These release notes contain information about IBM InfoSphere
Global Name Management, Version 5.0, such as installation notes, known
issues, fixed problems, and usage notes.
Note: For the latest version of the release notes, see the online
version in the
information center at ibm.com or the separate
release notes HTML file that accompanies the installation media and
is separately downloadable at the
product Support portal.
Upgrading to IBM Global Name Management
version 5.0
The installation program writes new configuration
files to install_path/data/*.config.template,
where install_path is the full
path of the directory where you installed IBM InfoSphere Global Name
Management. Existing configuration files are preserved and are not
overwritten by the installation program.
List of fixes and corrections
The
following list describes some of the functional improvements and corrected
problems that have been included in version 5.0:
- Fixed a problem where the exact matching logic for names was overriding
the built-in penalty for initial-on-initial matching.
- Added screening to weed out names that are over 128 characters
because these are too large for the system.
- Fixed a problem where periods were being removed instead of being
replaced with a space. This could have caused missed results due to
a lack of initial matching.
- Improved display and functionality of the GNM-NA user interface
on Internet Explorer 8.
- Improved handling of names with non-standard characters in AnalyzeForSearch
to avoid creation of multiple names when not needed.
- Added ability for NameWorks AnalyzeForSearch to suppress categorization.
- NameWorks AnalyzeForSearch now identifies alternate parse records.
- Improved level of information being returned in errors from NameWorks
Embedded Search.
- Improved parsing of names with tokens that are common Hispanic
names both as surnames and given names.
- Improved handling of business names with spelled-out numbers (for
example, “Eighteen”) and Web addresses.
Known issues when using the product
Review
the following information before installing and using IBM InfoSphere
Global Name Management version 5.0.
- Custom culture overrides
- Custom culture overrides should only be done to Global Name Management-supported
name phrases. It is strongly recommended that you contact IBM Services
for assistance with creating new culture categories and customizing
the current ones.
- Name Preprocessor full name parsing and processing
- If a full name is entered in the SN field and the doFullName tag
is set to True, the full name is parsed and the original (full) name
is not returned as the original parse. In this situation, where you
do not know how the name is parsed, Global Name Management will make
a best attempt to parse the name as you have entered it. It is also
possible for an additional alternate parse to be produced.
- NameWorks configuration file examples
- Documentation errata: Any configuration file examples related
to NameWorks that show the ONTERM= parameter
should not be left blank. This parameter requires that you specify
a file_path_name. See the topics "Comparison files section of the
configuration file" and "Embedded Search information in the configuration
file".
Also in "Comparison files section of the configuration file"
topic, the RegRules for Arabic and Pakistani have been consolidated
into SouthwestAsian RegRules. The presence of a "PNREG=arabicRegRule.ibm,arabic"
or a similar statement for Pakistani in a configuration file will
cause an exception to be thrown. Arabic and Pakistani RegRules no
longer exist. ONV=onv.ibm should also be included
in the example and in defined parameters list. ONV also specifies
one or more variant files that can be associated with a data list.
Variant files are applied to each name list that is associated with
a data list.
Known issues - Enterprise Name Search
Be
aware of the following additional information and instructions before
you install and use Enterprise Name Search.
- Enterprise Name Search requires UTF-8 encoding
- If you are working with data that includes non-ASCII characters
with the SOAP Web services, make sure that the locale environment
variables are set to use UTF-8 encoding. REST Web services are not
affected. If your systems do not default to UTF-8 encoding, do the
following:
Linux/UNIX environment:- Find or install the UTF-8 version of the locale for your language
and country. This locale must be installed on all machines used in
the ENS cell. Use the locale -a command to see
a list of installed locales on the machine. For example:
jksmith@din:~$ locale -a
C
C.UTF-8
en_AG
en_AG.utf8
en_AU.utf8
en_GB.utf8
en_US.utf8
ja_JP.utf8
POSIX
Find or install the UTF-8 version of the locale
for your language and country. For example, if you are in the US,
you can select "en_US.UTF-8" from the list.
- Make sure that the shell environment that starts ENS WebSphere
processes (runs the <ENS Install Home>/bin/start-<profile
name>.sh scripts) should have the following environment
set to the selected UTF-8 locale BEFORE starting ENS. Bourne shell
(sh) example:
export LC_ALL=en_US.UTF-8
export LANG=en_US.UTF-8
Example for an sh shell:
LC_ALL=en_US.UTF-8
LANG=en_US.UTF-8
export LC_ALL
export LANG
Windows environment:
Windows does not support
making UTF-8 the system-wide or user-wide default character encoding
and applications must be configured to use UTF-8 on a case-by-case
basis. For ENS, you must configure the Java JVM that runs WebSphere
and ENS to use UTF-8. IBM J9 Java JVM is bundled with WebSphere and
has an environment variable for specifying JVM options called "IBM_JAVA_OPTIONS".
