If Distributed Version Control and their related source sites improved my coding life massively this year then Dropbox was a little piece of everything magic that dropped into my life.
Dropbox is a distributed file share system, for someone with multiple machines running multiple OS that’s pretty handy. Previously I had been using Jets3t and Amazon S3 to try and keep files in the cloud. That’s a great system but it has a major problem that Dropbox fixes effortlessly.
Data objects in your S3 buckets have no revision history, there’s only one copy of the object and when you upload the wrong revision of the file you clobber the correct version and then if you don’t notice you can quite happily overwrite later revisions of the file on your other machines.
Dropbox versions your files and allows you to revisit previous revisions of your data. It’s like a file share and source control in one awesome package. Dropbox also handles the synchronisation for you seemlessly. As soon as a file has changed and is accessible it is whisked away to the net. Your other machines synchronise on startup or when they reconnect to the web. The file copy is very fast for me but I tend to have lots of bitty files rather than monster sets of images I am constantly working on.
It also has some features that I haven’t tried yet like creating public URLs to data for general download.
Using Dropbox has radically simplified my life, my encrypted password files now synch between home and work machines without my having to do anything. I can work on documents as I get inspired without having to wonder whether I uploaded the latest copy to S3. All my worries about clobbering files are over and even if I can’t run the nifty client application I still get to access my Dropbox via the web interface.
Dropbox, for me has been an unqualified success.
Echo One 
Thanks for the heads up, this is a great service.
The only problem I’ve come across so far is that, after I dumped 200,000 files into my Dropbox folder, it needs nearly 3gig of RAM of to sync them. Just about OK on my home box, not so good on much on my work machine
I really hope this due to the quantity of files added in one go, rather than the quantity present in the folder, and that it will therefore behave itself once it’s synced them, coz this only uses up 8% of the minimum quota they sell… if it needs this much RAM constantly it’s going to be totally useless by the time I fill up the space.
I haven’t tried that many files all at once! For me the Dropbox process sits at the same level of resources as my MP3 player.
I’d be interested to find out whether the client throttles itself on lower resource systems. Using up to the limit of the resources of a client machine when synching is much more acceptable than having a fixed set of resources per controlled item.
Yeh, sadly it turns out that, although it doesn’t use this memory all the time, it does while it spends the first half an our or so after boot up scanning all the files.
To be fair, the CTO of Dropbox looked into my problem himself, and said they probably be able to fix it within the next few months, and offered me a refund on the 50GB account.
I might write my own, Amazon S3 makes it all look pretty damn easy and it’d be a fun project to try it out on.
Are they a lot of small files? I have tried using Bazaar to stash things like websites to various servers via SFTP.