd t o regu late th e vol ume of the spe aker. Furthermore, to the extent that conference units also have associated functions, features such as voting buttons, the screen and the card slot may also prove necessary from a functional point of view.

H owever, as O HIM a nd the intervener have co ntended, t hose restrictions concern the presence of certain features in the conference unit, but do not have a s ignificant impact on their configuration and, therefore, on the form and general a ppearance of the c onference unit itself. In parti cular, it does not seem that a hinged e lement would be necessary in order to ensure any of the device's functionality.

That conclusion is borne out by the design corpus, as submitted by the intervener to O HIM, which s hows confer ence units of varying shapes and configurations that differ perceptibly from those used in the contested design. Therefore, depending on the model, the microphone is placed on a stem, or not, on the left, or the right or in the middle of the body of the device. In the same way, if the card slot is normally placed on t he ri ght, it is generally not integrated into t he speaker of t he conference unit but in the actual body of the device . In addit ion, the presence of any hinged element is the exception rather than the rule. "

Counsel for Dyson submitted, and I accept, that this passage demonstrated that the designer's degree of freedom could be tested by reference to the existing design corpus. As Jacob LJ put it in Procter & Gamble Co v Reckitt Benckiser (UK) Ltd [2007] EWCA Civ 936, [2008] FSR 8 at [57]:

Counsel for Dyson also submitted, and I accept, that evidence of design freedom could also come from designs produced after the date of the registered design. If a w ide variety of designs w as produced aft er the reg istered d esign, t hat is e vidence tha t t he de signer of the registered design had not been constrained to design the product in the way that he had.

The p arties we re divided as t o th e co rrect date a s at the ov erall imp ression of the allegedly infringing design should be compared with the overall impression of th e registered design.

Counsel for Vax contended that the comparison should be made as at the date of the alleged infringement. He subm itted that this fo llowed fro m the fact th at Article 9(1) of th e Designs Directive was expressed in the present tense and from the fact that design freedom could change over time. He accepted that this meant that the scope of protection of the registered design could change over time and could either diminish or increase, but argued that there was no reason why should this not be the case. He also submitted that, if Article 9(1) were interpreted as requiring the comparison to be made as at the date of registration of the registered design, then there would be a conflict with Article 5(1).

Counsel for Dyson contended that the comparison should be made as at th e date of registration of the registered design for four reasons. First, he submitted that it was immaterial that Article 9(1) was expressed in the present tense, since Article 9(2) must be referring to the degree of freedom of the designer of the registered design which pointed to a comparison as at that date. Secondly, he pointed out that recital (13) referred to the existing design corpus, which must be the design corpus which existed at the date of registration, and that recital (13) had been treated in the authorities as relevant to infringement as well as validity. Thirdly, he submitted that otherwise the scope of protection could be eroded by subsequent designs which adopted the striking features of the registered design one by one. Fourthly, he submitted that the comparison with Article 5(1) supported this conclusion, since if the scope of the monopoly could get broader

In my judgment the reasons given by counsel for Dyson for making the comparison as at the date of the registered design, and having regard to the existing design corpus as at that date, are convincing.

It is com mon grou nd th at, alt hough i t i s prop er to consider b oth similarities and differences between the respective machines, what matters is the overall impression produced on the informed user by each design having regard to the design corpus and the degree of freedom of the designer. In this regard both counsel referred me to the observations of Mann J in Rolawn Ltd v Turfmech Ltd Machinery Ltd[2008] EWHC 989 (Pat), [2008] RPC 27:

During the course of argument there was some debate as to whether questions should be referred to the ECJ as to the interpretation of the the Designs Directive. Although I consider that some of t he issues considered above involve points of interpretation of the Designs Directive which are not acte clair and which are likely to require resolution by the ECJ at some point, I have come to the conclusion that a reference is not necessary for the resolution of the present case because those issues are unlikely to be determinative.

Counsel for Vax submitted that the effect of the statement of novelty in the Registered Design was to exclude anything other than the "shape" and "configuration" of the article from the scope of the monopoly, and hence to exclude other aspects of the design of the article shown in the representations, including the fact that the dust collection bin is transparent. He nevertheless argued that the design corpus was to be assessed on the basis that the article shown in the representations was a cyclonic vacuum cleaner, and accepted that the only way in which the informed user could know that it was a cyclonic vacuum cleaner was because the user could see the shroud of the outer cyclone within the bin in the representations as a result of the bin being transparent. He sought to escape from the apparent paradox to which these submissions lead, that one must disregard the transparency of the bin and yet one can only see what is within the bin because it is transparent, by arguing that the court should not notionally treat the bin as opaque, but rath er sh ould i gnore t he tr ansparency of the bin when assessing the overal l impression produced by the Registered Design and comparing it to the overall impression produced by the Mach Zen.

