SweetSearch is a specialized search engine designed primarily to satisfy academic search queries. Owned and operated by Dulcinea Media, SweetSearch can be found at http://www.sweetsearch.com. As the self-proclaimed Curator of the Internet, Dulcinea Media’s mission is to help educators teach students how to use the Internet effectively, safely and responsibly. In accordance with this ideology, the design of SweetSearch is structured on the principle that high-quality results can be more desirable than a high-volume of results.
A staff of research experts, librarians and teachers are constantly reviewing and fine-tuning results and websites within the scope of a SweetSearch query. Sites from highly credible organizations such as the Library of Congress, the Smithsonian and PBS are given a prioritized ranking during this process. Currently, SweetSearch only searches 35,000 sites that have all been approved by the aforementioned staff.
With SweetSearch, it’s likely that a set of results will contain many primary sources, which can be beneficial for academic searching. As such, SweetSearch lends itself to research projects – especially those that are academic in nature – but it can also be expanded to use with more general searches. On the website SweetSearch is coined as ‘A Search Engine for Students,’ and while the utility of the site for students is unquestionable, anyone with a hunger for knowledge should be able to find some value in the service regardless of his or her schooling status. For a great video demonstration on many of the features described within this chapter, take a look at this lesson by Professor Scott Moore of UM’s Ross School of Business.
Academic Search Strategies
Before diving headfirst in to the fascinating specifics of SweetSearch, it’s important to touch on some general strategies for academic searching. If an academic search is approached the same way one attacks an everyday Google search, the content of the result set will suffer. An academic search differs from other searches in a few important ways that will be outlined throughout the next few paragraphs. Strategies for dealing with the differences are highlighted in each section.
Refine The Search Query
Often times a typical search might be based on a concept with a wide scope, or a confusing subject that the searcher knows little about. These kinds of searches can be funneled into Google, and with some perseverance and luck, a result set that contains some relevant information will be discovered. While blindly throwing general terms at Google may be effective for a quick query about a personal topic of interest – it will be much more difficult to retrieve results for academic use in this fashion. Academically oriented search engines such as SweetSearch won’t provide desirable results without a more fine tuned query. Using Google as a starting point can be a good option to help improve your search. After a little bit of help from Google, a searcher may have developed enough knowledge about the topic to generate a good query to try out on SweetSearch. Search operators can also be a good means of refining a search and they will be discussed in more detail later in this chapter. Berkeley’s Library hosts a well outlined document that describes in more detail how to really hone in on the appropriate query. It’s not specific to academic searches, but is still a valuable resource. It can be found here.
Identify Credible Sources
An academic search builds a content base for a project that will need to be credible and referenced throughout the project. Some academic searches are best served by a result set of primary sources (documents or physical objects produced during the original time of study). Others can be based on secondary sources (interpretations or analyses of primary sources). Regardless, its essential that a source’s credibility be verifiable after the project is complete. Frequently, it’s easy to tell if a primary source has high credibility while secondary sources require more analysis before credibility can be verified. It’s important to learn how to distinguish primary and second sources and quickly gauge credibility, otherwise maneuvering through a large result set will become impossible. SweetSearch has a lot of built in functionality to aid in this process that will be reviewed in more detail later in this chapter. Keeping a log of results is also good practice during academic research because it makes generating citations much easier as you complete the project. Many search engines now have tools to help with this as well and SweetSearch’s will be discussed in more detail later.
These are just a few general practices for succeeding in an academic search. They should be kept in mind throughout the rest of the chapter. With that out of the way it’s time to take a closer look at the various features of SweetSearch.
How To Use SweetSearch
Using SweetSearch begins similarly to any other search engine. The homepage features a simple layout with a search bar and a go button. Simply enter the query and you’re off. The nuances within SweetSearch are more apparent after performing a search. To demonstrate some of this functionality, research will be gathered on the topic of sorting algorithms throughout the next few steps.
Execute a Good Query
It sounds obvious and simple, but as mentioned in the ‘Academic Search Strategies’ section, this can be more difficult than one might expect. For the demonstration topic an obvious search might be [Sorting Algorithms]. When this query is executed on SweetSearch the results are not overwhelmingly positive. They can be viewed here. The first set includes links to pages within http://blog.findingdulcinea.com/webtech/, which is a blog by the same company that owns SweetSearch and most of the pages in question discuss how SweetSearch filters it’s results. Since the results are poor, the query must be refined. The two best ways to do so are either to include some search operators or get more specific with the topic.
