Manfred Kuechler


Finding a "persistent URL"

Last update:  3 Jan 2007

This document assumes that you are familiar with the content covered in the more basic document titled "Finding Information on (Published) Social Research Online".

What is a "persistent URL"?

If you already know what a persitent URL is, skip forward to the section on how to find one in various databases.


Typically, when we are surfing the web and find a page of interest, a page we may want to go back to or which we may want to send others to, we save the address of the page as a "bookmark" (Netscape/Mozilla/Firefox) or as a "favorite" (MS IE). The bookmark/favorite stores the address of the page -- more technically known as URL or "uniform resource locator" -- and we can easily revisit the page without going through the whole search process again. And the web address or URL stored in a favorite/bookmark is just what we see in the address or location box of the browser we use. Here is an example, shown with both Firefox and MS IE:





These URLs are "persistent" -- meaning they will still work if you use them in a couple days and possibly even in a couple of weeks, months, or years. Of course, some times the owners of a web site (in this example, the Census Bureau) reorganize their site and even a persistent will not work forever. Also, while there still may be a page at this address, the contents of the page may change -- but this is another issue.

In contrast, when you search a data base (like EBSCO, Lexis-Nexis, etc.) the address which you see in the address/location box is "transient" -- it may work only for the current browser session or maybe a day or two. After that, the address becomes invalid. Here are two examples of such transient or non-persistent URLs:



Here, you don't even see the whole URL which is:
http://web.lexis-nexis.com/universe/document?_m=081596080953636f3692434e025b5008&_docnum=9&wchp=dGLbVlb-zSkVA&_md5=1b4a39ec4e5049153200994c38debd25
This link will only work for a very limited time span.



Same here. The full non-persistent URL is
http://web19.epnet.com/citation.asp?tb=1&_ug=sid+AC303A8F%2DE516%2D4134%2DB154%2D79AFD4574BDA%40sessionm
gr4+dbs+aph+cp+1+7621&_us=frn+1+hd+False+hs+True+cst+0%3B1%3B2%3B3%3B4+or+Date+fh+False+ss+SO+sm+ES+
sl+0+dstb+ES+mh+1+ri+KAAACB2B00275329+A819&_uso=hd+False+tg%5B2+%2D+tg%5B1+%2D+tg%5B0+%2D+st%5B2+%2D
+st%5B1+%2D+st%5B0+%2Dillegal++immigrants+db%5B0+%2Daph+op%5B2+%2DAnd+op%5B1+%2DAnd+op%5B0+%2D+mdb%5
B0+%2Dimh+D21D&cf=1&fn=1&rn=2
Again, this URL will work for a limited time only.

So, whenever you are asked to document the source for some information found online, it is crucial that you supply a persistent URL rather than mechanically saving a bookmark/favorite or copying what you see in the address box.


How to find a persistent URL

Unfortunately, there is no general procedure to determine persistent URLs, each web site (data base) is different. Some data bases  do not offer persistent URLs at all and those which do offer  put them in different places and make it more or less convenient to secure them. So, the instruction differ from data base to data base and we can only cover a few data bases here. However, once familiar with the concept, you should be able to check other data bases for the existence of persistent URLs. Below are detailed instructions for:
Keep in mind that even persistent URLs may not always work from off-campus. When off-campus, you may want to establish access to the particular database first (by one of the three access methods) then use/test a persistent link. Also, there have been problems with persistent links to the CSA database in the past even from Hunter computers; hopefully, the vendor (CSA) has now fixed these problems for good.

Lexis-Nexis

In late August 2006, Lexis-Nexis finally started to provide stable ("persistent") URLs to specific articles in its database. It still takes a bit of work to construct such URLs, they are not posted with the article for convenient copy-and-paste and/or import into your EndNote database, but it is still a major step forward. The full details are in this 7 page document available at:
http://www.lexisnexis.com/documents/academic/academic_migration/LexisNexis%20Academic%20URL%20API%20Specification%202006_09_15.pdf

But here is a short summary:
The basic format of a persistent URL to an article in the Lexis-Nexis database is:
http://www.lexisnexis.com.proxy.wexler.hunter.cuny.edu/universe/api/version1/sr?csi={csi_number)&sr={search_terms}

Note that I have added the string ".proxy.wexler.hunter.cuny.edu" so that such an URL will work from off campus as well. However, an off campus user will be prompted for his/her Hunter e-mail address and PW to be authenticated by the Hunter EZ proxy server.

The two parts enclosed in {   ...   } vary with the article; these are the parts you must construct yourself.
A. CSI (Constant Source Identifier) number.
The number identifies a particular source (say the NYT) and it is the same for all articles for this source. E.g., the CSI for the NYT is 6742. A complete list of all CSI numbers is available at
http://www.lexisnexis.com/documents/academic/academic_migration/LexisNexis%20Academic%20A%20to%20Z%202006-08%20interim.zip
Simply unzip this file to extract the one .xls (Excel) file contained in this zip file to look up the CSI for any particular source (newspaper). Currently, there are over 6000 sources.

B. The second part consists of search terms which should uniquely define the article, e.g., the date and the headline for a newspaper article. Here is a concrete example:
http://www.lexisnexis.com.proxy.wexler.hunter.cuny.edu/universe/api/version1/sr?csi=6742&sr=headline(Metro Briefing New York: Manhattan: Raise For Cuny Chancellor) and date is September 26, 2006

Note that Lexis-Nexis is working on a new user interface to be released in July 2007. Most likely there will be some changes in the way  these stable URLs have to be constructed as well.  So, if these instruction cease to work, check this Lexis-Nexis web site for more information.

