Show HN: tidy.io: archive your Dropbox folders (built with Amazon Glacier) | Hacker News Reader

Show HN: tidy.io: archive your Dropbox folders (built with Amazon Glacier)

tidy.io

11 pointsalmost7 comments

Hey guys, this is my new app, it lets Dropbox users archive or backup their files. It's built with Amazon Glacier (and S3 and Simple Workflow Service). Feedback very much appreciated, I'll be around to answer questions for a few hours.

I did post this a few days ago as well but I messed up the timing and it sunk without a trace. Hoping people don't mind too much if I try again!

This seems intriguing. But it raises questions: Is not Dropbox based on AWS storage? Therefore if AWS goes down...

Secondly, given that Dropbox has native version control (via its web interface) how is a backup with tidy.io on Amazon Glacier a benefit?

Just saying... if tidy.io wasn't AWS-based, then you'd really have something here.

Hey, thanks for commenting! Dropbox does have a sort of version control but it only keeps files for about a month (and even then you wouldn't want to rely on it if you definitely wanted the files back!). And even if all the revisions of the files you want are still in Dropbox's memory you'd still have to figure out which which you needed for every file (potentially a problem in a folder with hundreds or thousands of files!).

AWS going down definitely isn't the only reason you might lose data from Dropbox, in fact it's probably one of the least likely. Far more likely is accidental deletion (delete it on one computer and Dropbox will delete it on all your computers!) or file corruption (which you might not notice for months). Even so tidy.io stores a copy of every file on Glacier which is a separate system to the S3 service that Dropbox uses

I think the idea is that if you have a large-ish volume of old stuff--archives from old projects or documents or stuff that you won't need for a while--and it's already in Dropbox, tidy.io lets you offload that stuff into cold storage easily. That frees your dropbox from clutter (and space).

You might also see tidy.io as a nice gateway to Glacier. You don't need to reupload files or use a CLI.

Btw, Dropbox is built on Amazon S3, which has a pretty impressive SLA[1] and service-record. That's not to say that it has perfect uptime, but in general, issues with AWS have been availability issues, rather than durability. Both S3 and Glacier have really good data-durability guarantees(99.999999999%)[2], so I wouldn't worry durability within AWS. With backups, you don't care about availability nearly as much as durability.

[1] http://aws.amazon.com/s3-sla/ [2] http://aws.amazon.com/glacier/#highlights

Hey thanks! You said it a lot better than I did :)

Thanks to you for an excellent product! :)

> given that Dropbox has native version control (via its web interface)

I'm guessing you've never had to do a large restore from Dropbox before? It's technically possible, but on a file-by-file basis. Delete a folder of 1000 photos? You've got 1000 restores to do.

A competent restore system has been begged for by the users for years:

https://www.dropbox.com/votebox/2215/restore-to-a-specific-e...

Frankly, it's almost sad that Dropbox's restore feature is so bad that people are able to make entire products based around making a good version of it.

Reply on news.ycombinator.com