Counsel for Dyson accepted that the fact that the bin was transparent was not in itself a feature of shape or configuration, but disputed that the statement of novelty required the court to ignore the transparency of the bin. He submitted that the transparency of the bin nevertheless affected the shape or configuration of the article depicted in the representations. Furthermore, he agreed with counsel for Vax that the informed user would know that the vacuum cleaner was cyclonic and that the only way in which the user could know this was because the user would see the cyclone within the bin. He disputed that the transparency of the bin should be ignored when assessing the overall impression produced by the Registered Design.

I have not found this issue easy to resolve. Speaking for myself, I find it difficult to see how the scope of protection of the Registered Design under the Designs Directive can be affected by an artefa ct of the pr e-implementation d omestic la w. Co unsel f or D yson did no t ar gue, however, that the statement of novelty should simply be disregarded. I must therefore proceed on the basis that it is to be taken into account.

That being so, I ha ve come to the conclusion that the solution to the conundrum is provided by the comment of the editor of Russell-Clarke and Howe quoted above. If one leaves the statement of novelty on one side, the transparency of the bin is plainly a feature of the design depicted in the representations which contribut es to its eye appea l. I do not consider that the statement of novelty requires this feature to be ignored even if transparency is "pattern" or "ornament" rather than "shape" or "configuration". (The m eans by whic h the transparency is achieved - such as the use of a transparent plastic - is clearly none of these, but in my view that is a separate point. ) The fact that it is common ground that the informed user would know that the article depicted in the representations was a cyclonic vacuum cleaner because the informed user would see the cyclone within the bin supports this conclusion.

The parties were sharply divided as to the relevant existing design corpus in this case. Dyson contends that it consists of cylinder vacuum cleaners. This is narrower than the indication in the registration and takes into account the nature of the product shown in the representations. By contrast, Vax contends that there was no existing design corpus at all in December 1994. This is on the basis that the relevant product category is cyclonic cylinder vacuum cleaners, and the first product to be marketed in that category was the DC02. Indeed, at times counsel for Vax appeared to be contending that the relevant product category was two-stage cyclonic cylinder vacuum cleaners.

I have to say that, from a forensic perspective, I find Vax's contention a surprising one, since it would apparently lead to the Registered Design having a broader scope of protection than if the design corpus were as Dyson contends. Be that as it may, I consider that Dyson is correct on this point. Although it is common ground that the informed user w ould appreciate the article depicted in the representations was a cyclonic cylinder vacuum cleaner, I consider that the informed user would re gard it as sim ply a cy linder vacuum cleaner with a partic ular dust separation te chnology. O n the e vidence, there is n ot a reco gnised separate su b-category of cyclonic cylinder cleaners even now. Instead, the distinction that is commonly made is between bagged and bagless cylinder cleaners.

The evidence demonstrates that in 1994 there were two main types of cylinder vacuum cleaner, described by Mr Gammack as "sledge" layout machines and "tank vacs". The former were shaped like a lozenge or bar of soap. The latter were barrell- shaped machines. The "sledge" layout machines were quite similar to each other in terms of their overall appearance. Differences between the registered design and the existing design corpus

Dyson con tends tha t th e R egistered D esign is striki ngly differen t both i n ov erall appearance and in numerous detailed aspects of t he design to the existing design corpus of cylinder vacuum clea ners in D ecember 19 94. In my ju dgment t he e vidence supports that contention.

Dyson contends that the evidence shows that the designer of a cylinder vacuum cleaner has a great deal of design freedom. Dyson says this remains so if one restricts one's attention to bagless cyclinder cleaners, and even if one restricts one's attention to cyclonic cylinder cleaners. In support of these contentions Dyson points to a series of cylinder vacuum cleaners that have been marketed since December 1994 that vary widely in their appearance from the Registered Design. In general, I accept that the evidence shows that a considerable variety of designs has been marketed.

That is not the end of this aspect of the matter, however, because Vax contends that the marketing of a variety of designs does not disprove the existence of relevant design constraints. Counsel for Vax submitted that the exercise undertaken by Dyson of collecting cylinder cleaners of different appearance was of little probative value for two main reasons. The first was that the degree of design freedom depended on the technical specification of each cleaner, particularly in terms of its dust separation technology, size, weight, electrical power consumption and suction power. As counsel for Vax put it, the freedom of the designer referred to in Article 9(2) is not the freedom to de velop a technically inferior product. The second was that the degree of des ign freedom depended on other intellectual property rights, and in particular patents, owned not only by the parties but also third parties.