Custom Newspapers (and other InfoTrac/Gale databases)

Click the title of interest on the results page:




Instead of clicking on the link labeled "Source Citation", you can simply scroll down the page. The "source citation" is at the bottom of the page:



This is the persistent URL, not what you see in the "address/location" box at the top of your web browser window.  As with other databases, you may want to insert the string "proxy.wexler.hunter.cuny.edu" so that the URL will work from a non-Hunter computer:

http://find.galegroup.com.proxy.wexler.hunter.cuny.edu/itx/infomark.do?&contentSet=IAC-Documents&type=retrieve&tabID=T004&prodId=SPN.SP00&docId=CJ149749951&source=gale&srcprod=SP00&userGroupName=nysl_me_cuny&version=1.0

EBSCO ASP (Academic Search Premier)

Finding a persistent URL in EBSCO is easy -- once you know which steps to take.  Start with the "results page" and click the "title link":


Of course, in a real search you want to first make sure that this is an article of interest and therefore check out the full text (here by clicking the "HTML Full Text" link) first. The persistent URL is on the next page, but sometimes you may have to scroll down a bit. The two following screen shots leave out (most of) the "abstract":




So, it take just one click  (and maybe a bit of scrolling) to get to the persistent URL. In EBSCO, all persistent URLs have the same format, the only things which changes is the "accession number (an)" at the end. Note that this number is already included on the results page.

With the new EZ proxy service offered by the Hunter library (when accessing the data base from off campus), the persistent link may be modified. So, in this example, the persistent link may appear as:
http://search.epnet.com.proxy.wexler.hunter.cuny.edu/login.aspx?direct=true&db=aph&an=18019198

This persistent link will only work when used from the Hunter campus or via authentication with a Hunter e-mail account and password. So, if you want to communicate with people outside Hunter, it is preferable to report the persistent URL  as on the original EBSCO page, not the modified version:
http://search.epnet.com/login.aspx?direct=true&db=aph&an=18019198

A final caution: You need to make sure that the article is available in full text independently. Otherwise, you get a persistent URL to the bibliographic details and/or the abstract only.
 

CSA databases (like SAGE Sociology Collection)

The  SAGE Sociology Collection is just one of many data bases offered by CSA which all use the Illumina user interface. So, the procedure for getting persistent URLs for other CSA data bases is quite similar. Again, we start from the results page and let us assume we have already determined that the first article listed is of interest (again, by checking the full text first):


This will bring up the following page:


Depending on whether you have set up your browser to block "pop-ups" or not, you will go to the relevant page directly or you will get this intermediate page:


This will bring us to the page where the persistent URL can be found -- though it is called something different:


In contrast to EBSCO, the persistent URLs in CSA are quite long; the screen shot above show the first part only. The full URL for this example is:
http://www.csa.com/ids70/gateway.php?mode=pdf&doi=10.1177%2F0011392105052716&db=sagesoc-set-c&s1=
643ce13fbe1bebb0d8a881b95a527a7f&s2=dbd6236f898354ae62d6ebaa2e4ea0db

JSTOR

In JSTOR a persistent URL (called "Stable URL") is shown directly in the search result list:


Note that a (stable) URL is shown only  if the full text is available in the JSTOR database. Otherwise, follow the "Link to Article" link and retrieve the full text in another database -- and get a persistent URL there.


ScienceDirect

Again, here is an example of a list of search results:





When you move your mouse over the link labeled "doi: ....", you will see a persistent URL displayed in the bottom-left corner of your browser (you can always see the actual URL underlying a link this way). To copy the URL, right click the link and select "Copy Shortcut" (MS IE) or "Copy Link Location" (Firefox), then paste the copied URL into any document. In this example, the URL is:

http://dx.doi.org.proxy.wexler.hunter.cuny.edu/10.1016/j.econedurev.2006.02.005

The "proxy.wexler.hunter.cuny.edu" part only appears if you access this database via the Hunter proxy server. If you access this data base from a Hunter computer, you bypass the proxy server and the URL would simply be:
http://dx.doi.org/10.1016/j.econedurev.2006.02.005
However, when you use the shorter URL from off-campus, you will get the abstract only, not access to the full text. So, it is preferable to use to include the "proxy.wexler.hunter.cuny.edu" part -- unless you are communicating with someone not affiliated with Hunter.  (Note that I show this string in color for emphasis only; the color is immaterial to the functioning of the URL.)

Note that DOI ("digital object identifier") links will never change even if the vendor (here: Elsevier) decides to rearrange its database.  So, these links are preferable, but not (yet) available in all databases. More on the DOI system.

Still, here is an alternative way to get a persistent URL (when using bibliographical software like EndNote or WriteNote). Click the "Export Citation" link:



Make your selections as shown in the screen shot above, then click the "Export" button.

What happens next depends on whether or not you have EndNote (properly) installed on your computer and -- possibly -- which browser you use.
When you are using bibliographical software, you can easily produce a list of all (selected) citations following a specific citation "style" (like "APA 5th" style). Whether or not the persistent URL stored in your EndNote/WriteNote "library" gets included in such a citation/reference list depends on the "style" you choose.  Many styles (including APA 5th) do not include URLs to the full text of published journal articles. You would have to add these URLs manually or -- works in EndNote, but not in WriteNote -- you could edit a style to include display of the URL.