Resource Review #8: Bing or bust: Can Microsoft Cure ‘Search Overload Syndrome’?

Link: http://search.ebscohost.com.ezproxy.library.wisc.edu/login.aspx?direct=true&db=lxh&AN=44864948&loginpage=Login.asp&site=ehost-live

Johnson, B. E. (2009). Bing or bust: Can microsoft cure ‘search overload syndrome’? Computers in Libraries, 29(10), 36-40.

This article is an overview of Microsoft’s new search engine Bing. The author starts with a good synopsis of the current state of search, both as a business and how effective its current state is. Most people are currently happy with search engines, but a “vocal minority insists that search is broken” (p. 38). Search ten years ago was about who had the biggest index of the web. Now, with the increased amount of material, Microsoft is trying to bring about a “new” concept of categorized search. The author notes that librarians have been doing this for ages, and this is nothing new. Still, it is new for web search engines.

It’s interesting to see search engines evolve along the same lines as libraries. Librarians figured out a while ago that subject and faceted search was superior to keyword search, and it seems that web search is figuring out this independently of librarians. As the author says, “it was librarians who carried the torch, stubbornly insisting on cataloging things in an age of metatags and keyword searching”  (p.40). I find it exciting to see “new” ideas interjected into search, but I would not be surprised to find many librarians frustrated with the current state of web search. The author in this case has a very calm, if a bit smug, tone about the whole thing. He does a very good job of laying out what Bing is doing that is different from Google, and questions what effect this will have. It certainly will be challenging to carve a bigger piece of the search pie, even for a major player like Microsoft, but competition is good. Even if Bing can get Google to expand their ideas of how people search it will be to all our benefit. I think the real challenge for librarians is moving past the idea that we did it first and if we can convince people so, they’ll give up their fling with Google and come back to us. In reality, it will be much more advantageous to learn from the popularity and usability of web search and incorporate this into library’s searching.

Resource Review #7: Will Google’s keyword searching eliminate the need for LC cataloging and classification?

Link:  http://search.ebscohost.com.ezproxy.library.wisc.edu/login.aspx?direct=true&db=lxh&AN=33957864&loginpage=Login.asp&site=ehost-live

Mann, T. (2008). Will Google’s keyword searching eliminate the need for LC cataloging and classification? Journal of Library Metadata, 8(2), 159-168. doi:10.1080/10911360802087366

Thomas Mann takes a stance against keyword search, which is emerging as the default for many journal databases and OPACs. He is mostly focusing on the Google Book project, which he refers to as Google Print, and how keyword searches within that system will be much less efficient than would subject searches. He recognizes that keyword searches are good for quick known searching for a particular word, but goes on to say the main weaknesses are inability to supply extra useful terms, synonyms variant phrases and different languages. He also notes that Google lacks the power of Boolean search mechanics, such as nested parenthesis, wildcards, and proximity operators.

I think Mr. Mann raises some good points about the differences between keyword and subject searching. For one highly skilled in subject searching, that method is very powerful. Inclusion of metadata along with the books in Google books would make them more more findable. However, Google doesn’t seem to be implementing this as of yet, thus making it difficult to find books within the project as easily as one would in a library with high quality subject searching.

Mr. Mann is using this example to highlight the continuing need for subject cataloging, to the point that the article seems like a response to a proposal to cut funding to subject cataloging. The final paragraph is more concerned with making a case for subject cataloging in the Library of Congress’s budget than with the downsides of keyword search. To a degree, it’s this kind of tone that makes this article harder to read. The author has an obvious bias towards old school cataloging, with a distaste for digitization and keyword search. He compares Google Books to the Tower of Babel on multiple occasions and uses strong language such as saying Google is “hamstrung by poor search software” (p. 160). I think he makes good points, but they’re taken away by the obvious bias and strong anti-digital language. With all of the metadata that has been added to items through catalogs, it is indeed very important to use this when it comes to refining the way we search. Keyword search, in my opinion, does have things to offer though.

