Dear sirs
I am grateful for the opportunity to respond to Ofcom's informal consultation on traffic management and note our previous conversations on this matter. This response notes your outstanding concerns that a regulatory agency, even one as august as Ofcom, is limited in its field of inquiry and is not suited to conducting an inquiry into the wider implications of network neutrality.
Indeed, given the broad policy questions about innovation policy that this inquiry suggests, it is frankly close to pointless (though elegantly and competently done) for Ofcom to conduct its walk-through of discrimination and two-sided market analysis.There is real value in your proposals for consumer transparency in Chapter 5, and it is here that I focus.
I attempt to answer the eleven questions posed, but also attach a working article on the wider issues that will be considered more fully by the European Commission and hopefully by your parent departments. As you are aware, I authored a monograph on these issues in Janaury 2010 which has now been downloaded more than 100,000 times and has been submitted as evidence to the FCC NPRM in March. It is also in wide circulation in BEREC, the CRTC, PTS Norway and the European Commission.
i) How enduring do you think congestion problems are likely to be on different networks and for different players?
On mobile, there is likely to be a more or less permanent attempt by the oligopolistic networks to block rival applications, especially streamed video, voice and peer-to-peer file-sharing.
On fixed networks, the cable network should be able to provide backhaul at reasonable levels with adequate backhaul investments. Its current policies intend to throttle for 10 hours in the 10am-9pm period which is not promising.
British Telecom should also be able to supply its own backhaul and that to its competitors at regulated levels, but this depends on the regulator showing the ability to recognise that exponential increases in backhaul capacity need to be matched by exponential decreases in wholesale costs to its competitors and retail arm. Neither I, nor other independent analysts (notably Burstein and Odlyzko) are hopeful of BT and other incumbents' incentives to achieve this, nor regulators' competence and vigour in pressing for such a solution.
ii) What do you think are possible incentives for potentially unfair discrimination?
See our 200-page report for BIS, DCMS and Ofcom in 2006: http://www.ofcom.org.uk/research/tv/reports/videoregulation/
We discuss at great length the potential investment-chilling effects of such regulation in the work Ofcom commissioned from RAND on the Audio Visual Media Services Directive in 2006.
I hold to those conclusions.
iii) Can you provide any evidence of economic and or consumer value generated by traffic management?
Note that it is a useful threat to block business cases for independent video or voice over IP offers, as we discuss at great length in the work Ofcom, BIS and DCMS commissioned on the Audio Visual Media Services Directive in 2006 - in this case, Ofcom substitutes the chilling effects of European regulation with the equally chilling effects of private ISP regulation. We find it bizarre that Ofcom cannot see the parallel.
iv) Conversely, do you think that unconstrained traffic management has the potential for (or is already causing) consumer/citizen harm? Please include any relevant evidence.
See our 200-page report for BIS, DCMS and Ofcom in 2006, detailing the effects of regulation on innovation (especially Appendix 1 on investment-shifting games): http://www.ofcom.org.uk/research/tv/reports/videoregulation/
v) Can you provide any evidence that allowing traffic management has a negative impact on innovation?
Can anyone prove a negative?
The questions you should ask are:
Can a developer rely on an open Internet in future?
Has an open Internet led to innovation in the past?
Did guaranteed Quality of Service on the telecoms and ATM networks lead to less or more innovation than that in the Internet in the thirty years since 1980?
vi) Ofcom’s preliminary view is that there is currently insufficient evidence to justify ex ante regulation to prohibit certain forms of traffic management. Are you aware of evidence that supports or contradicts this view?
Ofcom itself is well aware that ISPs - especially non-dominant ISPs - have been engaged in traffic discrimination for at least five years that has led to huge user frustration and complaints to ISPs - who relayed these complaints to Ofcom in 2006. See http://chrismarsden.blogspot.com/2008/10/charlie-dunstones-gotcha-moment-2006.html
Given that all ISPs have such incentives due to the incentives between layers in the Internet value mesh, the claim that there is insufficient evidence is a condemnation of Ofcom's evidence gathering and ISPs' failure to report complaints, not of ISP users.
vii) Ofcom’s preliminary view is that more should be done to increase consumer transparency around traffic management. Do you think doing so would sufficiently address any potential concerns and why?