First, the path of narrowing the search scope will be explored. In one of the results further down the page a specific search algorithm, ‘insertion sort’ is mentioned within a result. Perhaps this is the narrowed topic that will merit closer inspection. Adjusting the query to [insertion sort] produces notably better results than [sorting algorithms] did. All of the first 20 results come from a University page or a computer science research organization and all appear to have information specific to insertion sort. If none of the original results had mentioned this narrower sub-category, beginning with a Google search or even a Wikipedia search to set some bearings within the field of interest could be a useful strategy. SweetSearch even offers the option to execute a query on Google if the SweetSearch results aren’t satisfying. Simply toggle the radio button next to the search bar and the page will display Google’s results instead. For the [Sorting Algorithms] query that yielded poor results on SweetSearch, Google had a much more useful set:
Just like Google, SweetSearch includes a set of search operators to help fine tune queries. Going back to the original query, [Sorting Algorithms], it’s notable that in the set of poor results ‘sorting’ and ‘algorithms’ were not always a combined string. Applying one of the most basic search operators resolves this and greatly enhances the result set. The new [“Sorting Algorithms”] result set can be viewed here. Again, notice that virtually all the results come from Universities or science organizations. Exactly the kind of results one would be interested in viewing for this sort of topic.
Although it wasn’t documented on the SweetSearch website, I emailed the Dulcinea Media Help Staff and asked for a complete set of SweetSearch’s search operators. In the response, I was informed that SweetSearch actually uses Google’s technology to process its queries. As a result, any search operator that works in Google, should work the same in Sweet Search. In case you need to brush up on your operators, here’s some documentation covering the ones available with Google.
Navigate The Results
Now that some techniques for getting the best results with SweetSearch have been outlined, it’s time to explore SweetSearch’s functionality for exploring a result set. Here’s an example of one set generated during the [insertion sort] query and some points of interest are mentioned below:
- The main heading in blue, underlined text, provides the title of the page referenced by the result.
- Below the title, searchers are presented with a few lines of text from the page that contain the search terms, with the terms in bold.
- Below that, the URL to the page is displayed in green.
This section mimics many other search engines closely, but the content beneath it represents a SweetSearch distinction.
- Each cluster of text with a check-box to the left is a section of text taken from the page. This allows the user to gain even more insight before navigating away from their results.
- Clicking ‘More Results Available’ adds more text clusters from the page.
- Each search term is highlighted in a separate color.
- The check-boxes should be noted, but will be elaborated on in the next section of this chapter.
The text clusters in this section are quite a useful feature and there is even more to them than meets the eye. Clicking on one of these sections opens a pop-up of the result page and it jumps to the section from the page that contains the selected text. This allows a full on section-specific page investigation for each result without ever having to leave the result set. Here’s an example:
A page, by default, displays 20 results at a time. The top and bottom of the page feature a navigation bar, allowing the user to jump between pages of results. Navigation options include, ‘First’ to return to the first 20 results, ‘Prev’ to return to the previous 20 and ‘More results’ to navigate to the next 20. Unlike many major browsers SweetSearch has few advanced customizable settings, so if 20 results doesn’t seem ideal, there’s no way around it.
One other method of using SweetSearch is installing the SweetSearch browser add on. A large draw back of the feature is that it’s only available for FireFox 2.0+ and IE 7+. The world of Chrome, Safari and Opera users are out of luck. However, for those that can use it – simply click the large add-on button on the right side of the main page. It allows quick searches from the browser, even if the user hasn’t navigated to the SweetSearch page, much like similar browser search add-ons.
Use The Results
Another great aspect of SweetSearch are the built in options for managing results. This is where the previously mentioned check-boxes come into play. By checking the box next to paragraphs of note within interesting pages, the user has the ability to document or share results in a variety of ways. The console for these operations appears to the right side of the top of the result set. An image of the console can be viewed to the left of this paragraph.
As the description suggests, the console’s search bar allows the user to look for a specific text string within the result set produced by their query. Quite a useful feature for identifying a specific result (although it does essentially the same thing that a ctrl + f or cmd + f would do on most modern systems).
The icons below the search bar allow for result management. Except for the first check-box icon, they all operate exclusively on the page content whose check-box has been selected by the user.
- The check-box icon marks the check box of every item within the page.
- The star icon bookmarks selected results.
- The graph-node icon opens a modal box with content sharing options.
- The document icon allows the user to add content to a Google Document.
