Indexing the Indexer

[edepot]

One of the major goals of search engines is to provide good search results based on its proprietary algorithms and indexing methods. Many SEOs (Search Engine Optimization folks) try to gain up on the algorithm in order to place their webpages on a higher position or rank on search queries. And so many other companies are starting up trying to provide a better search result on the vast world wide web full of webpages.

What many people don't realize is that it is possible to index the indexer. In other words, provide a better search engine by re-indexing, or using a different algorithm based on the search engine data provided by ANOTHER search engine. For example, google.com provides many ways to allow you to look inside its vast data repository. You can do a site:website.com and it will list all pages it knows of inside website.com website. You can do a link:website.com to find all links linking to website.com (but not all of them). You can do + and - to remove search terms to get at webpages listed inside google.com. If a new search engine comes around trying to do something better than google.com, it does not need to search the whole internet and start from scratch. It can basically run its algorithm and tools on other engines, and thus save time and even space. Note that this is not like a meta search engine where you are basically putting another layer on top of other engines (although you can do that like metacritic.com). What this is pointing to is reorganizing data provided by another search engine, rather than just offering what it has produced.

Maybe someone can provide a neuro-net type of algorithm where it is very flexible and can be changed via weighed parameters, thus adaptable to different searches and data, adjusting to the mined data.