It may, but as your Chapter 4 analysis fails to comprehend the underlying reasons for discrimination (incentives between layers, rather than vertical integration by incumbents with market power), there appears inadequate understanding of the competitive dynamics between layers.
viii) Are you aware of any evidence that sheds light on peoples’ ability to understand and act upon information they are given regarding traffic management?
As you are aware from my work with Jonathan Cave and others on video, interactive gaming and mobile video http://www.ofcom.org.uk/research/tv/reports/videoregulation/, there is a pressing need to examine these individual sectors and their prosumers' response to traffic management. It is now urgent that Ofcom - or more properly BIS - carries out this work, especially in the field of interactive gaming, in which the UK formerly led the world in innovation and development, based on an educated user base. The reasons for our relative decline in that field deserve immediate inquiry.
ix) How can information on traffic management be presented so that it is accessible and meaningful to consumers, both in understanding any restrictions on their existing offering, and in choosing between rival offerings? Can you give examples of useful approaches to informing consumers about complex issues, including from other sectors?
See my proposed co-regulatory solution outlined in the book: Net Neutrality: Towards a Co-Regulatory Solution (2010) Bloomsbury Academic, London.
x) How can compliance with transparency obligations best be verified?
The only verifiable solutions, as Ofcom itself has made clear, is an independent website to offer QoS information on a quarterly if not monthly basis, accompanied by Ofcom and SamKnows (or another partner's) mystery shopper tests. Currently, many ISPs are engaged more in garrotting than throttling of bandwidth for their paying customers. The size of print, in marketing materials for new customers, that 'explains' this is miniscule and brief to the point of invisibility.
xi) Under what circumstances do you think the imposition of a minimum quality of service would be appropriate and why?
It is essential for both prosumers and developers that a minimum level of Internet access be provided. ISPs should not be allowed to advertise themselves as providers of public broadband communications if they intend to remove the majority of that bandwidth for large periods of the day from their customers. They appear to have forgotten that they are common carriers, and Ofcom needs to reassert its regulatory authority and remind them. The Code of Practice from 2008 - to be revised during this consultation period in response to obvious ISP failure to comply as revealed by Ofcom's mystery shopping as well as thousands of complaints made by customers through ISP fora online - is hopefully the beginning of that process.
I am always available to amplify these points. I congratulate Ofcom for provoking this debate and hope the new government will show some kind of leadership on the wider issues, not least freedom of expression and privacy, so poorly neglected in the previous silence on this issue, not least in the Digital Britain report and the mute BIS response on net neutrality to the latest EC Implementation report questionnaire.
Sincerely
Chris Marsden
Friday, June 25, 2010
Ofcom blows the whistle on itself on self-regulation
You may recall I discussed in 2008 just how 'voluntary' the ISP Code of Conduct on broadband really was? Well, Ofcom appears to have inadvertently told the truth in its net neutrality discussion:
"Ofcom has already introduced information requirements through our Code of Practice on Broadband Speeds...Within Ofcom’s Code of Practice on broadband speeds, there is also a requirement to provide information on policies on fair usage, traffic management and traffic shaping...We are planning to publish the new code after the summer." [pp37-38, Chapter 5]
So it is Ofcom's Code, not the ISPs', which was always obvious to insiders, but its good to have it confirmed. the ISPs intended to do precisely nothing without the regulators' cattle prod. Elsewhere, Ofcom suggests several alternatives for actually getting ISPs to spill the beans on QoS:
"5.48 We recognise that the use of traffic management techniques is, by its nature, difficult for consumers to detect and understand. It is therefore, critical that consumers are fully informed of any traffic prioritisation, degradation or blocking policies being applied by their network operator or ISP and that they are able to factor these in when making purchasing or switching decisions. This is important not just from the point of view of consumers’ rights, but also relevant to the question of whether stricter rules may be required on discrimination."
