Currently, field bridges can define whatever name they want for their fields, and this name is used exactly as is.

Two issues arise from this situation:

1. It's not easy for field bridge implementors to override the default field name (prefix + java property name) properly: they are only given the default field name, with no indication whatsoever on what is the prefix and what is the name. Even with string manipulation, it may be impossible to tell one from the other (when the prefix is overridden, for instance). Most of the time, the easiest solution is to only add a suffix to the default field name, which is probably the sanest choice anyway.
2. Field bridge implementors can easily bypass the prefix and add a field to the root of the document, which is a dubious practice at best.

When we design the FieldBridge 2.0, we might want to tacle this (especially the second issue).