Metasearch engine
A meta search engine is a search engine whose essential feature is that it forwards a search query to several other search engines at the same time, and collects and processes the results. A metasearch engine does not have its own index. Search engines that forward search queries to other search engines and at the same time have their own index are called hybrids.
history
The first Internet search engine called MetaCrawler was programmed by Erik Selberg and Oren Etzioni in 1995 as part of a research project at the University of Washington (USA). A year later, employees of the Regional Computing Center for Lower Saxony (RRZN), under the direction of Wolfgang Sander-Beuermann , developed MetaGer, the first German meta search engine.
Another well-known metasearch engine is DuckDuckGo . Some previously independent search engines, such as WebCrawler and the American edition of Excite , are now also active as meta search engines.
The current generation of meta search engines also allows syntax translations so that more complex search queries can be sent to the respective search engines.
technology
The results can easily be arranged one after the other. Complex post-processing, such as
- Aggregation of duplicates (the same find in several search engines is only displayed once, influencing the evaluation mechanism),
- Evaluation or ranking of the results,
- Consideration of the order in which the results were delivered by the individual search engines ( search engine ranking ),
- Grouping the results.
The results are then usually presented uniformly, corresponding to a search engine.
Meta search engines have no access to the indexes of the source search engines and thus also not to the information contained in their indexes that support a ranking according to relevance criteria. The source search engines only return a list of results to the meta search engine.
Traditionally, the server of the meta search engine has to wait for the answers from all search services to which it has forwarded the search in order to only then start displaying the results. This results in delays compared to a normal search engine. To counter this, an updated display can be made when different search results arrive or slow-reacting search engines can be excluded from the search.
Colloquially, search mask collections are also referred to as meta search engines. The difference, however, is that search mask collections display the results in individual external windows on the original interface of the respective search engine and do not combine the results like meta search engines. These search mask collections include the formerly also independent HotBot .
Legal admissibility
According to a decision by the European Court of Justice, meta search engines are illegal if a search form comparable to the platform being searched is offered, the search queries are forwarded "in real time", the search results display merges duplicates and the platform being searched does not agree to their use. In view of the reasons for the decision, the business model of unauthorized search and data analysis services (querying third-party databases without a contractual agreement) is fundamentally in question. In German copyright law, the judgment essentially corresponds to the protection of database works, whose claim to protection is derived from the compilation of the entire work and is a service worthy of protection, even if individual components are not worthy of protection.
See also
Individual evidence
- ↑ Diego Wyllie: Eight special search services in the test - Metacrawler: The search engine of the search engines. In: Computerwoche.de. April 22, 2012. Retrieved August 23, 2017 .
- ↑ Use search engines safely. In: Consumer Center.NRW. May 19, 2017. Retrieved August 23, 2017 .
- ↑ Search engines: the best Google alternatives. In: BR.de. May 3, 2017, archived from the original on March 7, 2018 ; accessed on August 23, 2017 .
- ↑ Decision C 202/12 of December 19, 2013
- ↑ The end of the metasearch engine?
- ↑ § 4 UrhG, collected works and database works
Web links
- Link catalog on the topic of meta search engines at curlie.org (formerly DMOZ )