Wednesday, August 20, 2008
Faulty towers
The problem with gazing out from the ivory towers of academia is that the real world is often out of focus, far beneath the gaze of the theoreticians.
Such is the case with some analyses of net neutrality. Andrew Odlyzko, of the School of Mathematics at University of Minnesota recently wrote a paper [
pdf] that started with a faulty premise:
Oh yeah.
Remember, the fact that traffic is already overwhelming some of the networks is the whole issue. The traffic on internet access and backbone networks has been growing, thanks to streaming media and file sharing and gaming, etc. This isn't building a stadium in Iowa and hoping they will come. The league is a success; the stadium can't handle all the fans and more games and concerts are getting booked and tickets keep getting getting printed. And there is a genuine shortage of real-time needs like food, beer and washrooms [your contributions to the metaphor are welcome].
Service providers are looking for ways to manage the quality of the user experience for all the wonderful applications and more that we haven't even heard about yet.
Even the abstract for the paper seems fundamentally flawed.
Real-time streaming may or may not may not be the right way to deliver movies, but it is hard to imagine another way to effectively deliver live action events like sports. Imagine, if you will, the appeal of each of us being able to choose which of the hundreds of cameras we want to follow at any given time from the Olympics. Sounds to me like a call for real time streaming.
The paper advocates use of "faster than real-time" techniques for video, suggesting that YouTube is a counter example for a statement attributed to John Chambers that transmitting video over the internet is "really, really, really difficult." YouTube quality isn't going to cut it for most of the content I want to see.
The key problem I have with his premise ("that delivering movies in real-time streaming mode is the wrong solution") is that neither service providers nor academics will get to choose what the right solution is for delivering movies over the internet.
Users and entrepreneurs, small time innovators and the market-place will decide based on all sorts of factors. For someone advocating net neutrality, it is surprising to see Odlyzko suggesting that there should be a "right solution" for video delivery. Every solution that customers choose will be a right solution.
That perspective shows an affinity for centralized control of internet applications - which is antithetical to the concept of the internet's open innovation.
Technorati Tags:
Andrew Odlyzko, net neutrality
Such is the case with some analyses of net neutrality. Andrew Odlyzko, of the School of Mathematics at University of Minnesota recently wrote a paper [
pdf] that started with a faulty premise: What if you build it and they don’t come? That is what happened with the landline and underwater cables of the telecom bubble of a decade ago, and many other seemingly promising technologies. And that is almost bound to happen if net neutrality is blocked, and service providers do what they have been promising, namely build special facilities into their networks for streaming movies.Internet video is nothing like the technology bubble of a decade ago. Building it and hoping they will come isn't the problem here. Video traffic on the 'net has already arrived.
Oh yeah.
Remember, the fact that traffic is already overwhelming some of the networks is the whole issue. The traffic on internet access and backbone networks has been growing, thanks to streaming media and file sharing and gaming, etc. This isn't building a stadium in Iowa and hoping they will come. The league is a success; the stadium can't handle all the fans and more games and concerts are getting booked and tickets keep getting getting printed. And there is a genuine shortage of real-time needs like food, beer and washrooms [your contributions to the metaphor are welcome].
Service providers are looking for ways to manage the quality of the user experience for all the wonderful applications and more that we haven't even heard about yet.
Even the abstract for the paper seems fundamentally flawed.
What service providers publicly promise to do, if they are given complete control of their networks, is to build special facilities for streaming movies. But there are two fatal defects to that promise. One is that movies are unlikely to offer all that much revenue. The other is that delivering movies in real-time streaming mode is the wrong solution, expensive and unnecessary.I could start with his suggestion that service providers don't already have complete control of their networks, but I'll let that comment go for today.
Real-time streaming may or may not may not be the right way to deliver movies, but it is hard to imagine another way to effectively deliver live action events like sports. Imagine, if you will, the appeal of each of us being able to choose which of the hundreds of cameras we want to follow at any given time from the Olympics. Sounds to me like a call for real time streaming.
The paper advocates use of "faster than real-time" techniques for video, suggesting that YouTube is a counter example for a statement attributed to John Chambers that transmitting video over the internet is "really, really, really difficult." YouTube quality isn't going to cut it for most of the content I want to see.
