I&#039;m working on a multilingual project with an Oracle 9i backend.<BR><BR>My client refuses to set up his database using UTF8 encoding. I told him this would cause problems as soon as we started accessing data using just about any non Western European character set, but they told me not to worry about it, and that I was wrong.<BR><BR>Well, the site for the Czech Republic went online last week, and sure enough, we&#039;re having problems. The Russian, Greek and Chinese data is on hold until we can find a solution.<BR><BR>So, to the meat of my question... The JDBC connectors from Oracle apply a built-in algorithm against raw output from an Oracle database (as I understand it). This algorithm corrects errors in the way Oracle returns special characters when an Oracle database is not set up to use UTF8 encoding.<BR><BR>Since we must use an ODBC connector, and Oracle&#039;s ODBC connector doesn&#039;t include this algorithm, I&#039;m wondering if anyone knows if this algorithm is in the public domain, or is relatively easy to recreate. All of my connections to ODBC are through a com component written in C++, and I would be interested to see if we can incorporate the algorithm inside this component. Have to find it first, though.<BR><BR>Other than that, any other tips or suggestions for solving this problem, other than telling my customer, "I told you so"?