- The trash can icon removes the selected results from the result set.
When sharing content on another service the modal box offers the options below for sharing. The interface and options for adding to a Google Document is visible below as well. Links to these services can be located in the appendix to this chapter:
The variety of social, research and email options available for sharing results is impressive. Even if a user isn’t familiar with all of them, it’s unlikely that he or she doesn’t frequently use at least one or two of the available options. Furthermore, the options regarding how to customize content in a Google Doc, make it a very valuable feature. The ability to append to an existing Doc or create a new one allows for a very organized system for gathering content.
Many of these advanced features are powered by yolink – which is, as they state on the product information section of their website,
…a next-generation search technology that helps you get things done quickly. yolink excels at multi-step or complex searches and dramatically increases the effectiveness of search across your favorite sites and services. yolink takes unwieldy results and surfaces key information so you can focus on getting things done.
The fact that the technology isn’t produced in-house by Dulcinea Media doesn’t make it any less valuable, but it does cause it to have a non-customized feel and some steps, like authenticating yolink to access a social service can require a few clicks that steer the user away from his or her results temporarily. Since yolink provides a lot of SweetSearch’s most exciting functionality, they deserved mention within this chapter.
Comparison with Google Web Search
It should be clear after reading through this chapter that a non-academic search or a query to obtain a quick fact isn’t well suited for SweetSearch. Looking for movie times, planning a trip or finding a band’s discography should probably be left to Google. That shouldn’t come as a surprise since SweetSearch is categorized as an academic search engine. Clearly, its main utility is geared towards academic research. More useful insight as to when SweetSearch should be used can be gained when comparing different academic searches.
Perhaps the biggest determinant of whether to go Google for an academic search is the scope of the topic in question. SweetSearch’s highly filtered base of 35,000 sites can greatly increase the quality of its results, but as demonstrated in the ‘Instructions’ section, if it’s early in the research process and the scope has yet to be focused – SweetSearch can prove ineffective. This can be mediated via its search on Google function, but using Google on SweetSearch prevents the user from access to main of Google’s advanced search features so many times it’s more practical to simply go to Google for the search.
This leads to another significant determinant for SweetSearch use. One of its largest short-comings is its lack of advanced searching features. Filtering by time period, looking for images and many other useful tools aren’t available within SweetSearch. While its select pool of sites to search from usually provides a degree of quality in its content, a sophisticated Google user can often use the aforementioned advanced features to obtain specific enough results to overcome the negatives that its vast pool of sites can bring to a result set. This fact, in combination with the previous paragraph, make the argument that extremely broad and extremely specific searches are frequently best left to Google – leaving the middle ground as the sweet spot for a SweetSearch.
Although it shouldn’t be used as one of the most significant factors when choosing what service to use for a search, it’s worth noting that the aesthetics of SweetSearch fall far behind those of Google. Many of the buttons on the page look like they could use significant anti-aliasing work on their pixelated rounded edges. The page seems to have very poor rules in place for how to orient itself in different screen resolutions and the results set is rarely aligned in the center of the page. As previously mentioned, it’s frequent that SweetSearch’s content pop-up tool will overflow from the browser window – which can cause some hindrances to its usefulness in addition to the negative impact it has on the page aesthetics.
Aside from its expert-pruned base of sites (which, as demonstrated previously, isn’t always a positive), SweetSearch’s biggest benefit over Google stems from the thick clusters of text within its results, the ability to browse a pop-up of a page without leaving the result set, and its wide range of tools for marking, sharing and saving result content. All together these tools allow for quick research gathering without ever leaving the SweetSearch page. For quickly putting together a large list of high quality academic resources during research – SweetSearch reigns supreme. Although it’s definitely designed to accommodate a niche searching style, the times when it’s appropriate to use should now be apparent.
References Throughout The Chapter
Professor Moore’s Academic Search Lesson: http://bit.ly/WS9DRF
Dulcinea Media: http://www.dulcineamedia.com/
Berkeley’s Search Guide: http://bit.ly/YegfG
Google Search Operators: http://bit.ly/snvf1l
Services Offered For Content Sharing
Yahoo! Mail: http://mail.yahoo.com/
About The Author
Dan Campbell is a senior at The University of Michigan’s Ross School of Business. In addition to pursuing a degree in Business Administration, he has also completed a minor in Computer Science. He currently works for the university doing web-development for UM ITS’s Print Team. After graduation, he will be moving to Chicago to work as a consultant for Navigant, Inc.