Yesterday App-Rising.com co-hosted its first event in DC, joining together with fellow co-host EDUCAUSE and conference organizer the Congressional Internet Caucus Advisory Committee to put on a preconference seminar called "The Future of Broadband: Moving from Why to How."
I'll work on writing up thoughts on the other sessions to post shortly, but first I wanted to review what was said on my panel
The intent behind this panel was to engage a voice that's too often ignored in broadband debates, that of applications developers, in order to discuss the demands their applications have for bandwidth.
The reason I feel this is an important topic to consider is that by understanding more of these specifics I hope to equip my fellow broadband believers with the ability to express the need for broadband in concrete rather than abstract terms.
And seeing as how video is the ultimate bandwidth hog, I focused this panel on video applications.
Joining me onstage were Tom Spengler, CEO of Granicus, a company that combines live and on-demand video of local government meetings with meetings notes, agenda, and votes to create what they've dubbed an integrated public record; John Hughes, CEO of GOSN, a company that is working to introduce a new era of Internet-empowered monitored video security for communities that actively prevents crime; and Gary Bachula, VP of External Affairs for Internet2, the big, big bandwidth private network that connects hundreds of university campuses, government agencies, and corporate research groups with massive connectivity.
After giving each the opportunity to share a 30-second elevator pitch about their respective companies, I asked each more specific questions about what they enable.
Tom shared that Granicus has more than 500 communities as customers and streams more than 1.5 million videos a month. He also alluded to the fact that many of their customers are finding new ways to use Granicus's platform to integrate online video into the workings of government, for things like distance learning, for example.
John described a typical SafetyBlanket deployment of motion sensitive cameras on buildings, pan-tilt-zoom cameras on light poles, all tied to 24/7 monitoring services. He also confronted the question of if his product is taking us to a Big Brother, nanny state by stating that the SafetyBlanket is an external security system that only sees what's already in the public eye.
Gary discussed the roots of Internet2 as an attempt to create a private network that was five years ahead of the public Internet, and he expressed his dismay over how much further than five years they have become through their continued investment in capacity. He also mentioned the fact that while Internet2 is a tremendous research tool, that its use is becoming woven into the fabric of students' and faculty's day to day lives.
Next we dove into the topic of bandwidth, exploring what the minimum and maximum connectivity each app requires.
For Granicus, that means 300Kbps for delivering video. They can serve customers slower than that, the video will just continue to degrade until bandwidth becomes so scarce that their system will flip users over to an audio-only stream. By doing this, though, they're able to serve even those customers still on dialup. He also shared that they could ramp up the bitrate of their video as high as a customer would want, the challenge is that the majority of their users are connecting at very low speeds and are therefore unable to receive anything higher.
For GOSN, while the SafetyBlanket is able to work over lower speed connections by sending JPEGs instead of video, it fundamentally is a high bandwidth application. In fact, to max out the capacity of the cameras requires 80-120Mbps per building.
With Internet2, I had Gary discuss the range of two-way videoconferencing technologies in use by people on their network. On the low end they start with traditional Polycom units streaming at 1.5Mbps. For those with newer units that number can jump to 3, 4, 5Mbps, or DVD quality headed towards HD. He mentioned the increasingly common use of uncompressed MPEG 2 video streamed at 30Mbps. Then he talked about the really neat use of video conferences where 30 people stream video to each other at the same time, requiring about 2-3Mbps per user. He finished by taking things to the highest end of uncompressed HD at 1.5Gbps and even the experimental quad-HD at 6Gbps for one stream.
I found these three examples to be a fascinating juxtaposition of the amazing things that can be done with low connectivity, contrasted against the tremendously bandwidth-intensive applications that in my mind put to bed any notion that we're not going to want at least a gig to every home in the not too distant future.
Then I asked the question: if you were king for a day, what would you do to improve our country's broadband infrastructure?
Tom shared his company's vision for expanding and enhancing participation in local government, and that to accomplish those goals we need everyone to have access to broadband and to get everyone onto the network.
John expressed his company's need for a more secure, stable Internet, one that can guarantee the delivery of video from his system to the monitoring center as he and his customers can't afford to lose signal in an emergency.
Gary's main comment was quite simple: you can never be too rich, or too thin, or have too much bandwidth. He backed that up with the fact that every time they've increased capacity, new demand has filled it up.
While there were many other interesting items shared during this session, the last two I wanted to cover here are:
- Tom's suggestion that if want to spur adoption of broadband use in government, we should be looking at creating something like an e-rate program that could fund government to purchase and integrate broadband applications.
- John's line of thought that historically alarm systems take over the phone line during an emergency, and that the SafetyBlanket does the same, only this time the broadband connection. This idea led him to defend his need to be able to prioritize traffic on a network in order to be able to guarantee video from an emergency can make it to the monitoring center, homeowner, and public safety officials.
All in all the message seemed to be clear: in order to support the continued development of broadband applications we need to make sure everyone has access and is connected, we need smarter more stable networks, and we need lots more bandwidth.
A good time was had by all, and I'd like to again share my gratitude with the Congressional Internet Caucus Advisory Committee for deeming this content worthy of a preconference event, and to EDUCAUSE, fellow broadband believers that I was honored to share the stage with.
If anyone has any questions about the information contained in this post, submit a comment and make your voice heard! I'll do my best to get whatever answers you need.