This variable can be set on Windows in multiple ways.
- From the Windows command prompt:
- Set the environment variable: IBM_JAVA_OPTIONS=-Dfile.encoding=UTF-8
- Run the start-(ENS Profile Name).bat script.
- Scoped to a specific Windows user:
This setting will
propagate to all IBM J9 based Java applications run by the user.
- Click the Environment Variables... button
in the Advanced tab of the system Control Panel.
On Windows 2008 Server go to Start > Control Panel > Advanced
system settings to be taken directly to the Advanced tab
of the System control panel). The Environment Variables dialog
appears.
- In User variable for <current user name> section,
click the New... button.
- The New User Variable box appears. In the
"Variable name" field, enter IBM_JAVA_OPTIONS.
- In the "Variable value" field, enter -Dfile.encoding=UTF-8.
- Click OK in the box. Then click OK in
both the Environment Variables dialog and in the System control panel.
- Log out of Windows and then log back in. Any Java application
started by this user who is using the IBM J9 JVM will now use the
UTF-8 encoding by default.
- Scoped to an entire Windows system
This setting will
propagate to all IBM J9 based Java applications started on the Windows
server.
- Click the Environment Variables... button
in the Advanced tab of the system Control Panel.
On Windows 2008 Server go to Start > Control Panel > Advanced
system settings to be taken directly to the Advanced tab
of the System control panel). The Environment Variables dialog
appears.
- In System variables section, click the New... button.
- The New System Variable box appears. In
the "Variable name" field, enter IBM_JAVA_OPTIONS.
- In the "Variable value" field, enter -Dfile.encoding=UTF-8.
- Click OK in the box. Then click OK in
both the Environment Variables dialog and in the System control panel.
- Reboot the system. Any Java application started on this system
that uses the IBM J9 JVM will now use the UTF-8 encoding by default.
Note: The above IBM_JAVA_OPTIONS environment variable solution
should also work on any other platform that uses the IBM J9 JVM to
run WebSphere, including Linux and AIX. It would not work for Solaris
because IBM does not use IBM J9 JVM with that operating system.
- Creating multiple profiles using the ENS CU utility
You should only create one profile at a time. Creating multiple
profiles using the ENS CU utility can take significantly more time,
especially with large numbers of profiles. Creating each additional
profile is slower than the last one because the newly created profile
causes an SSL encryption key to be generated between it and each existing
profile. These keys are required to support the SSL/HTTPS secure communication
feature between any two profiles. For example, if you create 35 profiles,
the 35th profile needs to create 34 encryption keys.
- One database driver for all systems in an ENS distributed environment
- As noted elsewhere in the ENS documentation, the database driver
must be the same and at one location for all of the ENS host and server
machines in your ENS environment (referred to as an ENS cell). You
can create a dynamic link to location of the database driver. If all
of the machines point to the dynamic link, you can make changes to
the database driver location without having to makes changes on every
machine.
To see the latest information about known
problems and issues
As problems are discovered and resolved, the IBM Support
team updates the Support portal. By searching the Support portal,
you can quickly find solutions to problems.
At time of publication,
there were no known installation problems. Check the Support portal
for the most current information.
Product documentation
You can
find product documentation version 5.0 in the following places:
- Version 5.0 information center
- Access at ibm.com: http://pic.dhe.ibm.com/infocenter/gnrgnm/v5r0m0/index.jsp
- Access on your local server, installed as part of the product
install:
Open a web browser and enter the URL for your server:
http://<servername>:<HTTP
port#>/help/index.jsp- <HTTP port#>
- The documentation port number that you specify during installation.
- <servername>
- WebSphere Application Server hostname or IP address.
- Culture reference for Name Analyzer
- Culture reference information is provided within the IBM InfoSphere
Global Name Management Name Analyzer tool.
- IBM product Support home
- Access at ibm.com: http://www.ibm.com/support/entry/portal/Overview/Software/Information_Management/InfoSphere_Global_Name_Recognition
- In addition to Technotes and other Support-related information,
contains links to the information center, PDF versions of the product
information, and the latest updates of the release notes.
Announcements
You can find
the latest announcement letter, which is linked to from the following
page at
http://www.ibm.com/software/data/infosphere/global-name-recognition/. See the announcement for the following information:
- Detailed product description, including a description of new functions
- Product-positioning statement
- Packaging and ordering details
- International compatibility information
Copyright and trademark information
IBM,
the IBM logo and ibm.com are trademarks of International Business
Machines Corp., registered in many jurisdictions worldwide. Other
product and service names might be trademarks of IBM or other companies.
A current list of IBM trademarks is available on the Web at "Copyright
and trademark information" at www.ibm.com/legal/copytrade.shtml.
IBM InfoSphere Global Name Management, Version 5.0