Speaking Engagements & Private Workshops - Get Dean Bubley to present or chair your event

Need an experienced, provocative & influential telecoms keynote speaker, moderator/chair or workshop facilitator?
To see recent presentations, and discuss Dean Bubley's appearance at a specific event, click here

Thursday, December 05, 2013

Telcos' digital services/OTT businesses make Net Neutrality more important than ever

Telcos actually need Net Neutrality to survive & thrive, even though some don't realise it.

The current US & European softening of regulatory stance about "managed services" and Net Neutrality potentially hands suicide pills to operators, who mistake them for candy.

The problem is that some telco executives - and most of their lobbyists and industry associations - are still living in the past, and haven't actually caught up with the realities of creating and delivering new services, working with customers' preferred behaviour, and interacting with the innovators and developers of tomorrow's propositions. 

In addition, many are being influenced by vendors still pushing an assortment of unworkable plans around "personalisation", "smart pipes", "differentiated QoS", "1-800 models", "sender-pays content delivery" and assorted other policy paraphernalia and "useless-cases". I've been in meetings listening to telco network folk regurgitate tired marketing slogans about "turbo boosts" and "monetising OTT services" recently and had to do urgent remedial re-education on the realities of what's actually feasible.

Until now, Net Neutrality (in some markets) or at least the threat of regulatory intervention, has acted as a safety valve protecting telcos from themselves. 

The problem is that some of the regulators are also living in the past, and there is a danger that they'll buy into some of the nonsense. Part of the issue is that many of the advocates of Net Neutrality are also from another planet, arguing from silly positions like free speech and sinister conspiracies. 

There's a new commissioner at the FCC, while the EU position espoused by Neelie Kroes is seen by many to have caved-in to the requests of the telecoms incumbents wanting to provide managed services. The FCC chair appears to have read a 2008-era report on two-sided business models (yes, I know I wrote about it around then too - one of my worse calls) and seems to think they can be workably applied to broadband. The usual "Netflix will pay for a fast lane" wishful thinking is being wheeled out again. The Devil is in the details, but this article from TelecomTV suggests that the cut-and-dried Neutrality stance is unlikely to succeed. (And that's even before the baffling US legal system finishes its wrangling with the previous FCC rules).

The truth is that almost all non-Neutral Internet models are unworkable. Rather than abdicating their role as consumer advocates and saying "no rules should stand in the way of new business models", regulatory authorities should treat telcos pitching non-neutrality as if they were exhibiting suicidal tendencies, and take measures to prevent them from harming themselves their customers, and cherished national broadband/economic objectives.

Yes, I know there are some isolated non-neutral examples like zero-rating, throttling of P2P, or national blocks on specific services, but by and large these remain in the spirit of the general perceived openness of the Internet. There are also a handful of emerging-market prepaid mobile data plans that try to offer app-by-app/site-by-site pricing, which are often aimed at featurephone users with very low ARPUs and equally low expectations of what the Internet offers.

But the oft-repeated "Internet fast lane" falls down on many grounds, much of which will end up with telcos ending up worse off than before, rather than foiling the dastardly customer-driven succss of Internet application and service companies.

The most obvious gotcha here is that the majority of non-neutral models (and proposed rules) are completely in contradiction to the recent, welcome trend of telcos to develop their own OTT/Internet-style applications, services, enhancements and enablers. Numerous telcos now have OTT communications or content units, offering everything from video streaming to VoIP to developer platforms. Several now have Telco-OTT messaging apps, cloud propositions and so on, as I predicted in my 2012 report. Sometimes these are "standalone OTT", while others are extensions of existing on-net services for users wanting access via WiFi, or devices running on other providers' connections. Even standardised failures like RCS/joyn now have OTT-extension capabilities over generic Internet access.

Ending Net Neutrality would kill these and many other promising service initiatives stone dead. In discussions with these groups, I have come across nobody who has even considered asking other telcos or ISPs for help in prioritising traffic. Most haven't even looked into QoS/priority for the times when users are actually their own access subscribers, on their own networks. There are numerous reasons for this, but chief among them is that these separate units or product people are usually staffed by Internet people, for whom this entire thing is an alien prospect.

