Ian Brooke wrote: > > Hi All, > I've been trying to mould my brain around this concept of File Ownership and I am truly stumped. Why is it that a file of data, originating from FreeBMD, transcribed by a volunteer to FreeBMD, Uploaded to FreeBMD is considered to be the 'property' and responsibility of that transcriber? I see little sense and little advantage in it at all and it seems to create far more problems than solutions. I would have thought that most transcribers would be quite happy to finish a page, upload it and forget about it - FreeBMD could then move it from their upload area to a centralised repository. If changes were needed to that file it wouldn't be too difficult to find a responsible, experienced transcriber(s) willing to take on the task on behalf of the project as a whole. We immediately loose the problems of AWOL transcribers with files that 'no-one' can get at easily when they need changing. We can forget about our (imagined) problems of security of files ("I can't touch 'his'! f! > iles"), we don't have complaints from ex (and even current) transcribers when they receive emails pointing out mistakes in 'their' files and life generally seems to become simpler. > What's the problem with this? I can't see one. The idea of file ownership does bring advantages! 1) You can load a page with some unreadable stuff, then replace it later once you have had chance to look at entries again. 2) When a query comes in, it is quite likely that somebody central will not have a copy of the fiche to check the transcription. 3) Splitting the files out by transcriber allows us to identify any potential cases of seriously suspect files and to discard them. 4) Identifying by transcriber ensures that we can show that double keying was truly independent. 5) Ownership ensures that at the stage where we apply corrections to the data, there is no prospect of a "rogue" checker overwriting the data held. Much of the problem seems to result from people not appreciating the data model. The path of asking a transcriber to correct his data is very much part of the early process in the months immediately after transcription, prior to double transcription. Facilities are already being developed for the later stages of the project, where corrections to records will be applied by other people. These facilities will NOT involve editing the submitted files. -- Dave Mayall