Following up on Jan van Barel’s talk, I am asked to discuss searching from a practical user’s point of view. I will be using a private practitioner’s terminology in talking about the "client" but I think that many of you in industrial practice now are also required to regard the companies in your group as "clients".

In preparing for this talk, I made up a word trade mark and a simple device mark to go with it. I sent a Questionnaire to a number of colleagues who are members of the Trade Marks Group of the FICPI CET. I have summarised their answers in a number of tables which are included with the written papers for this meeting. I would like to thank the colleagues who responded to my Questionnaire and I must apologise also to those members or countries which I omitted. I wish to make it clear that I am responsible for the summaries, which may not fully convey the information which was set out by the person answering the Questionnaire. I am aware also that even within some countries the approach to searching can vary from one professional firm to another and therefore the answers which I received could have been different if I had addressed the questions to another colleague in that country. Nevertheless, I hope that the results will be helpful for the discussion of this morning’s topic.

The trade mark which I invented is OLDCASTLE for cheese. It is to be used by a cheese manufacturer in the small town of Oldcastle, north of Dublin. The company has developed a flavoured cheese product which is expected to be popular internationally, particularly for children’s lunch boxes. The proposed trade marks are shown on the attached representation sheet:

1. Word mark OLDCASTLE, which will mainly be printed in a Celtic form of script.

2. Castle device, which will be used alone or with the word mark OLDCASTLE printed across it.

3. The 3-dimensional shape of each portion of cheese which will simulate a castle as shown in the representation.

I expect that most people in this audience are familiar with the searching of word marks. I will therefore be concentrating on "signs and designs" other than words. But I recognise that many device marks have words used alongside them or incorporated into them, as in the case of my castle device mark. The word may also describe the device itself or at least one of its major features. So we have to think about word marks as well as devices. By the way, I am using the term "devices" in this talk, rather than "designs" because at a later stage I will be discussing possible searching through registered designs, or designs and models, which takes us into a different field of Intellectual Property Rights. Instead of using the term "device marks" I could of course have referred to "figurative marks" or "pictorial marks".

The initial question in today’s topic as set out in the Working Programme is: "Is there a needle in the haystack?".

In every Search, we have to try to answer that question. However the task will obviously be much easier if the needle has a piece of bright red thread attached to it and the client can show us where the thread is lying on the surface of the haystack. Indeed we may not really be searching for a needle at all but for the sword of Damocles which is hanging by the thread over the client’s head and which the client has known about from the outset. The first essential step therefore in preparing for a Search is to get the right information from the client.

INFORMATION TO BE OBTAINED FROM THE CLIENT

There are some basic questions to be asked:

1. Is this a preliminary screening operation, for which there is a "shopping list" of proposed marks? This is often the case with word marks, much rarer with devices.

2. What is the origin or derivation of the mark? In the case of OLDCASTLE, the word mark is derived from the name of the town where the client is based. The device mark and the 3-dimensional shape are based upon an old drawing of a castle which existed in the same area of the country until about two hundred years ago. These sound like good bona fide reasons for adoption of the word and device marks. However we may have to question whether the local geographical and historical features are merely convenient excuses for using a mark which is intended to climb on the back of some other trader’s reputation.

3. What are the existing products or services with which the client’s product or service is intended to compete? What does the competitor’s product or printed material look like? These are particularly important questions in the case of device marks, 3-dimensional marks and get-up.

4. Has the client copied all or part of the mark from another source, either within or outside the field of competition? In the OLDCASTLE case, the client is aware that a leading international chocolate manufacturer sells chocolate products in the shape of a castle having four round turrets at the corners. This chocolate product has been on sale, particularly to children, for about seven years in some countries. The client says that the idea of making cheese portions in the shape of a castle came to him first, and he discovered the chocolate product on the market subsequently.

5. Even if there has been no element of copying, is the client aware of a business in a different field which uses a similar mark and which enjoys a reputation? In my invented case, there is an investment banking firm on Wall Street in New York which has used the name OLDCASTLE for many years and which has a leading reputation in the field of financial services. As my client is based in the town of Oldcastle, he has good cause for using that name and it seems unlikely that his use will take unfair advantage of, or be detrimental to, the distinctive character or the repute of the banking firm’s name. Nevertheless, the knowledge that the banking firm exists directs us towards a possible area of trade mark searching outside the field of food products.