I accept that, in general, the degree of freedom of the desi gner must be j udged by

I do not accept that the existence of other intellectual property rights is a relevant consideration, or at least not unless it is specifically proved that particular rights existed which imposed particular constraints on the freedom of the designer at the relevant time. In the present case Vax did not prove that any particular rights existed which imposed particular constraints at any relevant date.

Although stressing that overall impression is what matters, Dyson has identified nine respects in which it contends that the Mach Zen is similar to the Registered Design while the Registered Design differs from the existing design corpus. I shall consider these in turn.

A bin inclined at substantially halfway between vertical and horizontal. It was common ground between the witnesses that it is possible to orient a multi-cyclone separator at any angle from vertical to horizontal. There was also a considerable measure of agreement, however, that there are some technical considerations that favour a vertical orientation and others that favour a horizontal orientation, and that the trade-off between them favours an angle of around 45°. To

Counsel for Vax accepted that, at this level of generality, the Mach Zen was similar to the Re gistered D esign w hile t his feat ure w as n ot present in t he existing design corpus. H e submitted, however, that, given the restricted degree of freedom of the designer of the Mach Zen, the similarity in itself was not of great significance. I accept this submission.

Furthermore, he s ubmitted that in any event the informed user wo uld consider the design of the bin as a whole and notice a number of differences between the Mach Zen and the Registered Design in this respect. I agree with this. The principal differences are as follows:

The Dyson bin is a sim ple cyli nder s hape. The bas e is l argely i nvisible in the Registered Desi gn si nce i t i s hi dden by the bumper. Because of the shape of the housing surrounding the bin, the cylinder appears to flare out towards to the top. By contrast, the Mach Zen bin is "rocket-shaped", with three sections which step down in diameter towards the top. The larger diameter section flares out towards the bottom. It has a very visible frusto-conical base.

The handle on the Dy son is a sim ple tria ngle sh ape with a s lightly rou nded hypotenuse, forming a narrow angle where it meets the bin towards the bottom. It is a simple piec e of cl ear, flat plastic. T he handle on t he Mach Ze n is a m uch more significant design feature. It runs parallel with the wall of the bin and is formed from a semi-circular section length of plastic, concave where the thumb would sit. There is a grip area at the top, designed for fingers to pass through. The bottom area is filled in with plastic.

The Mach Zen has a highly visible hinge for the base at the bottom of the hand, so the user can open the bottom to dis pose of t he dust. The Dyson has n o hinged base. (Although one cannot see this from the Registered Design, the dust is tipped out of the top of the bin. )

The Mach Zen bin is significantly bigger than the Dyson bin in both absolute and relative terms.

The bin is transparent, and through it the cyclone shroud is visible. For the reasons given above I do not accept Vax's arguments that the transparent bin is not a feature of the Registered Design or that it is excluded under Article 7(1) of the Designs Directive.

In the further alternative Vax contends that the designer's freedom was constrained by the need for t he user to se e when t he bin is full. S o far as this is concerned, it is common ground that there are other possible solutions. One could have a window, one could have a tinted bin and one could have some kind of a detector with an indicator light or sounder. On the evidence, however, each of these a lternatives had drawbacks. A window is unsa tisfactory because dirt may not accumulate evenly in the bin, a tinted bin is unsatisfactory because it is less easy to see the dirt and a detector plus indicator involves added expense and complexity. It follows that ease of use and cost considerations both point to adoption of a transparent bin. Again, therefore, the degree of freedom of the designer of the Mach Zen was rest ricted if he wanted to achieve the best solution.

Counsel for Vax accepted that the Mach Zen was similar to the Registered Design in having a transparent bin while this feature was not present in the existing design corpus. He

As for the shape and appearance of the cyclone visible through the bin, this is completely different between the two designs.

A co-axial pair of wheels at the rear of the machine, oversized such that they extend to approximately half the height of the body of the machine. It is common ground that it is not necessary for a cylinder vacuum cleaner, or even a cyclonic cylinder cleaner, to have large rear wheels. V ax nevertheless contends t hat the designer's degre e of f reedom is constrai ned by technical considerations. The e vidence is that there is an advantage to having a p air of large wheels since this makes it e asier to drag the machine over obs tacles. Furthermore, there is a n advantage to locating these wheels close to the motor which is the heaviest part of the machine, and in both the DC02 and the Mach Zen this is at the rear of the machine.

Once again, counsel for Vax accepted that the Mach Zen was similar to the Registered Design in having this feature while it was not present in the existing design corpus, but submitted that, given the restricted degree of freedom of the desi gner of the Mach Zen, this similarity in itself was not of great significance. Once again, I accept this submission.

