Works for my workflow. I work on reasonably small files (my whole cloud storage is about 20 gb) and I like to have the storage available in my file manager at all times. Wouldn't want to run proprietary software or waste memory on it.
The biggest problem I have with these 3rd party clients is sometimes things just don't work right. Like for example gimp just fails to open files in the synced folder (google drive not the OP project) until I copy them out and open them. Not sure what the issue is though.
It's a good tradeoff since it lets me escape the official Dropbox client trying to upload my drives, photos, etc. to the cloud, and spam me with upsell offers on a daily basis (if you're operating with a nearly full Dropbox in steady-state).
I have 32GB of RAM and I would still do that trade off. The 60MB is there all day long doing nothing for me. The internet connection is actually being used for something useful which is synchronizing my files (the actual purpose of the software). I feel like I'm wasting more resources with the first option.
Also it's not even that bandwidth expensive with most users' workflow. Maybe this should be a concern for video editors. Majority of people who are just editing word documents (or code, for what I believe is most HN's audience), wouldn't notice the difference.
How is this any different than storing data in S3? Companies have petabytes in the cloud now. If their networking supports their usage then Dropbox is a good solution given all the features it provides.
The "imagine $X" construct has colloquially grown to mean "$X is irrational", which is strange to me here - the lack of binary diffing means if you change a single bit of your video editing project, the whole thing gets synced to the cloud. Dropbox is a common solution for small business VCS that's not VCS, I've seen it used for born exactly that way a half dozen times the past decade.
> the lack of binary diffing means if you change a single bit of your video editing project, the whole thing gets synced to the cloud
I know someone using Dropbox for video editing, they've been using the Synology Dropbox client which also uses the API. It hasn't been a problem.
The large video files (tens or hundreds of GB) are ingested once and never edited, the only edits to files are the small project files (megabytes), and some graphics files (hundreds of megabytes). Re-exports of rendered video causes the whole file to change anyway due to the nature of video compression so binary diff doesn't help there anyway.
I just think that Dropbox is an inappropriate file storage mechanism (especially for heavy artifacts being generated locally) for someone moving that much stuff in and out of a long term storage system, especially at those levels. I'm not someone that generates 500GB of CAD artifacts a day, or anything even remotely similar, but I ingest about 15GB of global market data and news a day and I laughed a little bit reading your response (I understand how you were responding to GP, and your response makes sense) because its just nonsensical to use consumer focused products for anything serious.
Just because I slap on a "business tier" sticker on a water bottle doesn't mean the water bottle is better equipped at solving enterprise needs.
