I am posting this in full with a link because it is so interesting:
AskEraser: Privacy Potential by Greg R. Notess Posted On January 3, 2008
As web searching becomes ever more common in all aspects of our lives, the issue of our online privacy, or lack thereof, is an increasing concern. Searches and visits can be, and usually are, tracked and logged by any website. The information recorded can include a user’s IP address, the previous URL visited, the browser and operating system used, information in cookies set by that site, and search terms used. While none of this information can necessarily identify a specific user, the August 2006 release of AOL search data (see the NewsBreak at http://newsbreaks.infotoday.com/nbreader.asp?ArticleID=17374) showed that some personally identifiable information may be extracted from just a series of search queries.
On Dec. 11, 2007, Ask.com (http://www.ask.com/) introduced AskEraser as one way to help combat concerns with searching privacy. It gives searchers the ability to choose whether or not to preserve their own privacy when searching Ask. By default, it is not turned on, but the AskEraser link is featured in the upper right-hand corner of Ask pages. Users can click that link for the option to turn AskEraser on. AskEraser is available on the main U.S. Ask.com site and at the U.K. version (http://www.uk.ask.com/).
While most search engines (including Ask) have previously stated that they will not keep search log data for longer than 18 months, AskEraser gives users the ability to have their data deleted sooner than that. Patrick Crisp, a spokesman for Ask.com, said that "with AskEraser, we’re in a very real sense laying down a strong industry ‘marker’ that we expect others will follow."
With AskEraser turned on, all Ask cookies are deleted from the browser and one new cookie is set, which only tells Ask that AskEraser is enabled. "Within hours" of a search being run, AskEraser promises that the search activity will be deleted from its log files. Once that deletion has occurred, Ask no longer has any record of those specific searches. The information deleted includes the following:
The IP address of the computer being used
Cookies containing user ID or session ID information
The text of the search query
Older search activity from before the launch of AskEraser or from searches run without an enabled AskEraser will be retained in the logs for 18 months after the search was run.
This is not a simple process. Note that since the concept was first announced in July (www.irconnect.com/askj/pages/news_releases.html?d=123324), it took Ask months to implement it. Part of the difficulty is that logs are automatically recorded for all searches, and getting specific search activity deleted requires additional actions from the search company. This is also why the deletion from the log files takes several hours and does not happen right away.
The Exceptions
Of course, there are exceptions. By no means does AskEraser create a completely anonymous web browsing session. All other servers beyond Ask’s can still track visit information. In addition, some information is passed to Ask’s advertising partner, Google, so that the text ads displayed are connected to the search query.
At first glance, this seems to defeat the whole purpose of deletion of search activity. If Ask deletes it but has to pass all of that information to Google’s ad servers, what privacy is gained? The difference is that all of the information that Ask would gather is not necessarily passed on to the Google ad servers. Unfortunately, Ask cannot comment on the specifics of which data is transmitted to Google. Even so, it is obvious that it would need to pass the query itself to be able to get ads connected with the query keywords. It also seems unlikely that there would need to be any passing of IP address or cookie information. Until either Google or Ask is willing to release more information about what data is passed, users should probably assume that Google ad servers are gathering at least some level of data, but not necessarily more than that which would be gathered from visiting any site that displays Google ads.
Two other unusual exceptions that Ask mentions in its About AskEraser page (http://sp.ask.com/en/docs/about/askeraser.shtml) are for critical technical issues and for a legal request from law enforcement. In the event that search activity data is need "to solve a critical technical issue … search activity data may be retained for a longer period. At the time of technical resolution all search activity data of AskEraser users that was retained will be deleted." As to the legal requirements, "even when Ask Eraser is enabled, we may store your search activity data if so requested by law enforcement or legal authority pursuant to due process." Both of these situations could occur at any other website. At least Ask is upfront about stating the possible exceptions.
AskEraser has also been criticized by the Electronic Privacy Information Center and other privacy advocates (see www.epic.org/privacy/ask/EPIC_%20AskEraser.pdf) for several perceived flaws, which the groups see as correctable. The complaints include the use of a cookie to keep AskEraser enabled and the inclusion of a time stamp within that cookie. Also, in the exceptions for court-requested tracking, Ask should alert that user that AskEraser is not functioning.
