What model was used for MathSciNet usage statistics? The MathSciNet usage statistics reporting system follows much of the COUNTER Code of Practice, release 1, (December 2002) for databases. For more information about the COUNTER Code of Practice, please visit (http://www.projectcounter.org/code_practice.html#start).
When are usage statistics made available? Monthly usage statistics are posted within two weeks following the end of the month. Email notification is sent to the contact email address that is attached to the subscription. Usage statistics for previous months and/or years can be accessed at any time.
How is access to the usage statistics reports authenticated? Access authentication is done via Internet Protocol (IP) address.
What is the definition of a session? A session is a successful request of an online service. It is one cycle of user activity that typically starts when a user connects to the service or database and ends when that activity is terminated either explicitly (by leaving the service through exit or logout) or implicitly (a default timeout due to user inactivity). The timeout period is 30 minutes of no activity.
What is the definition of a query? A query is a specific request for information from the database. Double-clicks and images are not counted as queries.
Why can't I see usage statistics for my site/consortium for periods before January 1, 2004? Prior to January-2004, the mechanism for collecting, reporting and presenting usage statistics was significantly different from this reporting system and the two are incompatible. If you need usage statistics for periods before January 2004, please send email with your request to email@example.com. Please remember that those older statistics reports cannot be used to compare against these new ones.
Why can't I get a further breakdown of my site's usage statistics by either IP address or named sub domain of department/institution? Privacy issues prevent us from reporting by IP address. The inconsistency in named sub domains, the number of changes made to these and the amount of time involved in breaking access down by these, prevent us from accurately reporting usage at a finer breakdown than by site.
I'm the contact for a consortium and some of the participating sites seem to be missing from the reports. Why? If the IPs registered under a single institution subscription include those for multiple campuses, statistics for all campuses will be bundled under those for the main campus. For example, Univ. of Herbert has a main campus in Hoover and several smaller satellite campuses in other cities, but has registered all IPs under the University of Herbert at Hoover. All usage of the database will be reported under the Hoover site. If you have registered IPs that cover more than one site under a single subscription and wish to have them reported separately, please contact Customer Services (firstname.lastname@example.org) and provide them with the appropriate information to make this finer distinction. If sites appear to be missing from your usage reports, please send email to email@example.com.
Why aren't full text downloads counted? MathSciNet is a secondary database containing reviews of published items. While there are a large number of links to full text articles, these articles are downloaded from the publishers' sites.
Why aren't turnaways reported? A MathSciNet subscription provides site-wide unlimited simultaneous access to the database.
I have a question not addressed above. Whom do I contact? Please send email to firstname.lastname@example.org for any questions or concerns regarding the MathSciNet usage statistics reports.
"Not everything that can be counted, counts; not everything that counts can be counted." -- Albert Einstein
Copyright 2004, American Mathematical Society