The key problem I have with his premise ("that delivering movies in real-time streaming mode is the wrong solution") is that neither service providers nor academics will get to choose what the right solution is for delivering movies over the internet.
Users and entrepreneurs, small time innovators and the market-place will decide based on all sorts of factors. For someone advocating net neutrality, it is surprising to see Odlyzko suggesting that there should be a "right solution" for video delivery. Every solution that customers choose will be a right solution.
That perspective shows an affinity for centralized control of internet applications - which is antithetical to the concept of the internet's open innovation.
Technorati Tags:
Andrew Odlyzko, net neutrality
Posted by Mark Goldberg @ 8/20/2008 06:32:00 AM 
Comments:
<< Home
Good analysis of a paper that doesn't look too appealing anymore.
It reminds me how happy I am to have left the narrow-minded acadmic world.
It reminds me how happy I am to have left the narrow-minded acadmic world.
Posted by 
: 1:53 PM, August 20, 2008
: 1:53 PM, August 20, 2008
Mark,Mark, Mark...
If you want to criticize Odlyzko's work, the least you should do is read the paper and, more importantly, the previous research that feeds into it.
Instead, you take a couple quotations, misinterpret them to set up a straw man, and then attack that straw man. Worse, you try to dismiss him as someone "gazing out from the ivory towers of academia", ignoring his long experience with Bell Labs and the strong empirical component of all his work on the Internet and data communications.
I won't try to repeat his extensive analysis of these issues (go to http://www.dtc.umn.edu/~odlyzko/ for a complete list of papers), but someone should respond to your most egregious mis-statements.
Odlyzko does not say that Internet video doesn't exist or is insignificant. His paper says the opposite. He does say that while video is a major source of overall traffic, it is much less significant as a source of revenue and doesn't need special treatment (through real-time streaming).
The mistake in the first bubble was equating the volume of data traffic with its value (either to consumers or businesses). A point he made before the last bubble burst -- again, based on looking at the empirical evidence -- and on which he was dead right.
This is not theoretical guesswork, but based on the actual behaviour of consumers and the capability of technology to deliver solutions. Dismissing YouTube because of the current quality level misses the point. The YouTube model (faster than real-time downloads) can improve quality up to the level of the bandwidth available.
What it demonstrates, along with P2P, is that consumer behaviour demonstrates this model is the most successful for delivering video. It is not the "right solution". It is the solution consumers have, to date, adopted.
Odlyzko also makes a broader argument that this will remain the preferred solution as technology evolves. If you disagree with this argument, challenge it. Don't ignore it while taking cheap shots.
Finally, the most humorous part of your post was using the Olympics as a counter-argument. Very little of the Olympic coverage is live. For NBC, much of the coverage has a latency of 12-24 hours... creating the ironic situation where, at one point on the weekend, they were running promotions for Tyson Gay's 100m race after he had not only been eliminated, but after the final had been run. Hardly a defence of the need for networks to deliver real-time live action sports. (And multiple camera angles don't require live streaming either.)
If you want to criticize Odlyzko's work, the least you should do is read the paper and, more importantly, the previous research that feeds into it.
Instead, you take a couple quotations, misinterpret them to set up a straw man, and then attack that straw man. Worse, you try to dismiss him as someone "gazing out from the ivory towers of academia", ignoring his long experience with Bell Labs and the strong empirical component of all his work on the Internet and data communications.
I won't try to repeat his extensive analysis of these issues (go to http://www.dtc.umn.edu/~odlyzko/ for a complete list of papers), but someone should respond to your most egregious mis-statements.
Odlyzko does not say that Internet video doesn't exist or is insignificant. His paper says the opposite. He does say that while video is a major source of overall traffic, it is much less significant as a source of revenue and doesn't need special treatment (through real-time streaming).
The mistake in the first bubble was equating the volume of data traffic with its value (either to consumers or businesses). A point he made before the last bubble burst -- again, based on looking at the empirical evidence -- and on which he was dead right.
This is not theoretical guesswork, but based on the actual behaviour of consumers and the capability of technology to deliver solutions. Dismissing YouTube because of the current quality level misses the point. The YouTube model (faster than real-time downloads) can improve quality up to the level of the bandwidth available.