Resource Review #6: Libraries and the Future of Search

Here’s the link.

http://www.webpages.uidaho.edu/~mbolin/odell.pdf

O’Dell, J. (2009). Libraries and the future of search. Library Philosophy & Practice, 11(1), 1-13.

Ms. O’Dell writes a good analysis of where search is going. She analyzes both Google’s mindset in where to go next as well as where libraries should be headed. The idea of the “smart search” is where we’re headed. This can take many forms, but essentially it knows what the users wants no matter how the question is formualted. This goal is far off, but were it to be achieved, there’d be not much use at all for users understanding how libraries organize information. Essentially they could have their own personal reference library at their fingertips. Efforts to move this forward are exploring more focused searching, more personalized searching, and using words as concepts instead of just keywords.

She also sees online public access catalogs (OPACs) as inefficient, as they don’t leverage all the data they could. She explores why users are preferring Google. The best point made here are that poor Google searches are more effective than poor OPAC searches. Additionally, users tend to like the interface better on Google, it is cleaner and there’s less pages to navigate before a results page is reached.

This article is a very good analysis or where OPACs and library’s search need to go. She makes a good point that “catalogs empower librarians, who understand how to use them” (p. 5). This makes search in library a much more powerful tool for librarians than it does for the patrons. I think she raises a good point here, that if the library is going to match Google in terms of use, it must reevaluate how search is being done. There must be a way to move towards simpler more user friendly search without sacrificing the quality scholarly research that library tools excels at.

Another good point that the author makes is that “librarians move away from teaching people to use tools and become more involved in the development of those tools” (p. 7). Librarians need to take a more active role in what is being used to search in libraries. OPACs are outdated by the web’s standard, and librarians need to move into the field of leveraging their previous knowledge of organizing information and retrieval with the new technologies and expected norms present. Instead of merely reacting for or against the new tools, it’s important to find a way to modify them for the libraries needs or create completely new ones. She ends with some commentary about those resisting the “Googlization” of libraries. She advocates for working with this new paradigm instead of fighting against it. Good advice.

This has a graph!

I love graphs. Here’s a link with one. Looks like the recent news of Bing hitting 10% market share on search is a bit overstated. Also, holy cow, really Google has like 75%, boggling.

 

Resource Review #5: Google @ Your Library

Here’s the link.

http://search.ebscohost.com.ezproxy.library.wisc.edu/login.aspx?direct=true&db=lxh&AN=39259512&loginpage=Login.asp&site=ehost-live

Grenzeback, M. (2009). Google @ your library. Nebraska Library Association Quarterly, 40(1), 4-9.

The author makes a good point that librarians either decide Google is the only tool really needed or try to completely ignore it and recede into the print world (p. 4). Ms. Grenzeback write an easy to read article with lots of citations on how Google is currently being used by librarians, and how it could be used, or not. It’s an article that is refreshing in that it takes an even approach where “sometimes Google IS the best choice; often it is not” (p. 5).

She points out that Google is a great tool for finding known information, when you know something is out there, just not how to get at it. It’s also great in that it has a very easy to use and understand interface. One of the downsides to current online public access catalogs (OPAC) search methods is the need to understand filters and Boolean search operators.

Beyond Google, some good points are made about its limitations. It’s interesting to note that “Google indexes only the first 101 KB of a website” (p. 5). This of course means that for very in-depth pages, one is not getting the full information when doing a search. Additionally, Google does not have access to what is know as the “Deep Web” or “Invisible Web,” which are portions such as sites that require a paid subscription. That’s where library search engines do much better. A library gives you access to academic journals in a different way than Google or even Google Scholar can. Although, perhaps that gap is closing.

The author recommends taking advantage of the popularity of Google and advocates for librarians to become Google experts. Most users know basically how to use Google, but not how to really get the most of out their searches. Librarians have a long history of aiding patrons in their own search. Learning and teaching how to use Google is no different. Additionally, it’s possible to set up link resolvers to help integrate Google’s results with the library. So that a user is pointed to a resource in the libraries own catalog after a Google search.

