@narrsd: Further, as above, you are going to lose something very important if you try anything but Scrivener’s Dropbox method: that it is able to do its checks, quite often automatic corrections, and even if not, complete safety against losing your work, for any issues with syncing.
I would just like to point out that copying a project from one place to another (a task this approach attempts to automate) is entirely safe. There is no risk with copying projects beyond the baseline level of risk that all technology has. There is no need for checks, automatic corrections, and you will not lose your work unless you do silly things—like copy the wrong project, or halt the copy halfway through the progress bar.
Bear in mind that a Scrivener project is a folder with files in it. It is no more mysterious than that! It does not take special technology, or piles of sync code, to copy folders with a bunch of files in them between devices and systems. It’s one of the most basic and rudimentary capabilities of any file system.
It is also the only method I ever use to get data on or off my device. I wouldn’t trust Dropbox (or any other sync) nor do I have need for that level of complexity for how I work. There are many safe ways to copy projects around between devices, and many of them are documented in the Scrivener use manual for Mac/PC. We even have an article on semi-automating through copies, which describes a method that is far safer than using live sync, precisely because it makes use of the dirt basic technology involved in copying projects between systems rather than syncing them.
Thus: these are all officially supported methods, and in fact could be argued to be the default methods, since you have to go out of your way to set up sync.
Now one might quibble that the transport mechanism one chooses to use for this copying is sync based, but if you’re using it to transfer whole copies, or to even better yet transfer zipped copies, then the specific mechanism is almost irrelevant. We aren’t “syncing” in the sense that one folder and its contents is being updated in pieces by multiple devices in a sequential fashion. That is where things get complicated in terms of technology, and error-prone in the sense that it is easy to mess up at a human level.
With copying, you copy “A” from iOS then all of “A” arrives on the Mac and becomes copy “B”. Edit it and copy “B” over to iOS and it becomes “C”. If you overwrite, at that point “A” ceases to exist because it is fully replaced from top to bottom with “B”. But when one uses the copy method you do not need to overwrite—unlike sync users, you can keep a date stamped log of copies on each device, just for a little extra safety net. I can make sure “C” copied over well, and then choose to delete “A” if I want.
Like the ones you make when you forget to close on one machine before editing on another.
That isn’t a complication when you are copying projects back and forth (nor is it really one for the mobile version, as it is designed to be safe to work on a project left open on the Mac/PC, specifically because many people use it to quickly jot some idea down while walking around the house or whatever).
@anxiousdog: I have about 5TB of space on OneDrive, so I’m wondering if I can use it instead… Have you had any issues so far? As you said, remembering to download the files and keep them updated before you open Scrivener seems to be the primary issue.
There were some noted issues with using it specifically as originally described, on account of bugs in the iOS file manager that make it so it can get out of date with the sync server. The other issue I spotted is that the trigger, as originally described, happens whenever you multitask. Depending on how you work that may be copying things around far too often, and with larger projects, may even cause issues if you switch in and out of things so fast that it isn’t getting a chance to finish copying.
I would say a manual trigger, run from Files, is a lot safer all around.
Do check out the article I linked to above in this post. It’s the safest approach, is not too much less convenient than built-in sync, and can be used with everything from USB cables to whatever sync service you prefer. It works hand-in-hand with your backup on close approach to working.
That method, with the exception of using a much safer “stack of zips” instead of replacing one single copy over and over is essentially what is being described though, and could be automated if one chooses to. Personally I’ve never bothered as dragging the latest .zip over to Scrivener’s folder on the iPad, and tapping it to unzip, really doesn’t put me out.