We also had some discussions about this here:
DBSIG: German Pork Butchers
http://www.rootschat.com/forum/index.php?topic=613454.0without coming up with a satisfactory solution

Some ideas so far:
[A 'submitter' is someone who submits i.e. enters a record, regardless of whether s/he is actually researching this family.
'A researcher' is someone researching this family. The 'submitter' may or may not also be a 'researcher']1) the 'submitter' "authorises" other 'researchers' to edit his submissions (Berlin-Bob)
2) topic with DBSIG corrections (SwissGill)
Topic could get unwieldy; needs an administrator to prune it after corrections have been made.
3) A comments/corrections field in the database where anybody can list corrections (Histres)
The submitter can then edit the relevant fields, and delete the correction note. A bit like a "post-it" note. I would suggest only allowing registered 'researchers' to add corrections and comments.
Following up on another idea from Histres ...
Suppose a record about Müller (#1234) has been submitted by someone who is not actually researching this name.
A 'researcher' now submits a record about this same person (#4321)
Comments could like these could also be added here
- "this is the same person as record #4321 , please only add comments there"
- "all data from this record has now been added to record #4321. Please delete this record"
- etc.
Feedback on the above, other ideas ??
Gruß,
Bob