What it demonstrates, along with P2P, is that consumer behaviour demonstrates this model is the most successful for delivering video. It is not the "right solution". It is the solution consumers have, to date, adopted.
Odlyzko also makes a broader argument that this will remain the preferred solution as technology evolves. If you disagree with this argument, challenge it. Don't ignore it while taking cheap shots.
Finally, the most humorous part of your post was using the Olympics as a counter-argument. Very little of the Olympic coverage is live. For NBC, much of the coverage has a latency of 12-24 hours... creating the ironic situation where, at one point on the weekend, they were running promotions for Tyson Gay's 100m race after he had not only been eliminated, but after the final had been run. Hardly a defence of the need for networks to deliver real-time live action sports. (And multiple camera angles don't require live streaming either.)
Mark, Goldberg & Associates is a Telecommunications Consulting firm and the Canadian sector is a pretty small space.
Reading your Blog postings I'm always struck by how dramatically pro-Bellhead they are which always leaves me wondering if you are currently employed by the Telco's you so reliably defend on your blog.
I think it would do your integrity a world of good if you openly disclosed your relationships which any of the companies you blog about in each posting.
Reading your Blog postings I'm always struck by how dramatically pro-Bellhead they are which always leaves me wondering if you are currently employed by the Telco's you so reliably defend on your blog.
I think it would do your integrity a world of good if you openly disclosed your relationships which any of the companies you blog about in each posting.
Posted by : 2:58 PM, August 20, 2008
: 2:58 PM, August 20, 2008
John -
Thanks for the concern about the size of the Canadian sector. The Canadian communications space is actually pretty big - more than $50B annually according to the latest CRTC reports.
I didn't know that my integrity needed "a world of good," but if some form of disclosure makes you feel better, let me be completely clear: the views expressed are my own, in respect of this posting and all 1103 of my blog postings over the past two years or so.
I am not doing work on net neutrality for any party other than my own interests in ensuring that the same commercial interests that have delivered the power of the internet that I have used for more than 20 years will continue to drive its development for the next 20 years. I will also confirm that I believe market forces are the best regulator, consistent with the Policy Direction to the CRTC.
You can check out my website to download an intro package, if you would like to learn about my typical clients and projects.
Thanks for the concern about the size of the Canadian sector. The Canadian communications space is actually pretty big - more than $50B annually according to the latest CRTC reports.
I didn't know that my integrity needed "a world of good," but if some form of disclosure makes you feel better, let me be completely clear: the views expressed are my own, in respect of this posting and all 1103 of my blog postings over the past two years or so.
I am not doing work on net neutrality for any party other than my own interests in ensuring that the same commercial interests that have delivered the power of the internet that I have used for more than 20 years will continue to drive its development for the next 20 years. I will also confirm that I believe market forces are the best regulator, consistent with the Policy Direction to the CRTC.
You can check out my website to download an intro package, if you would like to learn about my typical clients and projects.
Posted by 
Mark Goldberg : 4:11 PM, August 20, 2008
Mark Goldberg : 4:11 PM, August 20, 2008
Mark, whether or not your integrity needs an uplift is an open question which I shall leave to your readers. I realize you are not a journalist, never the less to be fair to your readers it would appropriate to disclose in your postings whether or not you are paid by the parties who benefit from the positions you take.
I appreciate the link to your resume which does disclose a client list which not surprisingly lists all of the major Canadian telecom providers.
As for your admonishment that the Canadian Telcom Sector is not small ("over 50 billion"), you helpfully provided link which shows you are good at distorting statistics.
First of all, your numbers are wrong. The report states that combined "Total broadcasting and telecommunications revenues [were] $51 billion in 2007". The telecommunications revenue (retail & wholesale) was $38 billion.
Not that revenue is what counts. The Canadian industry is "small"because there are only handful of players who pay telecom consultants.
From the same CRTC report: "$14.0 billion or 40% were related to wireless services and the remaining $20.7 billion or 60% were related to wireline services"
In other words, a very large proportion of the telecom revenue was split between the four incumbents you list as being your clients (Bell, MTS, Telus, Rogers). The only significant players missing would be the cable companies Shaw and Cogeco.
The point is simply this; if you want to make a living as a telecom consultant in Canada, you don't go around bad-mouthing the LECs.