While short, this article is very good. Ms. Grenzeback offers a very positive article about the promise of using Google inside a library. Her arguments are well laid out and she’s pulled from a lot of different research on how people and libraries are using Google or are scared of it. I agree with her that while Google can’t be the end all for search, it’s a great tool, one we must learn out to use better if we’re going to help out patrons and ourselves. Additionally, knowing its limitations lets us know when to make the jump to our deeper resources both online and in print.

Feature Battle!

Below’s a good rundown of how Google and Bing keep trying to out feature each other. Supposidly, Wolfram Alpha is integrating with Bing, which is awesome because Wolfram Alpha is awesome, but I have yet to have success doing it. One could always go here instead in case you need to know the calorie count of various cheeses and search the internets.

Bing vs. Google: Feature Wars from PCWorld

Resource Review #4: Google vs. Bing, Wolfram Alpha

Here’s the link.

Nicole, Kristen. (2009, October 22). Google vs. Bing, Wolfram Alpha:Pushing Search Evolution. ToMuse. Retrieved from http://tomuse.com/compare-search-engine-google-bing-wolfram-alpha/

This short article lays out a fairly basic opinion on new search engines Bing and Wolfram Alpha’s chances of competing with Google. The consensus being they’re pretty good, but not good enough. It’s an opinion piece, but brings up some good points on why Google is on top and why Bing and Wolfram Alpha won’t bring it down.

Google’s search was different than anyone else’s when it came on the scene and it quickly became the front runner. Since that time the company has expanded to a large variety of neat web tools that in a way encourage people to stick with Google, either by incorporating the search somehow or merely from a general sense of thinking Google is neat. The author is of the opinion that Google’s interface has become bogged down in adds and layout, this is where Bing steps in. Bing’s layout is a bit better in its general appearance and in mixing the different media types in one list. This is similar to what was discussed earlier with the search engine test. The interface is better, but not so much that it will draw many people away from the familiar Google. The author goes as far as to say that Google’s results are better, a claim that I’m not sure I agree with, and it’s unclear whether this is personal preference or based on something else.

Wolfram Alpha is a different beast altogether and I find it difficult to compare it to Google in a meaningful sense. Wolfram isn’t really a web search engine at all, as it has its own internal database that it pulls from. Instead of finding web pages associated with one’s search it uses data in its database to computer an answer, whether it be a mathematical formula or the nutritional information of your dinner. I agree that Wolfram Alpha isn’t set to take Google’s market share since it isn’t a web search engine. An interesting point that the author makes to strengthen her case to stay with Google is that Google is currently working on it’s own version of what Wolfram Alpha does in Google Squared. I didn’t spend a whole lot of time with Square, but it seems quite different than Wolfram.

I think the author’s point that new challengers to Google won’t threaten them at all is valid, but I am a bit hesitant as to her reasoning.

Resource Review #3: How Scholarly is Google Scholar?

Here’s the link.

http://search.ebscohost.com.ezproxy.library.wisc.edu/login.aspx?direct=true&db=lxh&AN=40507818&loginpage=Login.asp&site=ehost-live

Howland, J. L., Wright, T. C., Boughan, R. A., & Roberts, B. C. (2009). How scholarly is google scholar? A comparison to library databases. College & Research Libraries, 70(3), 227-234.

This study sets out to compare Google Scholar to that of traditional library databases, and also to find how this comparison stacks up across disciplines. The write up does a good job of explaining in detail the study. The methodology is sound and the conclusions reasonable. It also does a good job of noting in what ways it is limited.

In terms of methodology, seven subject libraries were asked to give a typical reference question in their field, the database they would use to search for the answer and the search terms they would use. The researchers then entered the same searches in Google Scholar. Taking the first 30 results from the library database and Google Scholar, they merged and scrambled them in such away that it would be impossible to discern from which database a source was from. Then they took the list back to the librarians and asked them to rate each source individually based on different factors that made up a “scholarliness” value. Then the answers were put through a formula to correct for individual tendencies to vote low or high and a final scholarliness number was ascertained. The categories within the scholarliness value encompass all of what is important in a source and the methodology seems to cover as many bases as possible in reducing secondary factors.

