Enter your email address:

Delivered by FeedBurner

About This Blog

Josh’s Tweet Stream

  • More tweets

« The Splinternet means the end of the Web's golden age | Main | Take our survey, score your HERO project (social media, mobile, or other tech) »

January 28, 2010

Proof the Splinternet is real

by Josh Bernoff

Splinters Since I published the post about the Splinternet Tuesday, a lot of discussion has been flying about. See the comments on that post, the post on Advertising Age, and my interview with Kai Ryssdal on Marketplace.

First, let's be clear. It's the Internet applications that are splintering, not the Internet itself. The most splintered of these applications is the Web, but there are others (tried running Skype on your iPhone lately?).

Second, it's wonderful that standards like HTML5 are coming along, but that won't change some things. Apple will still decide what apps can run on its iPad. Facebook will still require a login and much of its content won't be visible to Google. And the iPhone screen will still be a lot smaller than most Web pages expect. This creates dilemmas for site and application developers -- new decisions to be made. HTML5 will not miraculously cure that.

To clarify what I mean, let's invent the "Splinternet Index". This is purely notional (you can't accurately  measure most of these things), but ask yourself the following questions:

  1. What proportion of all Web page views on all connected devices yield an experience that isn't even close to what its creators intended?
  2. In absolute numbers, how many Web page views yield a page that doesn't render the way its creators intended?
  3. What proportion of content owners time (this includes content sites, marketers, anyone delivering interactive content) is spent deciding which platforms to support and customizing content for those platforms?
  4. What proportion of content owners' delivery of content goes to platforms for which standard analytics tools don't exist yet?
  5. What proportion of content owners' interactive budgets is spent on delivery to platforms where content is controlled by the platform owner (e.g. iPhone/iPad apps, Facebook applications, Xbox content)?
  6. What proportion of "interesting" Web content (content that the majority of people might reasonably be seeking) is hidden behind a login and inaccessible to search engines?

I contend that all of these numbers are the highest they have been since 1995. I also believe that they will be higher two years from now than they are now, regardless of what standards may become accepted. That is why the Splinternet is here to stay.

Photo credit: Lars Plougmann via Flickr


TrackBack URL for this entry:

Listed below are links to weblogs that reference Proof the Splinternet is real:


Steve Schildwachter

Josh, these two posts about the iPad and the Splinternet are brilliant. My opinion is that it doesn't matter if the iPad succeeds, because Apple fails wisely and they would use failure to invent something even better later. What DOES matter is whether the iPad permits us to engage consumers. Technology won't be a barrier -- if it's a good place to engage consumers, we will find the technology to allow it.

Larry Port

Since the conversation's migrated I'm putting my new comment over here.

I know that you really like this word Splinternet and are trying to create an idea, but you gotta have some meat behind it and some buy in from those of us on the front lines of web content delivery. To us, it seems like you're trying to jam an idea into discourse that has very little connection with reality.

Apple doesn't control all the apps on its platform. It only restricts the native apps on its platform.

An app delivered over mobile web is not restricted in any way. In fact, Apple has gone to great lengths to make Safari execute the same Javascript and HTML rendering we enjoy on a desktop machines.

And it takes minimal engineering to reformat content to adapt to different mobile devices.

So unless Apple gets all China on us and starts blocking content, anyone can develop a cross platform app for any device.

The only merit to the idea is the one about content being isolated behind logins like Facebook. But we have to call that the splinternet? Figure something else out to get your book deal.

Just my two cents.


Josh Bernoff

Larry: already got my book deal, and it's not about Splinternet.

140,000 iPhone apps are running only on Apple products. What were those app creators thinking? I suppose, that this platform was worthy of a special effort.

In other words, they fragmented their effort for a different platform.

If you think it takes minimal effort to turn a Web site into a mobile site into an iPhone app into an Android app, you're talking to different developers than I am . . .

Larry Port

You're right. We are talking to different developers. You never talked to me. I'm one of them.

I hate to break it to you but many of those 140k apps just point you straight to Safari. A bunch more leverage the SDK's embedded web browser.

The difference between creating a HTML mobile version and a native app is huge. Much easier for devs to leverage familiar tools than to learn a new environment for each platform.

Congrats on your book deal, let's hope the topic is less splintered than this one.

Just my two cents.


Ben Kunz

I agree with you Josh. The Internet is turning into a commodity -- like an electrical utility network -- and the new focus is on different ways to plug into it and different modalities for using it. It is splintering just as you could say electrical devices have splintered into a million different products. The web is no long a "thing" -- it is an information power source.

The emerging "splinters" you talk about really are major brands competing for different user modalities. Facebook wants to own your friend modality; Microsoft your work tools; Hulu and Apple and Netflix, your entertainment mode; Google, your search. Each of these companies is trying to build new gadgets that become entangling devices leading to their splintered information ecosystems, which really drive the revenue. Google's Droid phone is the perfect example -- a new cell phone with a hot key that boots up Google mobile search. The Kindle pulls to the Amazon splinternet, the iPad to Apple's.