For what its worth, I suggest they should allow consumers to measure their speed of access (I can do this currently with my 3 mobile connection), and think they MUST offer their own website detailing traffic management based on quarterly ISP metrics.
"Ofcom has already introduced information requirements through our Code of Practice on Broadband Speeds...Within Ofcom’s Code of Practice on broadband speeds, there is also a requirement to provide information on policies on fair usage, traffic management and traffic shaping...We are planning to publish the new code after the summer." [pp37-38, Chapter 5]
So it is Ofcom's Code, not the ISPs', which was always obvious to insiders, but its good to have it confirmed. the ISPs intended to do precisely nothing without the regulators' cattle prod. Elsewhere, Ofcom suggests several alternatives for actually getting ISPs to spill the beans on QoS:
"5.48 We recognise that the use of traffic management techniques is, by its nature, difficult for consumers to detect and understand. It is therefore, critical that consumers are fully informed of any traffic prioritisation, degradation or blocking policies being applied by their network operator or ISP and that they are able to factor these in when making purchasing or switching decisions. This is important not just from the point of view of consumers’ rights, but also relevant to the question of whether stricter rules may be required on discrimination."
For what its worth, I suggest they should allow consumers to measure their speed of access (I can do this currently with my 3 mobile connection), and think they MUST offer their own website detailing traffic management based on quarterly ISP metrics.
Ofcom: show us the money on net neutrality
As Ofcom rightly points out, based on conventional economic analysis, there is no evidence that fundamentally changing the nature of the Internet will slow down the pace of innovation. That said, they ignored the 2006 evidence in front of all their top brass (from 12:26) that net neutrality was a serious problem, so don't hold your breath, you can't prove a negative:
"4.50 It is possible to postulate a future scenario in which the introduction of traffic management and the kind of charging models described in this chapter lead to a different kind of internet economy, in which the space available for the ‘best efforts’ internet – and the low barriers to market entry and innovation that it guarantees – is reduced. If this outcome starts to emerge, it is likely that this would lead to pressure to regulate to avert this. One possibility would be to use the powers to impose minimum quality of service to define a ‘best efforts’ internet to which all network operators and ISPs would have to designate a certain proportion of network capacity. However, we re-emphasise that there is as yet no evidence that this problem is arising, and good reasons to doubt that it will arise, given the current competitive market structure and the incentives on network operators and ISPs existing within that market structure."
However, while we slaughter, or at least harass, the golden goose (whose eggs are in any case mainly harvested in the US, while Europeans have enjoyed consumer welfare increases based on telco value destruction by BitTorrent and Skype), is there not at least a bit of room to consider that there have been vast - simply vast - 'prosumer' (EU jargon) benefits from the freedom to tinker that a neutral Internet gave us? Yes, but that's not strictly within the bounds of the economic analysis performed according to Ofcom. This is a serious and seriously bounded economic analysis of net neutrality. Stupid monkeys abound!
That's not Ofcom's fault necessarily, its just a statement of why the government should be having direct citizen consultation on net neutrality rather than delegating it to a regulatory agency that necessarily has to limit its investigation. We were promised by the Tory opposition that there would be more policy making by government rather than Ofcom after they were elected? Not that such a prospect fills me with joy as the government is in denial, but it would at least consider options in view of the announced review of the E-Commerce Directive, for instance.
"4.50 It is possible to postulate a future scenario in which the introduction of traffic management and the kind of charging models described in this chapter lead to a different kind of internet economy, in which the space available for the ‘best efforts’ internet – and the low barriers to market entry and innovation that it guarantees – is reduced. If this outcome starts to emerge, it is likely that this would lead to pressure to regulate to avert this. One possibility would be to use the powers to impose minimum quality of service to define a ‘best efforts’ internet to which all network operators and ISPs would have to designate a certain proportion of network capacity. However, we re-emphasise that there is as yet no evidence that this problem is arising, and good reasons to doubt that it will arise, given the current competitive market structure and the incentives on network operators and ISPs existing within that market structure."
