Barbara, You are correct. US7ASCII is desupported for national character set. The default is AL16UTF16. Oracle 9i Upgrade automatically changing the NLS_NCHAR_CHARACTERSET from US7ASCII to AL16UTF16, but it is not changing the NLS_CHARACTERSET (Database Character set). This is why I would like to change it by myself to the appropriate one and trying to find some valid reasons why we should go from US7ASCII to WE8ISO8859P1 ? And also, trying to find if it is really required to convert from US7ASCII to WE8ISO8859P1, and relying on what the list experts say. -- Janardhana Babu -----Original Message----- From: Barbara Baker [mailto:barbarabbaker@xxxxxxxxx] Sent: Friday, October 29, 2004 8:33 AM To: jbdonga@xxxxxxxxxxx Subject: Fwd: RE: Which Character Set for 9.2 Babu: Here are some notes I made about migrating to version 9i: This is from the migration manual (A96530-02 As of Oracle 9i the sql NCHAR datatypes will be limited to the Unicode characterset encoding only (UTF8 and AL16UTF16). Any other NCHAR datatype will no longer be supported. When upgrading to 9i the value of the NCHAR is based on the NCHAR datatype used in the Oracle8 version. If the old National character set is UTF8, the new will be UTF8. Otherwise the national character set is changed to AL16UTF16. The change itself is done in step 31 by running the upgrade script Here's what I observed: Newly created 9i database: PARAMETER VALUE ------------------------------ ---------------------------------------- NLS_NCHAR_CHARACTERSET AL16UTF16 Our 8i databases: PARAMETER VALUE ------------------------------ ---------------------------------------- NLS_NCHAR_CHARACTERSET US7ASCII Our 9i databases after conversion: PARAMETER VALUE ------------------------------ ---------------------------------------- NLS_NCHAR_CHARACTERSET AL16UTF16 (So after using the migration utility, the Oracle utility changed our US7ASCII characterset to AL16UTF16) I'm still a bit confused about the characterset stuff myself, but I hope this info will help. Barb --- Janardhana Babu Donga <jbdonga@xxxxxxxxxxx> wrote: > From: Janardhana Babu Donga <jbdonga@xxxxxxxxxxx> > To: 'Niall Litchfield' <niall.litchfield@xxxxxxxxx>, > Janardhana Babu Donga <jbdonga@xxxxxxxxxxx> > CC: "Nelson, Allan" <anelson@xxxxxxxxxxx>, > rich.jesse@xxxxxxxxxxxxxxxxx, > gsais@xxxxxxxxxxxxxxxxxxx, > oracle-l@xxxxxxxxxxxxx > Subject: RE: Which Character Set for 9.2 > Date: Wed, 27 Oct 2004 15:47:55 -0700 > > Why is the 8 bits good and 7 bits bad? > Is there any valid reason? > > -- Babu > > -----Original Message----- > From: Niall Litchfield > [mailto:niall.litchfield@xxxxxxxxx] > Sent: Wednesday, October 27, 2004 1:58 PM > To: jbdonga@xxxxxxxxxxx > Cc: Nelson, Allan; rich.jesse@xxxxxxxxxxxxxxxxx; > gsais@xxxxxxxxxxxxxxxxxxx; > oracle-l@xxxxxxxxxxxxx > Subject: Re: Which Character Set for 9.2 > > On Wed, 27 Oct 2004 10:05:34 -0700, Janardhana Babu > Donga > <jbdonga@xxxxxxxxxxx> wrote: > > Dear List, > > > > If it is not required to convert from US7ASCII to > WE8ISO8859P1, I don't > > understand why so many people are changing their > character set to > > WE8ISO8859P1 from US7ASCII. Could some one explain > please? > > 7bits bad, 8 bits good. > > -- > Niall Litchfield > Oracle DBA > http://www.niall.litchfield.dial.pipex.com > -- > //www.freelists.org/webpage/oracle-l > -- //www.freelists.org/webpage/oracle-l