The funny thing about this is the old 1990s "portal" strategy in which firms such as AOL and Prodigy tried to build sticky content to keep their customers in their little private fields is now back again. The challenge I see is as the gadgets that are the front doors to these competing splintered ecosystems keep evolving, it is very hard to maintain leadership in any position. Google has the most to fear, I believe, as users begin using mobile handsets with single apps that get them online avoiding the old Google search window.

Very strong thoughts, Josh. Keep it up.

Mike McG

The author seems to be having a knee-jerk freak-out over the fact that there are now a bunch of isolated platforms that businesses can choose to target. "Should we make a Facebook app? An iPhone app? What about Blackberry/Android/Palm? What about a regular old website? Should it use Flash? Silverlight? HTML 5.0?" I don't think it's as out-of-the-ordinary as he makes it sound. This is a new, but small, frontier in digital communication. There are risks involved: Which platforms will be around tomorrow? Which platforms are inordinately expensive (in development resources) to develop on? Will the approaches our business takes to developing on multiple platforms be maintainable on, or scalable to, new platforms?

The real reason he is freaking out is that the traditional businss model is unstable due to are a plethora of new enabling technologies. But this instability is basic innovation/capitalism. The challenges described above can also be looked at as opportunities. For any consumer group, the company that can answer these questions most efficiently and effectively will ultimately reach and enrich the most consumers, and will dominate. In meeting these challenges, it is likely that they will innovate ways to mitigate/minimize the apparent volatile-platform risk of doing business on the "splinterweb" (likely including as-thin-as-possible platform clients, SOA and agile design, enterprise-wide UI conventions/frameworks per platform). At the same time, the platforms that provide the most value to business and consumers will dominate.

It's market economics, not magic. Innovation, consumer demand, innovation, ad infinitum.

Josh Bernoff

Mike, all I'm saying is that marketers that are used to counting on all this infrastructure on the Web are about to enter a whole bunch of new worlds.

It's all well to postulate that they will "innovate ways to mitigate/minimize the apparent volatile-platform risk" but in the meantime marketers need to get stuff done in this new world, and they haven't sufficiently understood how different it is.


Josh, I humbly suggest you're stubbornly clinging to your FUD in the face of testimony from knowledgeable folks that it just ain't so!

There's no splinternet. And there won't be. At the edges, some movers and shakers will try to keep content fenced in, but the meta-trend is open standards.

If various C-Level executives could stop being distracted by shiny tinsel for more than 20 seconds, they'd probably be able to see the solution is systemic in nature: Build for openness.

As it is, you'll have to have competent IT staff who understand the nature of information and networking. Not buzzword-spewing meme hounds hyping up important of the latest proprietary platform like Zune or MySpace which will go the way of the dodo all too quickly.

Mike McG

Ok, I think that's a fair and interesting point. It just sounded a little sensational to me.

Josh Bernoff

ReaderX says the megatrend is for open standards.

I say the megatrend is for incompatible devices and sites that people love.

Do you think the iPhone is going the way of the Zune?


With all due respect, many of the points raised to counter the author's argument are not counter at all.

Yes innovation and consumer are driving these developments, and sure they represent opportunities. And open standards or open APIs are beginning to underpin most well-designed applications in order to make them flexible so they can be 'ported' (ie not work seemlessly on) to a variety of platforms.

Neither of these arguments really prove that the splinterweb is jargon, in fact it seems to me that they just show that programmers and product developers are already addressing a splintered market, and a technology development path that is not going to result in the googleweb, much to everyone's relief I imagine.

As far as I can tell, the "open standards" vs. "proprietary platforms" divide is the dichotomy that is making the splinterweb.

Josh Bernoff

@Miriam Nice to see somebody who sees my point and doesn't think I'm an evil standards hater. I'm not!

Jeff Wilfong

The splinternet is amazing. Have you seen the latest drawings and analysis by Barabasi on the topology of the internet and web. Truly fascinating. A mixture of public and private necessitates the architecture. However, this by definition privileges info.

Shanty Mathew

When the Wild West, a new frontier was settled, fences first came up around the most productive patches of land. On the web, it'll be around the best content.

Up to now, content on web had been largely free. Now that this content is more in demand than perhaps any other medium, content creators will naturally look to monetize it - either directly or by advertising.

We don't know if it'll work... But we know that the web isn't the same as it was 10 years ago, and it probably won't be recognizable 10 years from now.


@jbernoff I like your use of the term "Splinternet", and this clarification of the earlier post as the splintering of Internet applications.

It made me think about definitions of interoperability -- I relooked at http://www.sutor.com/newsite/blog-open/?p=1372 -- and see the think about the overhead that is being introduced both to consumers (who will get locked into a platform) and to software developers (who will have to go through the expense of building for multiple platforms).

I agree that the Splinternet index is currently high. I'm unsure if it will remain high in two years, but would think that in 5 years, we might have revolutions through open standards. The philosophy towards open behind Google (e.g. with Android) and Ubuntu (e.g. I've noted that I should be able to access my iPod Touch under 10.4) demonstrates communities who maintain an interest in openness.

Thinking historically, we might consider the evolution of PDF. While Adobe still has the mindshare around the file format, alternative paths for accessing that content have emerged.

The comments to this entry are closed.