One way of looking at online research is as a project that starts with searches and ends with summaries. The online researcher searches for information to meet the needs of their client and produces a summary report for the client. Information aggregators help in the production and display of summary views and are therefore a natural tool for the online researcher. What kinds of information aggregator features would be most helpful for online researchers?

First, it would be useful to have help in collecting the information for the research project. Tools to capture links and selections from Web pages and to add notes would be useful. This information needs to be organized into a structure. Information aggregators can do this by providing a personal knowledge base or Web directory. The project is added as a new category in the knowledge base and the information collected can be organized into categories within the project. It would be useful to have a browser extension to help capture this information while browsing.

For example, the researcher does a search and is going through the results. Some of the links prove useful and can be captured into the knowledge base. On some pages, a particular selection from the page proves useful. This also can be captured. This captured information can be annotated and organized into categories. As the project progresses more information is captured, annotated and categorized. It would also be useful to be able to enter text and tabular information directly into the project within the same knowledge entries in the knowledge base. This is all part of the process of building up a summary view from the vast amount of information on the Web.

Of course, an online researcher may do similar projects for different clients, so it would be useful for them to be able to build up their own knowledge base of commonly used resources over time. They can go to these resources again and again on different projects. Notes can be included to remind them on the ways to use the resources. It would be useful to be able to copy information from these common areas into a particular project to reuse the work.

Since these sites change over time, it would be good to be able to get summaries of what is happening on the sites and be able to click through to see the details. Information aggregators can help here by providing personal pages containing a variety of modules that provide automatically updated information on what is new on a site.

Many sites provide summary headlines of what is new on their site as RSS feeds. An obvious use is for news headlines, but this is also used for more specialized information. An information aggregator should come with many built-in feed definitions and allow the researcher to easily add others that are specific to their research interests.

Many Web sites provide "free content" as an HTML fragment that can be added to a page. These serve a similar function as RSS feeds, providing an automatically updated summary view of new information available on the site. An information aggregator should include a number of these and allow the researcher to add new ones.

Another useful feature would be to allow the researcher to provide keywords that the information aggregator will use to periodically do a search on their behalf. The top search results can be shown on their page, and they can click through to see the complete results. This would allow them to monitor new sites of interest for keywords they often search on. It should be possible to set this up to only show the new results.

It would also be useful to be able to monitor commonly used sites to see if new content has been added to the site.

Finally, information aggregators should provide help in packaging the results of the project for the client. Export tools should be provided to allow the contents of the knowledge base to be edited in a word processor, for example. It should be as easy as possible to get the information out of the information aggregator into a form that the client would like to see it.

This area of packaging also presents an opportunity. Professional online researchers are always looking for ways to add extra value in the package they give to their clients. One idea is to provide the clients with "active" reports, that is, reports that automatically will provide updated information after they are delivered. Static reports produced with a word processor are a snapshot of the information at a particular point in time. An active report provides this same snapshot view but also has elements that can change over time from feeds.

These active reports will require the client to have their own tool to read the reports and to access the updated information. Some clients will have their own information aggregators. The researcher can then export parts of their knowledge base and the client can import it. Some information could be exported in a standard format like RSS so that the client can use any RSS reader they like. For more extensive active reports, the client will need some kind of package reader. This reader should be free, similar to how the Adobe Reader allows the reading of PDF files.

So the researcher could create an active report or information package using their information aggregator. They would deliver this to their client with a free information package reader that would allow them to receive the updated information. The information package could contain the knowledge base generated for the project along with pages that automatically update. The online researcher in this case becomes an "information package developer".

For clients who do not want to commit to a particular package reader, some of the active content can be exported as HTML pages that could be read by their Web browser and some as RSS feeds that could be read by any RSS reader. But to get the most active versions they would need to use a particular reader.

Many clients will prefer to continue to get their static reports. The information aggregator will still be there to help the information professionals produce these static reports as well as the active reports.

This article presents a target view of features for information aggregators that could be useful to online researchers. Some of these features are available in current products. Others will evolve over time. The bottom line is that information aggregators can be a great tool for online researchers.

About the Author
Ron Tower is the President of Sugarloaf Software and is the developer of Personal Watchkeeper, an information aggregator supporting a variety of ways to summarize the Web.
http://www.sugarloafsw.com