On 4/19/2012 3:23 PM, LeVan,Ralph wrote:
We see Unicode data pasted into MARC8 records all the time. It happens enough that my MARC8->Unicode converter takes a second look at illegal MARC8 bytes and tries a UTF-8 encoding as well.
Right. I see it too. I'm arguing that means cataloger entry tools, the tools which catalogers are using when they paste that stuff in, are not giving the cataloger sufficient feedback as to their entry. Flagging completely illegal byte sequences in the output encoding and not letting them be saved; make sure cataloger input is displayed back _as appropriate for the current encoding_, so they get immediate visual feedback if they're entering bytes that don't mean what they think for the operative output encoding.
I think it's possible _no_ cataloger interfaces actually do this. (although if any do, I bet it's MarcEdit).
If Connexion doesn't, for interactive cataloger entry, it'd be awfully nice if it did.