The results are a bit surprising in that generally the citations contained only in Google had a higher scholarliness value than those only in a library database. The researchers see this as a trend that will continue as more and more publishers share their content with Google Scholar.  Additionally, there was no meaningful difference across disciplines.

The researchers do a good job of identifying the limits of extrapolating their findings. Given that the number of librarians were 0nly seven, and all from the same university makes it hard to extend the findings. They also note they were limited by a three point Likert scale as opposed to a five or seven point. The biggest point made here is that the study was used by skilled librarians and not by students, who search in quite different ways. Therefore, it’s hard to know if a student would have better or worse luck than what is reflected in this study.  The very last point made is also important: Google Scholar acts as a discovery tool, and does not provide access to the content that databases do. Overall, the researchers did a good job of identifying the limitations of their study while presenting interesting findings in a very clear way.

Resource Review #2: Investigation of the Accuracy of Search Engine Hit Counts

Here’s the ugly link

http://web.ebscohost.com.ezproxy.library.wisc.edu/ehost/detail?vid=1&hid=11&sid=4ff1bf76-8665-4e49-af8d-5273c27ce970%40sessionmgr4&bdata=JmxvZ2lucGFnZT1Mb2dpbi5hc3Amc2l0ZT1laG9zdC1saXZl#db=lxh&AN=43405152

Uyar, A. (2009). Investigation of the accuracy of search engine hit counts. Journal of Information Science, 35(4), 469-480. doi:10.1177/0165551509103598

This article details a study on the accuracy of hit counts. It seeks to answer the question of whether or not search engines correctly report how many hits are retrieved for a particular query. Comparisons are made between Google, Yahoo and Microsoft Live Search. At first thought it doesn’t seem a worthwhile subject. Most queries turn up hundreds upon hundreds of results with people generally choosing from the first handful, so what importance could there be to knowing how accurate the hit counts are? The author points out hit counts have been used to map such things as musical genres. Perhaps it’s not useless after all.

The results are interesting in that for both single and multiple string queries all three search engines are inaccurate. Google was the most accurate on a single string, and all three engines’ accuracy decreased on multiple strings. Generally, for what it is, the study is straightforward and well done. It notes its own downfalls such that it only is looking at queries that return less than 1000 documents and cannot be generalized for queries larger than that. Additionally, despite being an article from this spring, two out of the three engines are or are soon to be obsolete. Microsoft Live Search is now Bing, and Yahoo’s search is soon to be powered by Bing.

Resource Review #1: Google vs. Bing Preference Study

Link.

Catalyst Group. (2009, June). Google vs. Bing Search Engine Preference. Retrieved from http://www.docstoc.com/docs/7723136/Catalyst-Group-Bing-V-Google-Usability-Study

The Catalyst Group, a usability group, set out to gauge users’ preferences between Google’s search engine and Microsoft’s Bing search engine. The sample size of participants was only 12, the number of searches only 2, and all the users were all already Google users making it inconclusive from a scientific standpoint. However, the conclusions and method do raise some interesting points about the difference between the two search engines and factors that users care about.

In terms of methodology, users eye movements were tracked while doing searches, they were then engaged in conversation about where they looked and their thoughts on the two different search engines. Finally, a survey was given asking about different factors between the two search engines.

The eye tracking results reveal more about the sites ad placement than how a user goes about searching. Bing searchers’ eyes lingered longer on or near the top ad bars than did Google searchers. For the survey section, arguably the most important result, relevance of results, came out as a tie between the two engines. For the other factors in play such as design of the site and filtering functions, results came out in favor of Bing. One would think from these factors that Bing would be the winner. However, when asked which search engine they’d prefer users stated they stick with Google in an 8 to 4 split. Rational for this was mostly centered around a familiarity with Google and no compelling reason to switch. As catalyst CEO Nick Gould put it, Microsoft “created something as good as Google and that is not good enough.” If Microsoft is going to edge in on Google’s share of the search engine market, they’re going to have to do a lot better than just as good.