Bitcasa: Infinite Storage (really?)

I don't care how smart your technology is, representing your product as "infinite storage" is inaccurate and misleading. There is obviously no such thing as infinite storage on a finite device, and saying you don't have to use a "clunky web service" implies it's not a web service, when it obviously is.

I respect their need to find a sexy way to sell a cloud-storage-and-sync product, especially since it's an increasingly crowded space, but this seems a bit sleazy.

Oh good, I can finally build a simulation of a Turing Machine (with an infinite tape).

When reading this so many red-flag warnings popped up that I could not decide if they were deluded or frauds. I will give them the benefit of the doubt and simply guess that they are deluded regarding the storage savings they will eventually realize via de-dupe. Given that the founding team includes a bizdev guy, a marketing guy, and a sysadmin/ops guy I am guessing that they can put together a nice pitch and powerpoint, have some easy answers to the operational problems of competing in this space, and have a lot of hand-waving answers to the harder technical problems. The latter problem will probably end up killing them unless we are just seeing a PR dump in preparation for a pivot to being just another online backup company.

One major fly in this whole "de-dupe" claim is that it will probably not even work out for them even if they did have some magic sauce to dance around the de-dupe/crypto conflict others have noted. The problem is that the files which users actually care about and want to back up are not the common files but the ones which make their data unique; it is not the mp3s or hollywood videos that matter, it is the data/content that each person has created that matters. If my disks crashed, my online and offline backups were corrupted, and I needed to rebuild my system I could get the common OS files in a hour, the mp3s and videos would take a few weeks of passive torrenting, but the pictures, home videos, and personal documents would be gone forever. It is these files that matter for a backup services, and they are not going to be something that you can de-dupe even if they were not encrypted.

Back when the term "de-dupe" did not exist and convergent encryption was something we were inventing before the term existed the thought was that a backup service employing these techniques would enjoy a massive savings in storage costs -- it turned out that people cared less about backing up the data that was easy to de-dupe and original data was a much larger portion of what users uploaded than we expected. This was back when pics were a meg or two and personal video was low bitrate, now that even a mobile phone is dumping multi-meg pictures and you can get a HD video camera for a hundred bucks I cannot imagine how anyone would convince themselves that de-dupe is going to make any significant difference to the operational costs of such a service.

Maybe it's like the yahoo mail infinite mailbox. They just severely limit the bandwidth in/out, but storage is infinite. Given an infinite amount of time...

I have to say that I find it a 'little' sad that the entire post reads like a Press Release, touting 'client side encryption' like something new and exciting and two thirds down it's revealed that CrunchFund is an investor.

Poor author didn't even know when she interviewed them.

I'm confused, they mentioned at techcrunch disrupt that they are able to offer infinite storage because they are deduplicating data. Fine. But they also said that they will be doing client side encryption. Contradiction - encrypted data cannot be deduplicated.

They really pushed the "infinite storage" point in their presentation. But didn't they mention something about predicting which files to cache before they are requested? I would have liked to see more focus on that. Otherwise, it doesn't seem much different from dropbox, which is more established.

We will see. Actually some backup companies also have such claims on storage (e.g. http://www.crashplan.com http://www.backblaze.com)

Well this is also backed by CrunchFund.... we would see (seen via TCDisrupt)

For, uh, small values of infinity? Or something?

I find that I don't run out of storage on my computers any more, so perhaps I'm not their target audience. Best of luck competing in that space, though--it seems pretty crowded already. :)

Just a reality check on your landing page: your #1 asset is a vimeo embed, which causes several problems:

  1) I'm on roaming 3G, I'm not paying that much to know what you're about
  2) Flash is blocked by default so all I see is a black square
  3) You're royally pissing on blind/disabled people.

Besides, the color scheme / general theming is generally sub-par and difficult to read.

The "Learn More" link is more informative, but still, I think you really badly need a designer.

From what I understand the concept is something similar to Dropbox, except instead of simply syncing files between the folders on different computers where all the files are physically present on each, it focuses on remote access (possibly with an AFS-like caching layer).

And by infinite, they probably mean "as much as you are willing to pay for," or, during the beta, "as much as you want until it gets ridiculous."

I watched them present on the TC Disrupt feed. I wish one of the panelists would have asked them how the file system stream impacts data caps imposed by many ISPs.

The problem with all these online storage services is uploading your data.

At my 1 mbit uplink it would take 97 days of non-stop uploading to upload 1TB. That's if my ISP doesn't ban me.

Ive got an "infinite stotage device" called /dev/urandom.

It does take a while to wait for your data to come back out :)