Last week I was eager to get some more information about how the librarian community dealt with google scholar. I discussed with few of them and found this interesting summary:
Finally, specialized databases are still valuable for many types of seaching, including searching for "scholarly material." Why? A few examples: + Limit to material published by date + Ability to view more than 1000 results. The Google cutoff of 1000 results is still in place with Google Scholar + Google is constantly crawling the web but we don't know how often the Google Scholar database is being refreshed with new content + Searching using a controlled vocabulary/subject searching + Ability to limit by publisher affiliation + You can limit by author with author: but you can only use a last name. First names and initials are not searchable when using author:. + Proximity operators + Gobs and gobs of content, Google Scholar still doesn't have it all. Example: While you're find John Grisham books, you will NOT find material (both current and archived) from MANY newspapers, trade publications, and general interest periodicals. + I'm sure you can add many more examples of what's NOT there
Some basic facts: + In a nutshell, Google has built an algorithm that makes a calculated guess at to *what it thinks* is a scholarly content mined from the OPEN WEB, and then makes it accessible via the Google Scholar interface. + Precisely what makes something "scholarly" enough to be included in Google Scholar, Google will not say. And this is not an insignificant omission
+ Material accessible via Google Scholar can also be in the main Google index. + Google Scholar results pages *will not contain advertising* -- at least for now. + Some examples of material from major publishers whose material you'll find (we know Google has been working with many)? Google will not provide us with a complete list, but look for content from ACM, IEEE, and yes, Open Worldcat material from OCLC. We also don't know precisely what is and is not available, date ranges, etc. * In some cases Google will be crawling and searching the full text of an article but users will either have to have a subscription to the content or pay for access to an individual articles. + VERY COOL! For many citations, you'll find a direct link to other articles in the Google Scholar database that cite the article you've selected. Yes, Google Scholar is a citation database too! This reminds me of two specialized databases that focus on specific types of scholarly content accessible on the open web that have been online for many years and remain EXCELLENT tools. 1) CiteSeer (focuses on computer science material, info tech content) 2) SmealSearch (focuses on business material)
Might this be a golden opportunity for the library community to tell people -- look, we have access to this stuff and MUCH MUCH MORE? We have better ways to search it, and you might not even have to pay for it? Well, yeah...but if what we've seen in the past is any indication, this is not going to happen. Maybe this time it will be different. Bottom line: It's very difficult to compete with the Google marketing machine. In the meantime, we'll be extremely interested in the response to Google Scholar from fee-based database publishers. Some might ask, are specialized database tools still necessary? Info pros know they are but we sure haven't done a good job of explaining why.