Nobody will win the storage war | Hacker News Reader

Nobody will win the storage war

blog.filepicker.io

48 pointsbrettcvz28 comments

> as Steve Jobs has proven, tight integration between components results in better user experience.

A note to the younger readers out there. Talking about "tight integration" like this is fine, but some people who have heard the term "tightly coupled" apply it to the same thing. It's not the same thing. Apple's APIs and SDK are very good because they are generally NOT "tightly coupled." "Tightly coupled" generally means that it's not so great about respecting encapsulation. (Sometimes the letter of the law is upheld, but the spirit is violated.)

(I like the term "tight integration" as it's the same general meaning of "tight" when people talk about bands.)

Also,this does not seem to be a market where small players are trying to unseat a large incumbent. Strategically, Google, Microsoft, Apple, Dropbox and Box are all in incredibly strong positions. Thus the market is unlikely to consolidate into one or two platforms for a long time. It may very well be that we see a similar pattern as we saw in evolution of PC platforms where it took decades before Mac, Windows and Linux emerged as the dominant standards. Even then, developers have had to work on cross platform compatibility for a while now.

i've always thought dropbox should play more on its strengths as a technology layer. When I think of a successful dropbox I think of all my critical applications having a logo in their startup banner saying "powered by dropbox" proudly displaying the dropbox logo. But thats where it could end. Why do I need to have a dropbox account? Why do I care about a quota? Can't the application just take care of all of that behind the scenes? All of things that make dropbox great could be built within the context of that application (sharing with other users of that application, live syncing with collaborators of that application). You could optionally sign up for a dropbox account which would just allow binding an app's dropbox-powered datastore to your global id (with permission from the app).

Except dropbox's technology layer isn't infrastructure. They do sync, but that's pseudoinfrastructure compared to what s3 really provides them.

Fragmentation of providers in markets can indicate that consumers see the product as a commodity.

Storage is becoming more and more of a commodity. It's not a good or bad thing, it just means that storage providers might benefit by adopting strategies that work in other commodity businesses.

I agree with both your premise and conclusion regarding the on going commoditization of storage. The OP lays out application integration and software integration as routes for storage providers to follow, which I tend to agree with, but the ideas can be better generalized as problems of horizontal and vertical integration (respectively). Different forms of commodity businesses exists in both the horizontal and vertical markets here, so as you point out, storage providers are likely to begin searching for customers with problems that they can build/grow into the function of their platform.

Consumers will.

Disagree.

As we continue to travel up (down ?) the layers of abstraction from actual files in a filesystem to "pipes between dropbox and box.net[1]" it will be more and more difficult to maintain any bearing on what is yours, where it's at, and how to use it.

Fast forward a few years from now when half of these players no longer exist, the other half have either been absorbed ("Dell Home Storage, powered by DropBox!") or have "pivoted" because they never made money in the first place (you know damned well that someone inside of dropbox is scheming up some "social" integration on a whiteboard).

The consumer, who circa 2012 has been deemed too stupid to even understand a folder, will lose.

[1] Whatever that means ...

Agree

Increased competition for a commodity will decrease prices for the consumer.

Or, we could end up with many different services with similar-yet-ultimately-disparate feature sets that people just give up. The consumer wins when she can use competing products together without jumping through hoops - Someone mails a word document, she saves it to Dropbox, and opens it with Pages on her iPad.

We need open standards and competing implementations. It's an absolute mess right now.

http://xkcd.com/927/

I just want open standards!

People often tend to forget that storage as offered by the likes of dropbox is not a product, its just a feature that makes a product better. With deep app platform integration that iOS, Windows have brought in, there will be less and less need for people to use an external store thus making these guys irrelevant...

What is Dropbox's product in your opinion then?

Would respectfully disagree. Dropbox and Box are building an awesome app platform as well and as long as users want neutral solutions (non iOS or Windows), they should be OK. In fact Box and Dropbox will start specializing as the author noted.

