September 16, 2007
Deep Web Search Engines
After using all of these different search engines, I was able to see how important a search engine can make; especially when trying to find deep web content like journal articles. I started by searching for [timber industry California] and compared yahoo and google results. Pretty much these two sites are identical with results and features. Google and Yahoo both provided searchers with local timber businesses in the area along with the search results. I think this was important since it allowed users to call local companies that are part of the timber industry in order to find out more first hand information. Therefore, I feel like its pointless to compare these two engines and its more of a preference on which format you prefer. Additionally, I searched on Yahoo Directory which provided me with categories that might relate to my search. This allowed me to narrow my search results to specific organizations, blogs, etc. I found this to be helpful since if I was looking for timber organizations, I was able to find them extremely fast after clicking the categories button. Generally I found all of my results from these three engines pretty good, but they did not provide me any journal articles or expert documents on this subject. The searches only resulted with articles that were posted on public websites (aka no deep web content). Obviously this was expected since these engines don't attempt to include those results. Therefore, I moved on to Scirus, Google Scholar, UM Library’s Search Tools, and CompletePlanet.
After looking through all of these results, I found a big difference in their usefulness. CompletePlanet was the worst since it did no provide me with any results. In this day and age, I feel like this is unacceptable so I crossed off CompletePlanet.
Google Scholar was also pretty bad since it provided a limited number of results and quality articles. This is most likely because of their limited databases that the engine searches. I was disappointed since I love Google products, but it explains why they still classify it as beta since it does not stand up to the google standard. I hope they improve this product in the future.
Scirus was the best free article search that I found. It found many results from different areas but I noticed mostly that the articles were related to public events about the timber industry. I did not notice many academic studies on the timber industry.
Obviously, the University Library search was the best since it is a search of paid databases. It provided me with many articles about the timber industry and studies about it. I was also confident in the authenticity of these documents because the databases were paid for.
In conclusion, I feel like it is best for web searchers to start with Google to find background data and research for the topic at hand. Then I think you should move on to Scirus and the UM Library Search since these provided the best and most diverse results, even when compared to each other. I hope more deep search engines come around in the near future, but I feel like the royalties for these databases will prohibit public release.
Posted by diamantj at September 16, 2007 02:04 PM