One other cookie-related issue relates to Ask users outside of the U.S. and the U.K. Such users can go into the Ask preferences to specify that they wish to use Ask.com rather than another international version, but if they then turn on AskEraser, it will erase the cookie that remembers that preference. So, AskEraser does not appear to be easily available to those users until it is enabled on the other international sites.
The Privacy Alternative
The recent emphasis from several search engines on searcher privacy in the last few years is in contrast to the opposite recent emphasis on search histories and personalized search. Ask introduced MyJeeves (now known as MyStuff) in September 2004 (http://newsbreaks.infotoday.com/nbreader.asp?ArticleID=16364). Other search engines soon followed (http://newsbreaks.infotoday.com/nbreader.asp?ArticleID=16219) with a variety of search history features.
No numbers are available for the percentage of searchers who log in and enable the tracking of their search histories. Nor is Ask sharing any numbers of active AskEraser users. Based on the time that each system has been available, it is likely that many more searchers use Ask’s MyStuff than AskEraser. Crisp notes that Ask is "happy with the usage of MyStuff" and that it "know[s] that users who are using it really love it."
AskEraser and MyStuff demonstrate how a search engine can offer users options in both directions. Those who wish to have more information tracked and recorded for their own use can log in, enable MyStuff, and be able to view their own prior searchers. For those more concerned with privacy, "AskEraser is designed for that subset of our users who would like more control over their data," says Crisp.For searchers who have personal or business reasons to keep their searching as private as possible, AskEraser offers a tool to help keep such information out of search engine logs. While it does not yet offer full anonymity of searching, it is one more new attempt to give users more choices in protecting their online privacy.
Greg R. Notess is the internet columnist for ONLINE and the author of Teaching Web Search Skills (www.notess.com/teaching) and SearchEngineShowdown.com.
Clearly there is no such thing as total privacy today. I am not suspicious by nature so I do not fear people knowing my online activity. It is quite tedious. Between my occasional gaming, online reading of library topics, and visits to Amazon and Ebay, I would not be the fodder of scandal sheets. I do represent information however. I am a consumer of goods. I am an information consumer too. Google likes to follow trends, so I am a water molecule in the bucket, so to speak.
There are two basic kinds of search engines.
The first are web crawlers. Google is the most successful of these. Each time a site is visited your IP address is recorded along with recent history of others sites visited. Google has a series of programs that gather this information from sites and records the trends. It has a listing based on the sites use. The more the sight is visited the higher it is on the results list, or indexes. When you or I uses Google we type in search terms that are used to pull up files (sites, that is their URL'S).he Its kind of a hit or miss. It is a crude way of getting results at times however. Tens of thousands of sites are ordered according to their popularity, and based on the terms you put in.
The other big kind of search index is a directory that is based on human indexing. Squidoo is a good example of this, as well as Stumble Upon. A site owner enters a description of the site and its content. Listing means nothing. What is important is the description of it. Sites that are well designed and have excellent info that is updated on a regular basis often get reviews of it and thus better coverage.
There are also sites that use a combination of these techniques. This was very common in early search engines. For more information on search engines and how they work, I have provided some extra links below.
Privacy may be better thought of as a contract. When we log on through a provider we are agreeing to its terms of use. When we visit a site the same applies. By my way of thinking it is absurd to think that we can enjoy the same level of privacy that we do in our homes. When I was a kid I had a brother and sisters. The only privacy I ever got was in the bathroom. I also think that as Americans, we have a rather inflated concept of human rights at times. We live in what is defined as a democratic state. The truth, however is not so black and white. Like the privacy matter we have shades of rights, and with that many responsibilities. What attracts me most to public libraries (my chosen field) is that they depend on democratic principles to operate in the interest of the public. This of course does not mean we can offer our patrons everything they ask for. That is impossible. The same is true of the Internet.
As the article above suggests, we can afford levels of privacy, but invisibility is not one of them. Its in the contract. And that is the way it is. It is left for you decide, and live with the choices you make.
Any thoughts?
http://newsbreaks.infotoday.com/nbReader.asp?ArticleId=40526
http://searchenginewatch.com/showPage.html?page=2168031
http://computer.howstuffworks.com/search-engine.htm
http://www.webopedia.com/DidYouKnow/Internet/2003/HowWebSearchEnginesWork.asp
http://www.lib.berkeley.edu/TeachingLib/Guides/Internet/SearchEngines.html
Email Greg R. Notess
No comments:
Post a Comment