If the poster has been making zipped backups right along, then recovering the project will be trivial, and words lost will be equal to those written since the last backup was made. This is because a zipped backup is a static snapshot of data, hence a true backup.
However, if the poster’s only ‘backup’ is that his project’s on Dropbox…well, who can say? It’s possible he might be able to put it back together again. Then again, maybe not. All he can do is roll up his sleeves and try. This is because Dropbox is a syncing service, not a true backup.
I wish that poster the best, and hope he’s been making zipped backups right along.
The poster should have zip backups in Dropbox or any other cloud service + another backup like ChronoSync or Time Machine. It’s not an either/or situation. Zip backups with unique names are a static backup in Dropbox. Did someone forget their belt and suspenders and get in trouble? Well, that’s a scenario no one should opt for anyway.
The whole point of a continuous backup is that it’s checking for updated files all the time. As soon as I close an application – whether Scrivener or anything else – the changed files get queued for backup. (In practice, I find BackBlaze runs about every fifteen minutes.)
Making a reliable backup of a file that’s actively being edited is fairly challenging for any application. If that’s your use case, I would recommend enabling Scrivener’s “backup on manual save” option. That creates a static archive at whatever interval you choose, and then the static archive can itself be backed up by Dropbox, Time Machine, BackBlaze, or whatever tools you prefer.
This thread has jumped all over the place, so I’d like to bring it back to reality by describing the scenario I commonly see.
User stores projects “in Dropbox,” (or iCloud, or OneDrive) but has no Time Machine volume or third-party backup service. Probably the user depends on a free-level Dropbox account, so no versioning either.
Machine is left running all the time, with Scrivener up and running, which means that Scrivener’s own automatic backups never run or run rarely. (Note that closing a laptop “sleeps” it instantly. Scrivener remains open and no backup is made.)
Data loss occurs for whatever reason: maybe Scrivener crashed, maybe a toddler pounded on the keyboard, maybe an evil ex-girlfriend/boyfriend wreaked havoc.
(As soon as the data loss occurs, Dropbox helpfully propagates the changes to their server and thence to all connected devices.)
User sends plaintive wail to Technical Support. “Scrivener ate my data!!!”
And there isn’t a darn thing I can do to help. No automatic backups, no third party backups, not a darn thing anyone can do.
I’m pretty sure that none of the participants in this thread is likely to find themselves in the same situation as the hapless user described above. But this user is why I am so adamantly opposed to the suggestion that “if it’s in Dropbox, you’ll be fine.”
Yes, of course a canny user has zip files and probably backs them up somewhere; I mean there is no static, uncompressed copy of an open project.
Neither the awesomeness of a backup (in the sense you mean it) nor the carelessness/ignorance of others means I cannot retrieve a backup from Dropbox. Preferably a zip backup in Dropbox. I don’t trust uncompressed project backups no matter where we put them.