However, while we slaughter, or at least harass, the golden goose (whose eggs are in any case mainly harvested in the US, while Europeans have enjoyed consumer welfare increases based on telco value destruction by BitTorrent and Skype), is there not at least a bit of room to consider that there have been vast - simply vast - 'prosumer' (EU jargon) benefits from the freedom to tinker that a neutral Internet gave us? Yes, but that's not strictly within the bounds of the economic analysis performed according to Ofcom. This is a serious and seriously bounded economic analysis of net neutrality. Stupid monkeys abound!
That's not Ofcom's fault necessarily, its just a statement of why the government should be having direct citizen consultation on net neutrality rather than delegating it to a regulatory agency that necessarily has to limit its investigation. We were promised by the Tory opposition that there would be more policy making by government rather than Ofcom after they were elected? Not that such a prospect fills me with joy as the government is in denial, but it would at least consider options in view of the announced review of the E-Commerce Directive, for instance.
Thursday, June 24, 2010
Ofcom consultation: nothing to see here
Ofcom has issued its 'do-nothing' consultation on net neutrality - they're not big fans, though a bit more consumer transparency is accepted and there's some analysis of discrimination and market power, which isn't terribly relevant to the n-sided market concern that there is simply no incentive for ISPs to support P2P applications. Rather cunningly, this consultation (deadline 9 September) was launched just as 800,000 concurrent iPlayer streams had all but overwhelmed ISP networks yesterday as In-gur-land proceeded erratically to their date with ze Germans.
As you would expect of Ofcom, its well written and argued, and beautifully produced - check out the 'garrotting' continuum on page6! Below that lovely diagram is the ugly assertion: "it is widely accepted that the blocking of illegal content (such as images of child abuse) is necessary and that steps taken to address issues such as online copyright infringement would be viewed as acceptable traffic management." Really? Aren't both forms of censorship that do nothing to solve the problem - i.e. removing the actual kiddie porn from source, and reforming copyright business models while actively prosecuting commercial filesharing frauds? Ah well....
More later. It really is a good read. Unfortunately it won't deal with wider issues: "questions of fundamental rights and industrial and public service policies are beyond the scope of this discussion document as they are matters for Government." [pp7-8]. It begs the question as to why BIS has not yet consulted directly on this issue.
As you would expect of Ofcom, its well written and argued, and beautifully produced - check out the 'garrotting' continuum on page6! Below that lovely diagram is the ugly assertion: "it is widely accepted that the blocking of illegal content (such as images of child abuse) is necessary and that steps taken to address issues such as online copyright infringement would be viewed as acceptable traffic management." Really? Aren't both forms of censorship that do nothing to solve the problem - i.e. removing the actual kiddie porn from source, and reforming copyright business models while actively prosecuting commercial filesharing frauds? Ah well....
More later. It really is a good read. Unfortunately it won't deal with wider issues: "questions of fundamental rights and industrial and public service policies are beyond the scope of this discussion document as they are matters for Government." [pp7-8]. It begs the question as to why BIS has not yet consulted directly on this issue.
World Cup and the Internet
A peak of 14million UK viewers (according to BARB, they can't count mass audiences in pubs, the 100,000 watching at Glastonbury's festival Pyramid stage etc. so add a million) watched England hang on against the mighty Slovenes yesterday afternoon - presumably the highest workday audience of all time? It was lower than the evening 17.7million versus nailed-on-guaranteed semi-finalists USA or the 19million versus pretty-but-scoreless Algeria.
More pertinent, the BBC was serving a record 800,000 concurrent UK streams on iPlayer - I 'enjoyed' it on SuperJANET and it was pretty superb quality with almost no jitter on that Gigabit network. That more than doubled the BBC's previous record and must have put a hell of a strain on the content delivery network and not a few office networks - but of course the wholesale ISP networks suffered: "Demon says there was a 55% increase in internet traffic during the match, while Easynet put the spike in traffic at kick-off at 226%."
