Why is this page text-only?

« The Germans Are Coming To Take Away Our Applications Developers (And Show Us The Future)! | Main | Complexity Prevents Change, Increases Opportunity »

April 28, 2008 12:22 PM

Developers Talk About Our 100Mbps Future at the Web 2.0 Expo

There’s constant talk among advocates for fiber about the need for America to realize its 100Mbps future as quickly as possible. While I’ve long been an advocate for fiber, I’ve also questioned what can really be done with all that bandwidth given the fact that the vast majority of what’s on the Internet today can work with far less. Back when the FTTH Council put out a call for a 100Mbps Nation, I responded writing about the possibilities of a 5Mbps Nation where everyone has and is actively using a robust, symmetrical 5Mbps connection.

The simple truth is that outside of P2P applications, high-end videoconferencing, and some home video security products like the SafetyBlanket, there isn’t much out there today that demands 100Mbps connectivity.

But that doesn’t mean there won’t be in the future. And I’ve long considered it part of my role as an advocate for fiber to try and inspire applications developers to start thinking about what’s possible in a full fiber future as networks are going into place that will feature this level of connectivity, such as in communities like Lafayette, LA.

So to that end, I set out last week in San Francisco at the Web 2.0 Expo to pose this question to developers: how would a world where 100Mbps is ubiquitous change what you are and can be doing with the applications you’re building?

I think my favorite initial reaction was from Danny Kolke, founder of Etelos, a company that provides a platform that makes developing, distributing, and customizing applications easier than ever. His first response to the potential of a 100Mbps future was, “Holy cow!”

He continued on to postulate that their “apps on a plane” technology they were promoting at the Expo, which allows web apps to eschew the need for connectivity and run as desktop applications, wouldn’t have been necessary as a world with ubiquitous 100Mbps is one where you’re likely never not connected. So here’s an example of how more bandwidth would allow developers to not have to focus as much energy on how to deliver apps in an environment where bandwidth is restricted.

Also interesting was Kolke’s somewhat worried observation about how all this bandwidth will likely lead to a new wave of networked technologies that will all demand more and more power. I tried arguing for the possibilities of how the use of broadband can reduce our overall energy consumption--as I wrote about recently here--but I can’t deny the premise of his concerns. And it made me wonder: will the new demands broadband-enabled technology creates for energy outweigh the savings its use can bring about, especially as we move into a world of 100Mbps to the home?

I’ve long stood by the observation that part of the biggest challenge of advocating for a 100Mbps future is that no applications developers are thinking about building applications for that kind of a bandwidth-rich environment. And while that generally held true at the Expo, where many of the people I spoke to seemed to have no idea that full fiber networks were taking hold in this country, I’m happy to report that at least one developer when posed this question immediately shared that that’s precisely the future they’re building towards.

That developer was Aaron Fulkerson of MindTouch (who I wrote about here). He and his team come from the world of distributed computing and hardcore research, so nothing excites them more than the thought of being able to operate in a world where bandwidth constraints go away. As we chatted briefly about the possibilities, Fulkerson began discussing the potential for a future where the wikis their platform enables are no longer stored in a server in the sky but instead the application could “literally exist within the network leveraging the storage of the individual machines of our users.”

He also touched on the economic potential of being able as an application developer to distribute applications from your basement. That’s long been a promise of the Internet but not as yet much of a reality due to bandwidth constraints. But in a world where 100Mbps is ubiquitous, there’s potential for servers to be located anywhere, meaning those entrepreneurs building the next Google in their garage could actually be delivering their apps right from that garage.

Michael Hughes from xtranormal (who I wrote about here) and a handful of others cited the ability for them to expand their applications into really high resolutions with that much bandwidth available. So no longer would xtranormal’s 3D movies be limited to YouTube-sized windows, but instead users could be creating, sharing, and viewing fullscreen videos, which is something he was quite enthused about.

While everyone was excited about the potential of 100Mbps, some struck a cautionary note. While at the SixApart evening reception I got to chatting with one of their programmers who’s first name was Graham but last name remains a mystery due to his lack of business cards. During an extended conversation on the subject he shared that one double-edged aspect of having that much bandwidth is that it may allow people to get away with using sloppier code since they won’t have to worry about squeezing everything down so much, but at the same time the challenge is that once something goes wrong in an application with that much bandwidth it’s likely to go really wrong.

The highlight of that evening for me came later on when I hit up the party at Rearden Labs and had the tremendous opportunity to chat with Steve Perlman. Listing all of his exploits in the tech world requires far more words than I have room for here, but suffice it to say he’s a true tech pioneer, one of the most effusively brilliant people I’ve ever met, and seemingly an incredibly nice guy on top of everything else.

His response to my question covered a lot of technical ground but the gist of it was that everything you use the Internet for today was designed with T-1 speeds in mind. And we’re not just talking about applications, even basic building blocks like the codecs that are used to compress video have been built for an era of bandwidth scarcity. So in his estimation taking full advantage of 100Mbps means having to basically reinvent everything that’s online. Alongside this he also observed that with that much bandwidth other bottlenecks will appear, with things like processing power now becoming a limitation.

But in talking about all this, he began alluding to his belief that broadband as we know it today is already reaching an inflection point. With speeds above 5Mbps and especially once they reach 10Mbps he sees the opportunity to enable revolutionary experiences that can already go above and beyond the expectations of the Internet of today without having to wait around for 100Mbps to show up.

In particular, the work his team at Rearden Labs is doing to enable a new era in motion capture is being built with this new era of abundant bandwidth in mind. While they were frustratingly tightlipped about what they were actually working on, I do know that it’s impact will range from movies to games to potentially interpersonal communication, allowing each of us to have our own digital avatar that’s mapped to our face to represent us in online worlds when communicating with others.

In the end, the responses I got were a mixed bag. On the one hand I confirmed that there isn’t much today that really needs 100Mbps and most developers are still focused on building bandwidth-light rather than bandwidth-intensive applications. But on the other there are some developers who have already set their sights on building applications that can take advantage of a big broadband future.

So overall I have to say I am encouraged. People are beginning to understand that this potential future is fast becoming a reality, and I’m confident that eventually we will have many applications that demand the capacity of fiber.

I’m going to continue asking applications developers this question, and I’d like to encourage any that are reading this post to add a comment about how a world where 100Mbps would impact the work that they’re doing. Because if nothing else, I’m a strong proponent for the school of thought that if we don’t dream big about the possibilities, they’ll never become a reality. So let’s think big and figure out what’s possible in a full fiber future.

Del.icio.us Digg Yahoo! My Web Seed Newsvine reddit Technorati


TrackBack URL for this entry:

Comments (1)

Try backing up a 100gig disk drive over a 5meg connection, and then tell us that "The simple truth is that outside of P2P applications, high-end videoconferencing, and some home video security products like the SafetyBlanket, there isn’t much out there today that demands 100Mbps connectivity."

Posted by Miles Fidelman on April 29, 2008 4:57 PM

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)