Assuming that the ECJ will not broadly allow the usage of third party trademarks as keywords (and, as a logical consequence, therefore also not allow the bearing of third party trademarks in the advertiser’s advertisement), both the LG München “POSTERLOUNGE” decision and the BGH’s “PCB”-decision prove inconclusive.
It is, therefore, subject to further discussion how predictable Google’s “Broad Match” and Yahoo!’s “Advanced Match” are in determining whether advertisers who booked generic terms as keywords can reasonably expect to have their (“Adv-“) ads be displayed next to a search result for a similar term which is itself a protected trademark.
1. Placing the burden upon the Providers of Keword Advertising:
Providers of Keyword Advertising give only vague clues to the range of their broad match functions, but fail to present detailed information and rules about them. An option that has many be called for my French courts [citation will be added soon] is to inform advertisers about possible “concurrences” between their desired keywords and protected trademarks.
2. Sharing of the burden between Adwords and the advertiser:
Following a recent French judgement Google Adword’s should have to offer links within its system to enable advertisers to check whether their desired keywords are similar or identical to any know/ registered protected French sign.
(the author attempted to test this function a month ago for Adwords being displayed in France but did not find any links guiding to any trademark databases)
Still, following this system Google has placed the burden onto the advertiser again and as for the broad match it will just be impossible for the advertiser to check out any possible combination/variation Adwords might find appropriate for its chosen keyword, given that most search engines for trademarks are in fact very slow and the use is truly cumbersome.
Following this an automated check by Google or the advertiser for a similarity between keywords and protected signs will in the opinion of the author effectively only be possible after the establishment of a single or a small network of highly efficient databanks, allowing to check in real time.
3. Use of automated Fingerprint Technology:
“Fingerprint technology” is a filtering technology actively and effectively used (e.g. by Google) to identify copyrights infringements on the internet. Using this technology not only copies of complete works can be found on the internet but also parts or segments of the copyright protected work. (For more information on this topic please consult the Website of “Ina” and take a look at their highly informative brochure on fingerprint technology) The suggestion of the author is to require the proprietors of protected signs to sign up to a fingerprint-database (so to speak a black list) which could then search the anomised Google Adword’s keyword database and disable infringing keywords before they ever get used to trigger advertisement.
Still this idea has got a significant downside: Such technology, as any filtering technology would then as a logical consequence also hinder freedom of speech, descriptive use etc. [Angelopoulos, Filtering the Internet for Copyrighted Content in Europe, IRIS plus, 2009-4]
4. Leaving the Burden with the Advertisers
At the same time, it seems highly unpractical for advertisers to include all -fairly similar- trademark protected terms in their list of “exclusive keywords”.
A fact that also should be mentioned at this point that users of search engines tend to make e.g. an incredibly high number of spelling mistakes. Search engines know that and have developed technologies to still deliver the results the users where originally searching for. (this fact is one of the main justification for the Broad Match technology) Should advertisers as a consequence also be banned from booking keywords in which protected signs where spelled wrongly, but which clearly indicate that users where searching for e.g. a trademark?