Furthermore, as counsel for Vax submitted, the informed user would notice considerable differences between the respective rear wheel designs:

In the Registered Design the housing is at an angle to the wheels such that the front half of the wheels is a pproximately flush w ith t he w heel arc h while t he re ar half protrudes increasingly outside the wheel arch and rear housing but not completely. By contrast, in the Mach Zen the h ousing is parallel to the wheels, and the front half of t he wheels protrudes beyond the wheel arch while the rear half is not covered at all.

In the Registered Design the wheels are located wholly within the housing and do not extend beyond the rear o f the body of the machine. By co ntrast, in the Mach Zen t he wheels do extend beyond the rear of the body of the machine.

In the Registered Design the "tyre" portion of the wheels is rounded and quite stylised. By contrast in the Mach Zen the "tyre" portion is more angular and less stylised so that it looks more like a real tyre.

In the Registered Design the wheel hub is smooth and featureless apart from a shallow central dimple. By contrast in the Mach Zen the hub bears a prominent central raised disc and eight radiating "sun rays".

The wheels are spaced apart to define the widest part of the machine. Vax contends there is little design freedom in this respect since the wheels need to be located at the outer part of the machine to give it stability. Dyson ripostes that the wheels do not have to be set so far apart. I conclude that it is advantageous for stability reasons for the wheels to be spaced apart and to that extent the designer's freedom was restricted. Furthermore, the effect of the first of the differences identified in paragraph 74 above is that in the Mach Zen the wheels appear to be spaced further apart than in the Registered Design.

Prominent wheel arches extending over the top of the wheels, part of each of which is formed by an elongated operational button. It is convenient to deal with the wheel arches and buttons or pedals separately.

So far as the presence of wheel arches is concerned, Vax contends that the designer's freedom is constrained by the need to prevent debris from dropping into the wheel mechanism. I accept this. Furthermore, the wheel arches are bound to have some prominence given the size of the rear wheels. Apart from that, however, the respective arches are quite different. In the Registered Design, as note d above, t he wheel arch extends almost flush over the wheel. In addition it is designed to align with the front and rear housings so that three parts together form a smooth curve around the wheel. At the leading end this curve is high up the wheel while at the trailing end it is low. By contrast, in the Mach Zen the wheel arch is separate from the main housing and shaped more like a bicycle mud guard. It extends from low down at the leading end to just beyond the top of the wheel.

As for the b uttons or pe dals, V ax contends that t here are a n umber of des ign constraints. One needs to have an on-off switch and a cable rewind activator. It is desirable, although not necessary, to have the former close to the motor and the latter close to the cable rewind. It is a lso desirable, although not necessary, to place them over the r ear wheels for stability reasons. Furthermore, it was common in the design corpus for buttons or pedals to be provided at the rear of the machine for these purposes. I accept all of this. Furthermore, the designs of the respective buttons or pedals are quite different. In the Registered Design they are formed as part of the smooth curve of the wheel arches. By contrast in the Mach Zen they are separate from and sit on top of the wheel arches, they bear ten prominent parallel ridges, they extend in an L shape around the back of the machine and the rear portions bear raised, moulded switch and cable symbols.

A cyclone top which increases in height towards the rear of the machine. Vax contends, and I accept, that the designer's freedom is constrained in that the shape of the cyclone top follows the air pat h between the cyclone and the motor. Of course this does not dictate the precise shape of the cyclone top, but the shapes of the respective cyclone tops are different in a number of respects.

A lower bin support formed from a sweeping and curved forward extension of the wheel arches. Vax contends that there is little or n o similarity in this respect. I a gree. One can certainly s ee this feat ure in t he Re gistered Design, a nd it is one of t he fe atures w hich distinguishes the Registered Design from the existing design corpus. But there is really nothing like it in the Mach Zen. It should be noted in this respect that, unlike the Registered Design, the Mach Zen is asymmetric along its longitudinal axis. Viewed from the one side there is a prominent run of hose along the bottom of the bin whereas there is no such feature viewed from the ot her si de. Thi s m eans t hat on e si de vi ew of th e M ach Zen i s more d ifferent to th e Registered Design than the other side view. Even if one focuses on the latter, however, I see little similarity in this respect.

A long arcuate handle sweeping from above the bin to the rear of the machine above and behind the wheel which overarches the cyclone top. It is common ground that the handle does not have to be longitudinal or central to the machine. Vax contends, however, that the designer's freedom is restricted in that it is advantageous in terms of ease of carrying to have a handle which is longitudinal and central. I accept this.

