Are we competing with Amazon?
As of a few minutes ago, Amazon web services launched an online storage platform for developers (thanks for the tip, Nik). Interestingly, we had been contacted by Amazon in early December to take part in an early project, which I’m now realizing was this. I turned down that offer because we were already 4 months into our new site and it was not reasonable to redesign our framework to support such a relationship. It’s definitely interesting to finally know what they were putting together.
So, the main question is, are we now competing with Amazon? In short, no. S3 is aimed at developers who want to tap into cheap storage for their applications through an open API. You see, Box.net’s storage offering is very different than this. And I quote:
Amazon S3 is intentionally built with a minimal feature set. The focus is on simplicity and robustness.
You see, Box.net’s goal is to let sites like NetVibes.com and Goowy.com give their users a place to store their documents, photos, and files. But more importantly, since we act as a neutral back-end (and front-end), the user can move between other applications which use our storage and still have access to those same files. That is very different from what Amazon is offering. Amazon is intending their storage to act, in effect, as a web host. If you’re looking for an easy way to offload large data from your site, this would definitely be the place to go (of course you will still need a server to get the data to them).
Now that we’ve established that Amazon and Box.net are not competing against one another, you may wonder why we don’t just use their storage. Beyond reliability issues, technical limitations of the API, and the troubles of outsourcing, lies the issue of bandwidth pricing. Amazon is charging $.20/gb for upstream and downstream transfer. Here is some simple math that I just posted on techcrunch regarding the costs of a free box.net user:
Files need to pass through our servers before they are stored or distributed back to the user. This would make our bandwidth costs double (for each direction). We would be paying for Amazon’s cost, and our own. So if you wanted to store and retrieve 1GB in a month, the net effect is .40 (Amazon’s BW) +.15 (Amazon’s storage) +our BW and server costs.
Now consider sharing, where this same user needs around 10GB transfer/mo for photos/clips/files, and we end up paying $2.20+.15+our BW and server costs.
If these were our costs for an average free account, we, or anyone who offered it, would be in business for no more than two months. Uncoupling storage from the application makes it impossible to have a practical business when you’re paying for the bandwidth. All in all, this is why neither Amazon nor a web services API developer will be competing with Box.net (or google).
Update#2: John has informed me that there may be around the double-bandwidth scenario. This may be possible in situations when the user needs to get straight to their file, however we do a lot of localized processing of the data before the user or the storage sees it.
Update: Michael Arrington and Rob Hof have their comments on this release.
By the way: Pending a company-wide decision, I think that “Sync without the think” will be the winner of our sync slogan contest.
Amazon 推出 S3 (Simple Storage Service)
Amazon 推這個有什麼目的 ?_?
剛看到報導，說是 Amazon 推出了一項新的 web service API 叫做 “S3” 的，取其 “Simple Storage Service” 之意。這是一個收費服務，提供了 SOAP 和 REST 的 API，每…
March 14th, 2006 at 10:07 am
It looks like Amazon is trying to address the bandwidth issue by including an interface to BitTorrent’s protocol. Is that a feature Box.net will implement? Will competitors like Google have to offer BT as an option too?
March 14th, 2006 at 10:14 am
Yay! My slogan’s getting used
March 14th, 2006 at 10:20 am
One of our “sister” companies is Redswoosh, a p2p provider similar to BT- so it’s definitely a route we’ve considered. The main issue is that most of our users share their files with a limited number of people, so P2P is not always efficient in this scenario. Will google go that direction? It seems that with their investments in broadband, they might just end up owning the pipes. More on this in a future blog post
March 14th, 2006 at 10:23 am
You’re position statement kinda makes me sad. I was hoping to use box.net for storage for my browser extensions, syncing files like say with scrapbook (http://amb.vis.ne.jp/mozilla/scrapbook/) and even building a simple remote drive… Guess I’ll check out amazon.
March 14th, 2006 at 11:12 am
Check out our query string authentication mechanism. It’s designed to solve the exact “double paying for bandwidth” problem you mention. You should be able to sign a REST request with your application’s identity, then serve your customer a link or tag directly into S3.
http://www.amazon.com/gp/aws/sdk/104-1775788-0244713?v=2006-03-01&s=AmazonS3
March 14th, 2006 at 1:30 pm
hey..
so which entries are in the short-list for the company tagline competiton?
March 14th, 2006 at 2:24 pm
Andrew,
I should have been a bit more specific about my post (I have now revised it to make it very clear what I mean). Box.net is not intended to be the equivalent of buying 100gb on a server somewhere. We are focused on consumers and consumer applications. That means we cannot benefit in any way out of someone using us as an invisible backend to store files from their site. That is what a web host (and now amazon) does. We look for opportunities where a user will rely on us for their photos, documents, and important files for remote access, backup, or sharing.
Building a simple remote drive via our API is exactly the type of functionality we are looking for. It is something we can eventually promote in our “widgets/modules” section… Let me know if you have anymore questions.
March 14th, 2006 at 2:42 pm
Interesting. I’m glad you didn’t pair up with Amazon, because like you said, Box.net and Amazon’s services are different.
March 14th, 2006 at 4:21 pm
Trying to wrap my arms around this but I’m having difficulty. I suggest that you try one more time for those of us who don’t quite get Box.net versus Amazon.
Here’s my stab at it, which of course tells shows you what I think you are saying:
Box.net = “hard drive on the web”; for humans; general access via browser but also through various web applications (still mostly by humans); web site and application independent
Amazon = web hosting / storage web service; for web sites and applications (machines and developers but not end users); access via APIs and code, web site and application dependent
March 15th, 2006 at 1:32 pm
[...] Well, box.net now seems to be open. In true Web 2.0 style, it has banners and a blog. An example of a box banner appears in this very post. To add some detail to the banner, box.net is “a virtual storage space for your computer files.” An example of box blogging is this post on how box.net differs from Amazon’s new S3 service. [...]
March 16th, 2006 at 9:55 am
I think I’m finally getting it. I believe things will get clearer as folks build apps on this.
March 16th, 2006 at 5:37 pm
Aaron,
Great post! keep it up the good work!
Your friends goowy!
March 16th, 2006 at 11:08 pm
我觉得你们比亚马逊做得好
March 19th, 2006 at 8:29 am
[...] I’m doing a lot of research on the online storage market recently. There is a lot going on here. Still no real reliable provider yet – I mean really reliable. What struck was this statement by one of the very successful consumer storage-providers. Box.net’s Aaron states. If these were our costs for an average free account, we, or anyone who offered it, would be in business for no more than two months. Uncoupling storage from the application makes it impossible to have a practical business when you’re paying for the bandwidth. All in all, this is why neither Amazon nor a web services API developer will be competing with Box.net (or google). (Box.net:”Are we competing with Amazon?“) [...]
September 3rd, 2007 at 5:15 am