Regular readers will note that Paraguay and its mobile surfers are playing the surprise package of New Zealand's 'All-Whites' this afternoon. This blog as Paraguay supporters predicts a dull 0-0 or a 1-0 Paraguay win and they will sail into the last-16.
More pertinent, the BBC was serving a record 800,000 concurrent UK streams on iPlayer - I 'enjoyed' it on SuperJANET and it was pretty superb quality with almost no jitter on that Gigabit network. That more than doubled the BBC's previous record and must have put a hell of a strain on the content delivery network and not a few office networks - but of course the wholesale ISP networks suffered: "Demon says there was a 55% increase in internet traffic during the match, while Easynet put the spike in traffic at kick-off at 226%."
Regular readers will note that Paraguay and its mobile surfers are playing the surprise package of New Zealand's 'All-Whites' this afternoon. This blog as Paraguay supporters predicts a dull 0-0 or a 1-0 Paraguay win and they will sail into the last-16.
Wednesday, June 23, 2010
Private lobbying in FCC on net neutrality - shocking, shocking...
I was shocked, shocked to hear that the FCC was triangulating between Google and the Bells....
No, not really, I didn't think so either. Will Washington ever work out what it finds politically expedient to state about net neutrality? Its not often that Brussels feels smug about its fast-moving dynamic pro-innovation lead over DC...
No, not really, I didn't think so either. Will Washington ever work out what it finds politically expedient to state about net neutrality? Its not often that Brussels feels smug about its fast-moving dynamic pro-innovation lead over DC...
Sunday, June 20, 2010
John Naughton on the Internet's effect on us
The ever-thoughtful and engaging John Naughton has an excellent article in The Guardian about 9 lenses through which to view the Internet's effect. Alongside John Seely-Brown, I think he's an essential read on where we might (and might not) be going - sagacious.
Here's a small sample:
"The first printed bibles emerged in 1455 from the press created by Johannes Gutenberg in the German city of Mainz. Now, imagine that the year is 1472 — ie 17 years after 1455. Imagine, further, that you're the medieval equivalent of a Mori pollster, standing on the bridge in Mainz with a clipboard in your hand and asking pedestrians a few questions. Here's question four: On a scale of one to five, where one indicates "Not at all likely" and five indicates "Very likely", how likely do you think it is that Herr Gutenberg's invention will: (a) Undermine the authority of the Catholic church? (b) Power the Reformation? (c) Enable the rise of modern science? (d) Create entirely new social classes and professions? (e) Change our conceptions of "childhood" as a protected early period in a person's life? On a scale of one to five! You have only to ask the questions to realise the fatuity of the idea. Printing did indeed have all of these effects, but there was no way that anyone in 1472, in Mainz (or anywhere else for that matter) could have known how profound its impact would be."
ERRATUM: John writes for The Observer but as I consume almost all my news (Private Eye the exception) online, and The Guardian website consumes its uneasy 'partner', I "read it on" TheGuardian.
Here's a small sample:
"The first printed bibles emerged in 1455 from the press created by Johannes Gutenberg in the German city of Mainz. Now, imagine that the year is 1472 — ie 17 years after 1455. Imagine, further, that you're the medieval equivalent of a Mori pollster, standing on the bridge in Mainz with a clipboard in your hand and asking pedestrians a few questions. Here's question four: On a scale of one to five, where one indicates "Not at all likely" and five indicates "Very likely", how likely do you think it is that Herr Gutenberg's invention will: (a) Undermine the authority of the Catholic church? (b) Power the Reformation? (c) Enable the rise of modern science? (d) Create entirely new social classes and professions? (e) Change our conceptions of "childhood" as a protected early period in a person's life? On a scale of one to five! You have only to ask the questions to realise the fatuity of the idea. Printing did indeed have all of these effects, but there was no way that anyone in 1472, in Mainz (or anywhere else for that matter) could have known how profound its impact would be."
ERRATUM: John writes for The Observer but as I consume almost all my news (Private Eye the exception) online, and The Guardian website consumes its uneasy 'partner', I "read it on" TheGuardian.
Subscribe to:
Posts (Atom)