The August launch contained many new features, appropriate for our web search dominated environment, but retained the power search features unique to classic Dialog. It is this combination that, ProQuest undoubtedly hopes, will restore its luster and prove the value of this pioneering online resource. I was particularly pleased to see Dialog adopt the now web-standard double quotation marks (" ") for exact phrase searching. Left-hand, embedded, and right-hand truncation was also a huge leap forward.

Included in the first release was term highlighting, previews, “search within results,” automatic de-duping (removing duplicate results), filtering by date using a slider, filtering by document types and, in some databases, by specific descriptors, setting up alerts, translation for some languages, and citation exporting to RefWorks, ProCite, EndNote, or Reference Manager. ProQuest also added a MyResearch area to which searchers can store both documents and search strategies, organize research projects, and collaborate with other searchers.

The February 2011 release of the new platform considerably expands its capabilities. Searches can now be conducted in 69 databases, containing 200 million documents, across nine subject areas: Aerospace & Defense, Automotive, Chemistry, Energy & Environment, Engineering & Technology, Food & Agriculture, Healthcare, Pharmaceutical & Biomedical, and Telecommunications & Computing. I find the subject areas reminiscent of Dialog’s OneSearch categories. Free trials of the new platform may not allow access to all 69 files, but the full list of databases is at the ProQuest Dialog site. Although ProQuest Dialog’s focus is on migrating the scientific, technical, and medical databases to the new platform before the business and social sciences, the February release includes British Education Index, ERIC, PROMT, PsycINFO, and Social SciSearch.

Setting Up Search

In the newest release, searchers gain the ability to build sets and stack multiple index or field codes within their search strategies. Set building—think of it as communicating with databases through modifying results components—is critical for professional researchers and was a hallmark of the Dialog and DataStar systems from their inception. To open the screen for set searching, click the down arrow next to Advanced Search at the top of the screen, and then choose Command Line. A rather large search box appears, inviting the searcher to enter terms.

Below the search box are check boxes to limit your search to Full text or Scholarly journals, a dropdown menu to limit by date, and options to display results by relevance or in date order.

To the right of the search box are search tools. Click on Thesaurus for subject terms specific to individual databases, View field codes help for an extremely helpful list of the codes, and Search tips. Experienced Dialog and DataStar searchers will need to scrutinize the field codes carefully as some have changed. Most are still 2-letter acronyms, but others go to 5 letters. The syntax is field code, no space, term(s) in parentheses. Thus to search for articles written by Ojala, enter AU(ojala). For articles about a person, the syntax would be PER(ojala). As a veteran Dialog and DataStar searcher, I’ll admit the codes will take a bit of getting used to. I am also learning the new proximity operators, which still retain the searching power of Dialog and DataStar.

After the initial search request, a screen of results appears. Click Modify search, found directly under the search box at the top of the screen, to return to command line searching. Now you'll find set numbers to the left of the queries. Click Clear form and enter the query that will become the next set number. With multiple sets displayed, scroll down to the Combine searches box to manipulate the sets using Boolean operators. Relevant results can be saved to the MyResearch area of ProQuest Dialog.

Dialog Develops Web Characteristics

This newest iteration of the ProQuest Dialog platform moves the system into the world familiar to searchers who rely on web search engines such as Google and Bing, while still retaining (reinventing might even be the better word) the power of set searching, controlled vocabulary, and sophisticated proximity operators relied upon by professional researchers.

There is more to be done as the remainder of the Dialog and DataStar databases move to the new platform. Determining relevancy, for example, is always tricky, as even Google is discovering. What one person finds relevant might be irrelevant to someone else. A business researcher and a science researcher might be looking at the same topic, nanotechnology for example, but the former is looking for marketing opportunities while the latter wants a scientific breakthrough. Thus far, neither commercial databases nor the free web have solved this problem algorithmically. It remains the searcher’s responsibility to make the judgment.