On the other hand, there do not appear to be any design constraints which militate in favour of the choice of an arcuate handle. In this respect there is a similarity between the Mach Zen and the Registered Design which is not present in existing design corpus and in respect of which there is design freedom.

Nevertheless there are a number of differences between the handle designs which the informed user would notice:

The Dyson handle is open and accessible from front to back, enabling the machine to be picked up at a variety of positions including the top of the machine. The Mach Zen handle is only a true handle at the back. The section over the bin (and part of the section over the main body) is filled in. The result is that the handle not only looks different, but also does not enable the user to pick up the Mach Zen at the top of the machine. Furthermore, in the Mach Ze n t he ac tual handle does not overarch t he cyclone t op, a s op posed to the continuation of the handle.

The cross-sections of the handles are different. In the Registered Design the handle has an "I" section. The Mach Zen handle has a semi-circular section.

In the Registered Design the handle terminates well before the rear of the machine. The Mach Zen handle extends over the rear, terminating in a r ecess for the t ool kit which projects out still further.

The arc of the handle on the Registered Design is less than the arc of the handle on the Mach Zen.

A body portion which slopes downwards from the cyclone top towards the rear of the machine. This area houses the air ducts from the motor to the cyclone. Vax contends that, since these ducts pass from a high position (above the cyclone) to a low position (the motor), it is technically sensible for the housing to follow the same line. Of course, one could build up the housing, but this would take up unnecessary space and material. I accept this. Furthermore the details of the respective designs are different.

Vax contends that, in addition to the differences identified above, there are a number of other differences between the Mach Zen and the Registered Design which would have an impact on the perception of the informed user. I agree with this. These differences include the following.

The rear of the machine. The rear of the Registered Design is a smooth curved cover, following the curve of the top of the housing and extending beyond the wheels. The recess for the plug is tucked away right at the bottom underneath the cover. The rear of the Mach Zen is quite different. There is no curved cover. It comprises a vertical wall, with seven parallel plastic ribs and a ventilation grille. The plane of the wall is inside the rear of the wheels. The plug recess is prominent in the middle of the wall.

Hose connector position. In the Registered Design the hose connector is positioned at the top of the machine. This is a prominent feature which serves to emphasize the height of the machine, and enables the front of the machine to be finished off with a smooth, curved bumper. It also involves a swivel joint which is another prominent feature. The Mach Z en has the hose connector right at the front of the machine, emphasizing its length, not its height. There is no

Front wheels. The Registered Design has a pair of castor wheels at either side. The Mach Zen only has only one in the middle, of different appearance.

Cut-away feature. The base of the Registered Design has a right-angled cut-away. This enables the cleaner to sit on a stair and is a striking feature of the Registered Design. There is no such feature in the Mach Zen, the base of which is close to the ground.

Footplate feature. The Mach Zen has a prominent "footplate" jutting out from the main body of the machine and running parallel to the ground on both sides. There is no such feature in the Registered Design.

Wand handle. The w and h andle is t he s ubject of a separa te repres entation on t he Registered Design. It is therefore an aspect of the Registered Design of some significance. The wand handle is roughly banana-shaped with a trigger by a grooved area. The base of the wand handle swells to a joint with a hose collar fastened with a clip. The Mach Zen wand is quite different from the Dyson wand. It is m ade up of three intersecting sections, creating a storage area for tools so t hat they are made easily available to the user. This results in a triangular configuration. There is no trigger, grooved section, clip or hose collar. In addition, the Mach Zen wand handle has a rather longer nozzle.

Standing back from the details and considering the overall impressions of the respective designs from the perspective of the informed user in the light of the existing design corpus and the degree of freedom of the designer, my conclusions are as follows.

In my view the informed user would notice that there were certain similarities between the two designs, in particular the inclined transparent bin, the large rear wheels and the curved longitudinal central handle. For the reasons given above, the informed user would not consider these similarities to be par ticularly significant. T he informer user woul d als o notice m any differences between the respective designs, including features present in the Registered Design which are not present in the Mach Zen and vice versa. For the reas ons explained above, the informed user would consider a number of these differences to be significant, particularly the rear view, the hose connector position, the bin handle, the wand handle, the cut-away feature and bumper of the Registered Design and the asymmetry of the Mach Zen. The overall impression produced by the Registered Design is sm ooth, curving and elegant. The overall impression produced by th e Ma ch Ze n is rug ged, angular a nd in dustrial, even somewhat brutal.

Even on t he basis that the Registered Design is e ntitled to a fa irly broad scope of protection because of the differences between the Registered Design and the existing design corpus and because of the degree of freedom of th e designer, in my judgment the overall impressions produced by the two designs are different.