The client may be reluctant to answer our questions, particularly concerning the nearest products of competitors and the possibility that the mark may have been consciously or unconsciously copied. But we have to get the client to understand that a well-directed search is the best search. It may be substantially cheaper if it enables us to do a name search for marks owned by a particular competitor or indeed to search for registrations of a particular mark or design which is already known to the client. There is no point in carrying out a comprehensive and expensive search in order to uncover information which the client really knew from the outset.

But let us not be negative. The client has a good mark and a novel device which to the best of our knowledge is not copied from or taking advantage of someone else’s product. The next step is to decide on the geographical spread of the search.

WHAT TERRITORIES NEED TO BE COVERED?

The client may want world-wide clearance for the proposed mark. For word marks, a World Identical Screening Search can be done but it has limitations, particularly because it does not extend to similar marks. For device marks, a world-wide search would be prohibitively expensive, so we have to be realistic about the countries which can be covered.

We start with the home territory, and other territories in which the client is already active and for which it has a potential marketing interest in the foreseeable future. We can take a step-by-step approach, looking at certain key Registers first and then expanding the search if the mark is not knocked out in the early stages. For example, for an Irish company we would probably do Ireland, Community Trade Marks and the UK initially, and then move on to International (Madrid) Registrations and key continental countries such as Germany, France and the Benelux. If there is even a small possibility of the product going to the United States or being offered via mail order catalogues or on-line shopping, we need to carry out full Searches there, including Federal, State and Common Law Searches. The penalties for neglecting to search in the US can be severe if litigation arises later.

We have to think also that offers to trade on the Internet may reach into every country on the globe. We have to assess the risk/cost balance, which will be discussed later in this presentation.

HOW RELIABLE CAN SEARCHES OF REGISTERS BE?

In my Questionnaire to colleagues, I asked whether a Search for word marks would be carried out on an on-line database and I also asked how reliable are the records in which Searches are carried out. As you will see from the summary table, the majority of the respondents indicated that a database would be used, although in some cases the database would be available on a CD-ROM rather than being on-line. The real issue as Jan van Barel has pointed out is whether the search parameters, for example to determine phonetic similarity, are applied as part of the computerised system or are devised by the Searcher in order to extract the relevant information from the database. We can easily do an initial screening for identical marks in such a database, and the answers which I received from my colleagues indicate that in most cases they regard the reliability of the databases as high. However the reliability of a full search result depends upon the searcher, whether it be computer or human, and ultimately on the analysis of the search results.

DEVICE MARKS

When we look at searches for device marks, it is an entirely different situation. Device marks can be held in a database as digitised images and the classification of the elements of the device can be stored in a manner which facilitates retrieval. However the on-line delivery of digitised images is extremely slow and it seems at present that distribution of device marks on CD-ROM is the most practical option.

Good CD-ROM systems are available for the Community Trade Mark and also for UK Trade Marks. The images can be blown up on screen and are reasonably good if the device is not too complex in its details. Before publication of a mark in the Bulletin, this is the only image which is available; after publication one can go to the printed version in the Bulletin also.

Jan van Barel has mentioned the Vienna Classification for the Figurative Elements of Marks. This is the system used by a substantial number of countries, even though they may not have formally ratified the Vienna Agreement. The Vienna Classification is based on 29 categories, each of which is split into divisions which in turn are split into sections. In the case of the castle device of my Questionnaire, we would be looking at the following:

Category 7: Constructions, structures for advertisements, gates or barriers.

Division 7.1: Dwellings, buildings, advertisement hoardings or pillars, cages or kennels for animals

Section 7.1.1: Castles, fortresses, crenellated walls, palaces.

The OHIM uses the Vienna Classification down to section level and it is ideal for a simple device such as a castle. However the sections are very narrowly defined and at least one of the CD-ROM products only classifies CTM device marks down to division level so that, for example all device marks for buildings are brought up by entry of the search parameter.