When you say neutral solutions, you are assuming that the API developed by Apple and Microsoft can only be accessed from these platforms. I know that that's not true at least for SkyDrive API.

Given these why would I choose a neutral API?

Try using any Apple product on Windows, or any Microsoft product on Linux. It might work, but it's often quite painful.

I definitely benefit primarily from Dropbox's neutrality. Much as I benefit from the neutrality of lastpass and xmarks. I can use them at home on the latest Mac and Linux, I can use them on personal VPS instances, I can use them at work, even in a fairly locked down network environment.

iCloud is useless at work, and my officemate has switched from SkyDrive to Dropbox for the same reasons. We now both doubly benefit from the wide-ranging accessability and sharability of Dropbox.

are you confusing neutrality with it being a web app? What part of the above things you could not do with say skydrive?

Basically I don't see a difference between dropbox and skydrive (or gdrive or....) as a developer, if I had to choose, I'd probably choose a platform from a company that has a long history of building platforms (Microsoft or Google or any other big co) as opposed to dropbox which is a tiny startup somewhere with no viable long term life expectancy.

Dropbox has a Linux version. It's not a web app. I've used it in Unity/Gnome/Nautilus. It feels exactly right. It also works perfectly on OS X, Windows, Android and iOS.

dropbox is a hell of an amazing, slick product. Although I see it getting cheaper, I don't see it being turned irrelevant by the likes of iCloud or Skydrive (although Skydrive is making a decent job on being platform independent - but so far, it's just "another dropbox")

I want google drive but with the widespread use of dropbox.

You're making two big assumptions.

First, that Microsoft or Apple are able to perfect these services, which so far they have not been able to do, and there is good reason that it may be impossible.

But assuming that it's good enough, you still need to commit entirely to one ecosystem. Certainly some people can do that, but some people can't, particularly as BYOD becomes more common in business.

This is the magic of Dropbox. They make a dead simple product, they make it available everywhere, but they don't overreach. So in the end you have something that's harder to use and can't get the novice adoption that Apple brings, but on the other hand, most people understand what a folder is these days, and what Dropbox does is conceptually several orders of magnitude easier than trying to figure out exactly how iCloud works and what edge cases are going bring about catastrophe.

The old "that's a feature not a product" line is clever, but it's been overused to death. I don't know the source of the quote, but I'm pretty sure it was originally to address the fact that a startup needs to have an asymmetric attack to succeed against large incumbents. If something can be trivially replicated then it can be copied by the large company and smothered simply by the marketing heft of the goliath. If on the other hand, the very nature of the product is something that has a parity mismatch with large companies, then they can not directly attack it. Dropbox has a bit of that because making a cross-platform service like that just isn't worth the money for an Apple or a Microsoft. It doesn't tie into their cash cows and it's not a big enough market by itself. Dropbox is not a direct competitor to them anyway, it's more of a symbiotic relationship, so who's going to kill them?

>you still need to commit entirely to one ecosystem

I don't know why you said that. Skydrive API at least are available to most popular platforms it seems.

Because I was thinking of iCloud and I don't really know anything about SkyDrive. Forgive me if I'm skeptical that Microsoft is gonna blast that one out of the park.

How is it not a product? Technically I would call it a service but regardless, how is it not a product or service?

There's some truth to that w/ respect to the "cloud" storage services out there. Basically it costs $X dollars/month to host your data in a datacenter (with bandwidth and cooling costs). So what we have now are several providers that basically offer the same service (a special folder and/or syncing) for roughly the same price.

Look out in the next few months for companies that will disrupt classical client/server datacenter based storage with peer-to-peer protocols and in home devices (which don't have the same cost constraints that datacenters have). These product offering will in some ways be similar to dropbox, icloud, etc, but they will be an order of magnitude cheaper and an order of magnitude faster.

I work at one of these companies. http://spacemonkey.com/ If you're interested in P2P storage and distributed systems check us out, we're hiring.

A system like that will never work.

Reply on news.ycombinator.com