There's a broad set of other reasons to consider sweeping changes to Net Neutrality law being actively suicidal for telecom operators. Potential unintended consequences include:

  • Lots of expenditure on DPI/policy infrastructure & use-cases that don't actually work as well as vendors make out, given the constant shapeshifting of the application space
  • Loss of focus on things that customers actually want (cool new services, whether in-house or re-sold/partnered) vs. stuff they don't (attempts to sell them half the Internet rather than the whole thing)
  • Probability that regulators will insist that non-neutral offers are made on a non-discriminatory basis, ie to anyone who requests them. Should be fun when Skype, competing telcos or adult content firms get in touch.
  • Hideous nightmare of OSS/BSS integration to put in systems to monitor all of the new non-neutral stuff, bill for it, support it and so forth.
  • Numerous ways to game the system, either legitimately or fraudulently. I can think of plenty just off the top of my head (feel free to ask for a workshop for ideas!)
  • Probability that non-neutrality will cut both ways, with powerful companies like Netflix, Google and Facebook instead charging telcos for access to their servers/features rather than money flow going to the ISPs. Want your subscriber to get HD? How about avoiding the 10 second delay on search that gets implemented on your new shiny non-neutral network? Pay up. It'll soon be obvious who wields the real consumer power to force churn.
  • Likelihood of new OS's moving to default-on hard encryption for all IP traffic. Good luck picking out and prioritising traffic when that happens.
  • Web mashups making a mockery of what gets accelerated / prioritised
  • All sorts of entertaining corner-cases with radio networks, eg working out how to balance the needs of a Gold customer with a Platinum application at the edge of a cell's coverage, versus a 100 Bronze users using a Gold application next to the base station.
  • Proliferation of "how neutral is your ISP?" traffic-light reporting schemes. Some regulators already offer these.
  • Incentivising application developers and device/OS supplies to push users to open, 3rd-party WiFi wherever possible
  • Embarrassing and easy comparisons between "prioritised" traffic on-net, versus potentially better performance on other unmanaged networks. Will a refund mechanism be in place for unnecessarily paid QoS?
  • Huge scope for cost and lousy execution around the sales and marketing of non-Neutral Internet services.
  • The risk that lots of money gets spent, lots of time & management effort wasted and diverted from real value-added services, and then any (massively politicised) laws get changed 15 minutes after the next set of elections.
Yes, there are some isolated examples where this might work. Enterprise cloud services for home-workers on fixed broadband. In fact in general, it's easier to see non-Neutrality apply to the fixed/cable network where there's a much clearer distinction between Internet and non-Internet broadband services anyway - plus it's possible to use a second physical or logical connection to ringfence the two. But these are small niches, which don't warrant the huge moral hazards of risking the general "open Internet" principle which is the only thing driving 2+ billion people to buy data access. 

It would probably be worth governments sponsoring second broadband lines to support some of the other services, rather than caving in to the illogical and dangerous "multi-service" rhetoric from some pundits who spend too much time perfecting mathematics, rather than customer and economic needs.

Overall, the current swing back away from Net Neutrality looks as if years of lobbying has born fruit. Unfortunately, the lobbyists' and protagonists arguments are now years out of date with reality. Most non-neutral business models are not just irrelevant, they are actively harmful to the telecoms industry at precisely the point it needs to focus on service innovation, not pointless network tinkering. It will lead to greater loss of money for operators, and ultimately faster consolidation and poor outcomes for the consumer.

You get the picture. Basically, out in non-Neutral Internetland there are lots of bear-traps. And lots of bears. Be careful what you wish for. Regulators should be acting as rangers, rather than encouraging naive telcos to go blundering about in the "polyservice" forest.


Ben Klass said...

Excellent article, one of the most nuanced I've read on a blog.

You may be interested to learn that there is a formal complaint being considered by the CRTC up here in Canada, concerning Bell's prioritization of its own OTT mobile video service.

I think the complaint passes your criteria for "workability."

This will be the first time Canada's net neutrality regulation is applied to mobile OTT services, to my knowledge.



D.B. said...

I was one of those mindless marketing people at my last employer talking about "monetizing OTT services", but only because they paid me to do so. I didn't really believe what I was saying, and probably did a bad job. Hence I was selected to leave in a RIF, for the benefit of the rest. Perhaps I wasn't mindless enough.

InfoStack said...

Comparing apples (high, average priced, vertically integrated models) with oranges (low, marginally price, horizontally scaled).

Balanced settlements (not 2-sided revenue takings) actually making a ton of sense. We need price signals to clear supply/demand north-south between apps and infrastructure and east-west between networks and apps/services.

Dean Bubley said...

Benzo - Thanks, I will check it out.

D.B - (Good initials!) Thanks for the commet

InfoStack - those various fruit need to be compared, because the vertical models are being undercut by their own lack of investment in service innovation, and the stupid reliance on antiquated standard/federation models. You only have to look at the farce that is RCS to see how weak that approach is in the modern world of customer choice.

Not sure I entirely follow you on North/South and East/West. Any specific examples? All the ones I've heard (eg paid QoS, sender-pays data) are unrealistic on multiple grounds.

In particular, the very common use of 3rd party WiFi makes access/app price signalling essentially impossible much of the time. No app company wants to deal with 150 million independent access network owners, and the vision of QoS being enforced by either operators or WiFi consortia is unlikely/impossible.