While access to images on a practitioner’s desk is very useful and an initial screen can be performed by using the Vienna Classification, I think it is clear that device searches need to be carried out by expert searchers who are familiar with the classification and the mind-set of those who allocate the divisions and sections to elements of marks.

3-DIMENSIONAL MARKS

3-dimensional marks are included within the same classification system as device marks. The guide to the Vienna Classification contains an interesting set of examples of bottles and flasks which are allocated to division 19.7 and various sections of that division. The Community Trade Mark system provides for the identification of a mark as being 3-dimensional but this does not necessarily indicate that the distinctive character is associated with the shape of the product. The representation may show a distinctive get-up applied in a 3-dimensional manner on a product or package of conventional shape. It would be helpful in these cases if the OHIM would make greater use of article 38.2 CTMR and request a disclaimer of any exclusive right to non-distinctive elements of the mark so as to reduce the doubt as to the scope of protection. However that is another topic.

When the trade mark really does consist of the shape of goods or of their packaging, then the scope of searching may have to be extended beyond that of device trade marks. Many of you will probably be familiar with the recent case of Yakult v. Danone in the District Court of The Hague [1998] E.T.M.R. 465, which was concerned with small bottles for holding a fermented milk drink. The Japanese company Yakult Honsha had not only registered the shape of the bottle as a trade mark in class 29 but they had also protected it under the Uniform Benelux Act on Designs or Models. The case was decided on the basis of infringement of Yakult’s right in the 3-dimensional trade mark, not the design registration, because in that case there were doubts about the novelty of the design at its date of registration. Nevertheless the case illustrates the overlap between trade mark protection and design protection for 3-dimensional packaging or products of this kind.

In my sample case, I suggested that cheese portions would be made in the shape of a castle and that the client already knew that a chocolate product was being marketed in a similar shape. The majority of the respondents to my Questionnaire indicated that it would be advisable in this case to carry out a search through Registered Designs, although this would not generally be part of a trade mark clearance. In a number of countries design searching is done in the Registry, and no search facilities exist for outsiders. Other countries do provide a database, although the problem of providing images of the registered designs is obviously even more acute than with device trade marks. Designs can be classified under the Locarno classification, with 32 classes, each divided into sub-classes, which is suitable for computerisation.

COMMON LAW RIGHTS (Unregistered Marks in use)

The need for searching word marks which are used but not registered clearly varies substantially from country to country. In the United States, a Common Law Search is almost automatically a part of a trade mark Availability Search. In the event of litigation, the absence of a Search may be very costly to the defendant, in terms of damages or of attorney’s fees. I understand that a case on this issue between International Star Class Yacht Racing Association and Tommy Hilfiger, Inc. [1998 U.S. App LEXIS 10642 (2nd Cir. 1998), extracted in INTA Bulletin, Vo. 53, No. 15] is still oscillating up and down between lower and appellate courts with no end in sight. In that case, Hilfiger’s attorneys had screened the mark STAR CLASS for clothing in the Federal Register only. They advised the client of the need to conduct a full search but it was not done until after the Association sued for infringement of its unregistered mark. The courts are considering whether Hilfiger acted in bad faith, giving rise to an award of damages and attorney fees.

Systems are well established in the US for doing Common Law Searches, with considerable help from on-line databases. Internet Searches can also be carried out to find occurrences of particular words in association with specified goods or services. However it should be recognised that Internet Searches are far from conclusive because they depend upon the extent of coverage of the particular Search Engine used, which may be only one half of the total number of pages on the Web.

Common Law Searches are also recommended in other countries which recognise rights in marks arising from use, but they are carried out less frequently than in the USA and less information is available on databases, although Internet Searches can of course be done from anywhere. Company Name Searches will generally form part of a Common Law Search and they are also recommended for most of the other countries covered by my survey. However from a Practitioner’s view point, Company Name Searches are difficult to assess if little information is given about the nature of a company’s business or whether it is actively trading.

Domain names are the subject of another session at this Forum. It is sufficient now to say that domain name searches should be included on the same basis as Company Name Searches but with the greater convenience that Domain Names can be searched world-wide (at least on an identical basis) by a simple and economical computer search.