I'm not saying you should start criticizing your clients, but I don't think it's fair to claim your postings are independent when in fact they are bought and paid for by your clients "the same commercial interests that have delivered the power of the internet that I have used for more than 20 years"
Aside from the fact it is impossible that you have been getting internet for 20 years from the same "commercial interests", it is also a gross distortion. Most people would agree the "power of the interent" is delivered by companies such as Google and the suppliers of the "pipes" are only important in so far as they don't impede our ability to access that power.
This requires net neutrality, which is something you oppose. Which not surprisingly is also Bell's position.
I appreciate the link to your resume which does disclose a client list which not surprisingly lists all of the major Canadian telecom providers.
As for your admonishment that the Canadian Telcom Sector is not small ("over 50 billion"), you helpfully provided link which shows you are good at distorting statistics.
First of all, your numbers are wrong. The report states that combined "Total broadcasting and telecommunications revenues [were] $51 billion in 2007". The telecommunications revenue (retail & wholesale) was $38 billion.
Not that revenue is what counts. The Canadian industry is "small"because there are only handful of players who pay telecom consultants.
From the same CRTC report: "$14.0 billion or 40% were related to wireless services and the remaining $20.7 billion or 60% were related to wireline services"
In other words, a very large proportion of the telecom revenue was split between the four incumbents you list as being your clients (Bell, MTS, Telus, Rogers). The only significant players missing would be the cable companies Shaw and Cogeco.
The point is simply this; if you want to make a living as a telecom consultant in Canada, you don't go around bad-mouthing the LECs.
I'm not saying you should start criticizing your clients, but I don't think it's fair to claim your postings are independent when in fact they are bought and paid for by your clients "the same commercial interests that have delivered the power of the internet that I have used for more than 20 years"
Aside from the fact it is impossible that you have been getting internet for 20 years from the same "commercial interests", it is also a gross distortion. Most people would agree the "power of the interent" is delivered by companies such as Google and the suppliers of the "pipes" are only important in so far as they don't impede our ability to access that power.
This requires net neutrality, which is something you oppose. Which not surprisingly is also Bell's position.
Posted by : 2:38 PM, August 21, 2008
: 2:38 PM, August 21, 2008
John - I have yet to see anything that addresses the substance of the post "Faulty towers." Instead, you have now attacked me and my integrity twice.
Your comment that 'it is impossible that you have been getting internet for 20 years from the same "commercial interests"' shows a puzzling understanding of the industry considering that I have enjoyed internet, email and newsgroup access continuously since 1986 - 22 years by my calculations.
Your statement that only a 'handful of players ... pay telecom consultants' demonstrates a similarly superficial view of the communications (telecom and broadcast) business. I work for large and small ISPs, large and small ILECs, CLECs, content and application developers, fixed and mobile wireless companies, manufacturers, end users, landlords, government agencies. I am engaged by more than a 'handful of players' precisely because I do not simply parrot the party line. I am an advisor, drawing on 28 years of global experience in technology development and service operations, R&D, business planning, regulatory affairs and other areas you read on my intro package.
I write what I believe based on this body of experience. Period.
If you wish to comment further without resorting to ad hominem attacks, feel welcome.
Post a Comment
Your comment that 'it is impossible that you have been getting internet for 20 years from the same "commercial interests"' shows a puzzling understanding of the industry considering that I have enjoyed internet, email and newsgroup access continuously since 1986 - 22 years by my calculations.
Your statement that only a 'handful of players ... pay telecom consultants' demonstrates a similarly superficial view of the communications (telecom and broadcast) business. I work for large and small ISPs, large and small ILECs, CLECs, content and application developers, fixed and mobile wireless companies, manufacturers, end users, landlords, government agencies. I am engaged by more than a 'handful of players' precisely because I do not simply parrot the party line. I am an advisor, drawing on 28 years of global experience in technology development and service operations, R&D, business planning, regulatory affairs and other areas you read on my intro package.
I write what I believe based on this body of experience. Period.
If you wish to comment further without resorting to ad hominem attacks, feel welcome.
Posted by Mark Goldberg : 6:23 PM, August 21, 2008
Mark Goldberg : 6:23 PM, August 21, 2008 << Home
Syndicate
