A great partner recently asked me about a newsgroup post mentioning Unicode CTC files. The post mentioned that Unicode CTC files could be compiled and used in a package satellite dll in order to accomplish localizing the dll. This is not true. The CTC compiler does not support Unicode files.
The CTC compiler only accepts multi-byte files. In the past, to localize a satellite dll, one could install an OS (like JPN or DEU), author the CTC file on that OS, run the CTC compiler on that OS and then build the satellite dll on that OS. This required a different OS and build for each localized language you wanted.
Another option is to take the CTO binary blob that the CTC compiler produces and localize that. The only problem with this method is the difficulty. You'd need to know the binary format and we've never published that.
Since Whidbey Beta2, we've improved this considerably. The CTC compiler now accepts a switch to specify the code-page of the file being compiled. The file must still be a multi-byte file, but by specifying the code-page on the CTC command line, you don't need to have the localized OS on the machine. You simply need to know the code-page of the CTC file as it was authored originally.
Unfortunately, we do not have a sample in the VS SDK (formerly VSIP) that shows how to do this. If you think a sample would be helpful, please let me know.
Thanks and happy localizing,