For Device Trade Marks and 3-Dimensional Marks, there is no practical way of carrying out Common Law Searches by computer. One has to go back to the basic questions to the client, and ask for information about competitors or traders in other fields who are known to be using devices or get-up which could be regarded as similar. Internet access can then be used to review the websites of the other traders to see whether their logos and product get-up can be seen. However ultimately it will probably come down to a survey of the market, carried out by an Investigator or, at least initially, by the client itself.

WHAT DO RECENT IMPROVEMENTS IN ACCESS TO DATABASES MEAN TO THE PRACTITIONER?

For searching word marks, life has become much easier because searchers do not need to spend time in the Trade Mark Office. Initial scanning of word marks can be done from a practitioner’s own desk. For device marks and 3-dimensional marks, databases on CD-ROM are a great convenience as a screening tool. Internet searching can also be done but it is likely to require too much time and to produce too much information for the Practitioner’s needs. Searching and analysis by specialist searchers are still, in my view, to be preferred.

WHERE IS THE RISK V. COST BALANCE?

For Trade Mark Searches you have to decide how many classes to cover. In my OLDCASTLE example, it is clear that a full Search is required in classes 29 and 30. Some colleagues added classes 32 and 33 for beverages, and class 42 for "providing of food and drink". Others included class 36 for "financial affairs" because of the investment banking firm called OLDCASTLE. An identical Search in all classes is recommended. This is feasible for word marks, but extremely expensive for device marks.

You also have to decide on the number of countries or regions to which the Search should be extended. This can be done on a step-by-step basis but the client will probably want to know the likely overall cost for a satisfactory Clearance Search.

In some cases, you may recommend that an application should be filed in order to get the results of Official Searches. In the United Kingdom, for example, search results are now available within a few weeks after the filing date. Search Reports on a Community Trade Mark application also provide useful information, although they can scarcely be regarded as Clearance Searches and at present they suffer from a long delay.

If the client knows the market well and provides good information about it, Common Law Searches may not be necessary except in the United States. Searches in Registered Designs and Models will only be required in the case of a product or packaging whose outer shape or ornamentation appeals to the eye.

The risk to the client depends on the circumstances in each individual case. Ultimately the cost to the client can be the withdrawal of the product from one or more markets and the need to redesign all packaging, labelling and advertising material. But search costs will not usually be regarded in this way, but will be considered as a proportion of the funds available for developing and launching a new product or service.

Clearly if the client is planning to do a marketing campaign in a particular country, to appoint a distrubutor or even to set up a subsidiary company, then full searches in that country need to be done. On the other hand, if the client is merely hoping that the product might eventually find customers in a country, then it is reasonable to omit searching at the initial stage. As I have mentioned, particular problems now arise with products or services which are to be traded over the Internet. No one can guarantee that a mark is clear in every country of the world.

I understand if an unexpected conflict does arise, it may be possible to restrict access to the client’s website from a particular country where there is a prior right so that goods/services are not supplied via the website to users in that country. If the client wishes to continue selling into that country via the Internet, the consequence and resulting cost could be the need to rebrand the product/service (and redesign its packaging, labelling and advertising material) with a non-conflicting mark and to set up a separate area on the website specifically for users from that country from where the goods/services with the non-conflicting mark are supplied. Whether the mere availability of the client’s website in the problem country can constitute an infringement is a separate question. If it is an infringement, the possible ultimate cost to the client could be the cessation of trading under the mark via the Internet altogether. Obviously, these questions involve an area of great legal complexity and uncertainty which will depend on the applicable law in a particular case and further detailed discussion is outside the scope of this paper.

The cost/benefit ratio in searching for device marks is very much higher than it is for word marks, but the risks of confusion between device marks are less because no question of phonetic similarity arises. The more complex and unusual the device, the smaller is the chance that it will run into a conflict.

Ultimately the client has to make the decision. It is a matter for the practitioner to make it clear to the client:

(a) What Searches have been done and the limitations of those Searches e.g. time delay in entry of marks in the Search records; Convention applications which could be filed later; and possible errors or subjective classification decisions in the search records.

(b) What further Searches could be carried out and whether these are recommended in the circumstances of the case.

QUESTIONNAIRE and Survey RESULTS