Finding valid and timely information on the Internet is a difficult task.  It requires a structured process with specialized techniques and skills.

There are three (3) traditional methods to verify the validity of information:

	First, if it is written and/or issued by an authoritative source such as the federal government or a reliable organization, it is generally accepted at face value as having validity.
	Second, if it is authenticated as part of an editorial or peer review process by a known and established publisher, it is generally accepted as reliable.
	Third, if it is evaluated by experts, reviewers, or subject specialists/librarians as part of collection development, it is generally accepted as authoritative.

Since almost anyone can publish anything on the Internet, using the traditional methods of evaluating information validity are not very useful for the Internet. Information on the Internet may be important and interesting, but it is not necessarily reliable or current.

Here are some simple checks which may be used to help evaluate the validity of Internet information:

	the expertise and reputation of those connected with the website
	evidence of lack of bias (specially important when seeking information on potentially contentious matters)
	evidence of equal emphasis (ie, no omission of information) towards all aspects of the search topic
	check to see whether the source is a moderated or an unmoderated list
	check online directory sources for affiliations and biographical information
	check and compare this source to other sources, both similar and different ones
	check online for other works by this author
	check online reviews of the source

Here are a few techniques to determine if a website contains timely information:

	check the "Last revised" date on a website. This is usually found on the first screen. Note that this may only indicate the last time a change was made to a particular part of the website and that it does not imply that all the content was updated on that date
	look to see how recent are the publication dates of the documents referred to on the site
	look out for dates quoted within text and make an estimate of currency
	look for a "Forthcoming events" list where all dates should be in the future, or in the very recent past

If no explicit date information is available to determine the timeliness of a website, you might try some indirect methods:

	look for regulations or laws mentioned in the site and find their dates of introduction.
	look for the names of institutions or Government departments. These do change name and can be used to give a date range
	look for the names of personalities involved in the topic which can show timeliness
	simply look for reference to recent news items which will date the site's information

Search engines are constantly trying to improve the algorithms that can yield the best matched keyword results.  However, there are many limitations to an algorithmic search approach for indexing the Internet.

Given the size of the web, many of the publicly available websites have not been indexed by search engines.  Each search engine captures a different percentage and nobody knows the percentages.  However, it is safe to assume the percentage is probably not very large.

There are many databases on the web which search engines do not index.  Since these databases generate dynamic pages based on a user request, conventional search engines cannot index them.  For instance, the US Patent Office can only be searched by a user query into the site itself.

Some sites block search engines, usually due to copyrighted material available to members only, while other sites use search engine optimization (SEO) techniques to increase page rankings.  Neither extreme helps the search engine user to find what he wants.

Some types of content are beyond the reach of the search engine because the media.  For instance, search engines may not be able to access content formatted for Macromedia flash, Adobe PDF, JPEG/GIF, an audio/video stream or podcast.

Knowing the context of a search is a fundamental problem encountered in any algorithmic search model.  Furthermore, even if the context of a search is known, most web pages don’t present data indicating what context they intend to address.   This results in search engines deliberately presenting a diverse set of results which is not as useful to the user.  In general, most user queries don’t fully explain the context of their search request.  For instance, a user searching the term “diabetes” could be looking for one of the following types of information:

	a patient looking for treatment information
	a doctor looking for research information
	a drug designer at a pharmaceutical company looking for drug trial data
	a medical authority looking for related regulations

A number of studies on web searching reveal why it is important to have a professional searcher perform this work:

	Experienced users are more systematic in their search planning and execution.  Novice searchers adopted more haphazard strategies and were more affected by the external presentation of the information presented.
	Many web users do not use advanced or Boolean search features and interact with the Web as if it was another person.
	A search engine user’s level of experience and cognitive style affects search performance.
	Over time Web searchers adopt browsing as opposed to searching strategies.

Just like any type of specialized work, you need trained professionals to perform the work in an effective and focused manner.  Expecting in house employees to have the needed search expertise and focus to perform quality searches is not realistic.  Not only do employees lack the search skills, but they are susceptible to the human weaknesses while “surfing” the web for information.

Cyber Search uses highly trained search specialists to provide a turn key Internet research service.  Where needed, Cyber Search specialists are also highly trained in the required specialty (ie, business, medical, legal, etc).  By using search specialists and structured queries, Cyber Search offers a complete turn key outsourced search service.  Unlike the search engine model (even a specialized search engine), the Cyber Search customer does not have to be familiar with the specialty nor the best search techniques for that specific search engine.  By intimately understanding the terminology and requirements of a discipline, Cyber Search specialists understand exactly what information is needed and where to find it.  Cyber Search customers can take full advantage of the powerful combination of search skills and specific industry knowledge.

Cyber Search specialists are trained the most advanced search techniques and methodologies to insure the optimum results for our customers.  Here are a few of the simpler techniques Cyber Search specialists use:

	Multitasking Simply by having scripts which simultaneously enter searches specifically designed multiple search engines, Cyber Search can significantly reduce search time.  This technique may reduce the search time by half.  Also, it makes the speed of the internet link needed less important.
	URL Analysis A web address holds so much more information than a country code or organization type.  It may also tell us the publisher, type of author, information format, date, likely publishing paradigm, and more.  This technique allows us to move more rapidly through the internet, guessing the contents of possible destinations as we search.
	Field Operators By accessing advanced search features thru fields (ie, link:, etc), specific searches may be crafted more easily and may be automated via scripts.  Of course, this reduces search time.  This technique is most useful when you want to search horizontally thru the Internet.
	Manipulate the URL Once a site of interest has been found, the URL may be manipulated to find more interesting information horizontally or vertically in a website.  For instance, the URL may be hacked by cutting off directories and guessing page names.  Also, a search engine could be asked for all the pages found in a given directory.
	Narrow search Instead of jumping into a search with a few keywords, we can be more precise before we start searching.  We can narrow down the likely author, likely publisher and likely access points of the information we seek. We can use the length, age and format of the information we hope to find.