Communications Insights: Trends and Cool Stuff - Latest Comments

Re: Water, Air, and Wetting One’s Pants

Tedsprague — Thu, 18 Nov 2010 11:52:25 -0000

As an aside on the Android issue, see Steve Wozniak's comment that Android will outpace the Apple iOS: http://news.cnet.com/8301-1350...

Re: The Uncommon Unlicensed – A Licensed Commons

Kenneth R. Carter — Wed, 01 Sep 2010 10:48:15 -0000

Bill, thanks for your insightful comments. Many incorrectly view the licensed/unlicensed as an either or dichotomy. This is a left versus right view of the world. More accurately, the FCC's unlicensed regime is viewed as an over versus under distinction. The jurisprudence of the Part 15 Rules has historically been that if the field strength of the emissions is sufficiently weak, they do not rise to the level of being "spectrum". So, as you point out the nuanced question is not whether we allow unlicensed operation, but rather how much unlicensed operation should we allow.

Re: The Uncommon Unlicensed – A Licensed Commons

Wlehr — Wed, 01 Sep 2010 07:48:32 -0000

Ken makes very important points. Nuanced spectrum reform needs to recognize that unlicensed is not a single regulatory framework and does not reflect a binary choice between property rights v. unlicensed. They are ALL property rights regimes, with different implications for different contexts. The are not cast in stone and can and should evolve over time.

Re: Taking the Roof off of the Internet

Kenneth R. Carter — Mon, 23 Aug 2010 03:17:50 -0000

I stand corrected.

Re: Taking the Roof off of the Internet

Rudolf — Sun, 22 Aug 2010 18:04:02 -0000

:quote: Bill-and-keep is the way most Internet traffic is exchanged (peering). It works well when the networks are roughly equivalent in size, traffic flows, and cost-causation. :/quote:

This was the bit I had a problem with. It's just not correct. networks don't need to be roughly equivalent in size or traffic flows and cost causation.

Also I fully agree that we need to stay away from terminating monopolies and horror of horrors cascading payments. These have stifled competition for ages now.

I do feel that if parties stick to peering and transit in the way it is done now, there will be less of a problem with terminating monopolies. A network can ask to be paid for a peering connection, but this can be refused too and it will be refused if the price is above the price of transit.

Re: Taking the Roof off of the Internet

Kenneth R. Carter — Sat, 21 Aug 2010 09:57:22 -0000

Christian, thanks for your comment. Prof. Medini Singh at Columbia B-School uses the Depuit quote in his Advanced Service Operations course. (Prof. Singh is the one of the best, certainly funniest, professors I have ever had). It also appears in a paper or two by Odlyzko, who is nothing short of a genius.

Thanks for cross-posting to your blog. What do they say, fools seldom differ...

Re: Taking the Roof off of the Internet

Christian — Fri, 20 Aug 2010 20:57:58 -0000

I love that Emile Dupuit quote. Francois Bar put it on a transparency when I took Media Economics from him.

Clearly we are thinking along the same lines about both network neutrality and railroad metaphors. My recent post:

http://futuretense.publicradio...

I swear I had to turn that post in earlier than it was put online. I wasn't copying you or the carriages!

Re: Taking the Roof off of the Internet

Kenneth R. Carter — Fri, 20 Aug 2010 15:02:06 -0000

Rudolf, thanks for your comments. They are, as always, well informed and very insightful. I had to noodle this through for a while before I could respond. What I am left with is even if peering is not the dominant arrangement how does that change my analysis?

I have no issue with transit providers. Transit tends to be a competitive market and even if both parties pay to the transit provider, there is no termination monopoly problem. If you do not like the cost of one transit provider, you can probably find someone else to give you a circuit on the same route. Similarly, I have no problem with “content” networks (we can stop beating up on Google) developing competitive advantages by building out distributed networks which move content closer to the eyeballs in the “eyeball” networks in order to improve the delivery performance over their competitors. I have said so in a previous Cool Stuff: http://kennethrcarter.com/Cool...

Further, I agree completely that the eyeball network (access network) and the content network will agree to peer if the cost of peering is less than the cost of transit for both of them, regardless of the traffic flows. It doesn’t take two MBAs to figure that one out.

