Cloud Sync: One Approach

There were people in this world, I knew, who were totally on top of syncing their computers and using cloud storage.  And someday, I believed, I would be one of them.  But the truth was, so far I had not used cloud (i.e., Internet) storage very extensively, and I had yet to work out a system for it.  I was mostly a desktop kind of guy, with my local hard drives doing the data storage.  At the time when I was writing this post, I wanted to break through that barrier.  Not very far, to be sure; I was not about to upload everything I owned.  This was more a matter of working out the details of synchronizing a couple of folders.  But it was a start.

For this purpose, I needed cloud storage.  There seemed to be numerous options.  As noted in another post, recent articles led me toward iCloud, Box, Google Drive, SkyDrive, and DropBox as the leading current examples, though the list also included Amazon Cloud Drive, Wuala, and numerous others.  For the time being, I was working with SkyDrive.

Brief experience with several of these cloud services yielded the impression that the basic idea was to synchronize subfolders and files, located in a designated folder on my computer, with copies thereof uploaded to the cloud storage service.  Then, if I changed anything in the designated folder on my hard drive, that would be reflected in the cloud storage and, conversely, if I was out somewhere, using my laptop or some other computing device, I could make changes to the cloud copies and those changes would be propagated back to the hard drive on my home computer, next time I turned it on and connected it to the Internet (or immediately if it was already on and connected).

The specific problem confronting me had to do with the ideal way of using the cloud storage and the designated folder back home.  There were two options.  I could do my work directly in the cloud, or directly in the synchronized folder; or I could do my work in some other folder, on my desktop or laptop, and then copy the changes to the cloud folder or the desktop’s synchronized folder.

I decided that I would not work directly in either the cloud or the synchronized desktop folder.  One reason was backup.  Out on the road, the cloud might contain my only copy of a document I was working on.  If I screwed it up, that could be the end of my work until I got home again.  Similarly, at home, I figured I would be using the cloud for various projects.  Cloud storage was small compared to desktop hard drive storage, so I could only upload materials that I needed access to when I was away from home.  I did not want to have to try to mesh that awkward synchronized folder into the groups of folders and subfolders in which I might organize a given project, especially since the cloud services generally seemed to require me to specify where the synchronized folder was – and if I forgot, at least some of these cloud services (e.g., Google Drive) would get confused and would not work properly until I had done various disconnecting and relinking steps.

It seemed easier to preserve my preferred file and folder structure on the hard drive at home, and to treat the synchronized folder as a separate place that I would synchronize with certain folders on the hard drive.  In other words, I would have original documents in various folders on my home computer.  I would want to make some of those folders available in the cloud.  Therefore, I would synchronize those particular folders with the designated cloud drive folder elsewhere on my home computer; and then I would use the cloud service to synchronize that designated cloud folder at home with the cloud service out there on the Internet.  The specific folders linked to the cloud-synced folder on my hard drive would change, as I went into different projects, but the location of the cloud-synced folder would remain the same.

This two-link sync chain could sound complicated, but I believed it would be virtually unnoticeable in ordinary usage.  The cloud storage provider would handle synchronization between my desktop and the cloud.  I just needed to come up with a way to synchronize the designated desktop cloud-linked folder with some other folders on my desktop machine.  (If needed, I could then also use that same technique to update the cloud when using my laptop or some other computer while on the road.)

In a previous post, I had somewhat clarified the difference between synchronization and backup systems.  For this purpose, I wanted synchronization – that is, two-way updating.  When I changed something in my home workspace, I wanted the option to carry that change over to the folder that would then synchronize with the cloud; and when I changed something while working away from home, I knew the cloud service would bring that update back as far as the synchronized folder on my home drive, from which I would then have the option to perform a synchronization with my home workspace.

In my own existing arrangement, I was using two programs for backup and synchronization.  Neither of these programs was freeware, though I believe they may have had freeware versions that would have sufficed for some purposes.  Neither was expensive – around $40 in both cases, if memory served – and both had been worth every penny.  These two programs were Beyond Compare (which I was using largely for one-way backup) and GoodSync (which I was using to synchronize two computers at a single desktop).  (There were also other programs, mentioned in some of the posts just linked, that would provide similar capabilities.)

I wanted the synchronization at home to be manual.  That is, I didn’t necessarily want everything that I put into a given folder on my desktop computer to go running off immediately into the cloud-synced folder on my hard drive.  On the other hand, I also wanted this thing to go pretty quickly, at those times when I would be running out the door and would realize at the last minute that I had not yet uploaded my latest work to the cloud.  So I wanted the synchronization to be all ready to go at the touch of a button.

Either of these programs, Beyond Compare or GoodSync, would probably have been able to do this very well.  In my own arrangement, Beyond Compare was not always on, but GoodSync was.  So GoodSync was my first choice for the job.  That said, all of my other GoodSync jobs were set up to synchronize automatically.  I would have to set this one up to do its analysis often, and frequently, but not to do any automatic synchronizing.  That’s the button I would press when I was ready.

GoodSync would not be my tool of choice on the laptop, however.  GoodSync was a tremendous resource hog.  Even a relatively fast desktop would slow to a crawl when it was running.  This was part of the reason why, as described in one of the posts linked above, I had set myself up with two separate desktop computers, with one of them functioning more or less as a server handling most synchronization, backup, and other maintenance tasks.  GoodSync could have its way with that secondary computer, and my work pace on the primary computer would rarely be affected.  On the laptop, I would probably use Beyond Compare, but that was a question for later.

Once this decision was made, it was pretty straightforward to just add another job to GoodSync, configure it as just described, point it toward the desktop work folders du jour, and tell them to analyze, and be prepared to synchronize themselves with, the cloud-synced folder on the desktop.  As long as I had backup, it seemed like this would work pretty well.  I did the initial synchronization with some manual adjustment, and after that, I hoped, it would be smooth sailing.