pjbuk007, as OP, since you stepped in against it, I've got to step in for it again, just like I did 4.5 years ago.
To say I'm not completely against you...
Merging with existing work, I am all for NOT allowing it. Way too many problems there. I've done a LOT of merge work since 2008 (lots of trades), and I've found I've had to hand-edit 95% of entries, find my own sources, or scrapping data entirely for a few metric tons of unsourced material. People, as a whole, don't source. Its sad.
That said...
I still think merging with an offline tree that has was originally uploaded solely as an online version of the offline file, and changes back and forth synced is a great idea. It's not like it is a novel idea by any means. Source code isn't all that different (open a gedcom...). Source code uses a choice of Subversion, Gitorious, or Mercurial. All of them make a dead simple offline-online sync for something that's supposed to be synced already. SVN (subversion) would be super easy to use as a basis. As per how it works, you start from whatever changeset (1 being initial sync), and then send a 'diff' (what changed). It updates the server, or you receive the diff and update your copy. As many people as you want to have work on it could theoretically work on it, and they'd receive the diff file for the changeset, and apply it and 'viola' its synced.
Its not novel, it's not new, it's just using an already established system and putting it to use in a way to help a new crowd, genealogists. If you want to argue whether it is reliable... some form of collaborative source control (what would likely be the basis for syncing) is used for EVERYTHING you do online. Google, Microsoft, heck... any game or program... they all use some form of source control where patches can be submitted, syncing the master source tree with offline source trees.
The other nice thing, is if it's stored in a format like a gedcom then it's all easily plain text, which is highly compressible, so it'd be next to no server load to even submit massive changes.
The only caveat is, with all of the various options, you have to start from an already synced copy.
I heard the new version (2013) had some form of syncing, I haven't checked yet. Due to inability to sync, I've been using Legacy (if I can't sync, I'd rather use a more robust-for-sourcing program, and FTM's (I've only been trying them since they went to years, 08,09, etc) haven't been able to hold a candle to Legacy).
I should look into the new FTM, it's possible syncing was added already. Only posting back here, since your post said down with syncing and it's the only reason I moved away from FTM in the first place, and I still think it's the Best Idea Ever™.
To say I'm not completely against you...
Merging with existing work, I am all for NOT allowing it. Way too many problems there. I've done a LOT of merge work since 2008 (lots of trades), and I've found I've had to hand-edit 95% of entries, find my own sources, or scrapping data entirely for a few metric tons of unsourced material. People, as a whole, don't source. Its sad.
That said...
I still think merging with an offline tree that has was originally uploaded solely as an online version of the offline file, and changes back and forth synced is a great idea. It's not like it is a novel idea by any means. Source code isn't all that different (open a gedcom...). Source code uses a choice of Subversion, Gitorious, or Mercurial. All of them make a dead simple offline-online sync for something that's supposed to be synced already. SVN (subversion) would be super easy to use as a basis. As per how it works, you start from whatever changeset (1 being initial sync), and then send a 'diff' (what changed). It updates the server, or you receive the diff and update your copy. As many people as you want to have work on it could theoretically work on it, and they'd receive the diff file for the changeset, and apply it and 'viola' its synced.
Its not novel, it's not new, it's just using an already established system and putting it to use in a way to help a new crowd, genealogists. If you want to argue whether it is reliable... some form of collaborative source control (what would likely be the basis for syncing) is used for EVERYTHING you do online. Google, Microsoft, heck... any game or program... they all use some form of source control where patches can be submitted, syncing the master source tree with offline source trees.
The other nice thing, is if it's stored in a format like a gedcom then it's all easily plain text, which is highly compressible, so it'd be next to no server load to even submit massive changes.
The only caveat is, with all of the various options, you have to start from an already synced copy.
I heard the new version (2013) had some form of syncing, I haven't checked yet. Due to inability to sync, I've been using Legacy (if I can't sync, I'd rather use a more robust-for-sourcing program, and FTM's (I've only been trying them since they went to years, 08,09, etc) haven't been able to hold a candle to Legacy).
I should look into the new FTM, it's possible syncing was added already. Only posting back here, since your post said down with syncing and it's the only reason I moved away from FTM in the first place, and I still think it's the Best Idea Ever™.