Yet, I am still concerned about the termination monopoly problem. Let me be clear. I am not saying this is happening, nor that it will. However, the competitive threat remains. My concern is that the “you’re nuts if you think you’ll use my pipes for fee” extortion of payments could change the network economics. In the absence of peering, how is the content network paying the access network for managed (premium) service on its access network not like calling party pays, with the corresponding problems?

Finally, I can’t offer much advice on the alcohol. Maybe try “interconnect” the meetings with another. “Hi. I my name is Tier 1 Network, and I’m an…”

Re: Taking the Roof off of the Internet

Rudolf — Fri, 20 Aug 2010 04:11:14 -0000

sorry, kenneth, but Peering is not the default on the internet, transit is. Getting transit is easy. Getting peering is hard. Most people don't have enough traffic to sustain a peering agreement and are better off with transit. Getting peering also requires regular attendance at meetings where a lot of alcohol is present.

Furthermore peering has nothing to do with the direction nor with the balance in traffic. Yes I know there are idiots who believe in traffic ratios. Ask Google, their incoming is much lower than their outgoing and still everyone peers with them. Why is that. Well transit is paid for based on a given capacity measured these days in Gbit/s/month, regardless of direction. Both eyeball networks as well as content networks pay to the transit provider. Many content networks, like Google or Akamai may even pay less for transit than eyeball networks as they are bigger and can buy at cheaper locations, such as Amsterdam. So in order to get a peering the content network and eyeball network will look at the amount of money they can save from peering with the other network and cutting the transit network out of the loop. Many eyeball networks may reach this point earlier than some big content networks. So there you are, it doesn't matter which way the traffic flows.

Re: Taking the Roof off of the Internet

Kenneth R. Carter — Fri, 20 Aug 2010 04:00:16 -0000

Kind words on this post by my good friend Stefano Quintarelli on his Quinta's Weblog: http://blog.quintarelli.it/blo.... Please read, or at least Google Translate.

Re: Solving Network Neutrality

georgeou — Wed, 18 Aug 2010 07:56:41 -0000

I should also point out that the actual regulatory fight in the US is not over whether ISPs can block or degrade (in undue ways other than fair responses to normal Internet congestion) certain websites. We know that it would be suicidal of an ISP to do this on a PR level and political level.

The real impetus comes from companies that invested in CDN prioritization technology (e.g., Google) who want to kill off competing router prioritization technologies through regulatory or legislative means. That's why they specifically want to target ISPs offering these services to Business to Consumer (B2C) websites.

Re: Solving Network Neutrality

georgeou — Wed, 18 Aug 2010 07:52:27 -0000

I've explained my comments on unbundling (mandatory wholesale at government regulated prices) in detail in the links I provide. I would not be able to paste all of it here.

When you change the rules of the game and require wholesale of fiber at government set prices after a provider builds the network under a regime that didn't require unbundling, that is essentially taking private property and giving it to someone else at a non market rate.

Also, the current US FCC and congressional debate isn't actually around unbundling or the reversion to a Title II regime. Some people would like to see that happen but it's extremely unlikely the courts would allow this.

The regulatory issue boils down to an arbitrary and capricious targeting of ISPs but not other premium content delivery providers that use CDN technology. My first link makes all of this very clear (or so I tried).

Re: Solving Network Neutrality

Kenneth R. Carter — Wed, 18 Aug 2010 07:27:59 -0000

George, thanks for your comments. While I agree with much of what you say, let me take issue with a few things in order to bring the debate into sharper focus.

I would not go so far as to say that Network Neutrality is devoid of logic. There have been some serious deviations from neutrality in previous years. To my mind the most serious has been the AfterDowingStreet.org kerfuffle. That said I am not prepared to substitute my judgment in lieu of the market as to what is the appropriate level of prioritization and product tiering. That said, the market must be sufficiently competitive.

I am not sure exactly what you mean by the unbundling vision of Network Neutrality. I assume you mean network unbundling to achieve competitive entry. Unbundling, in and of itself, is not a taking. Even if it were, it would be permissible under the power of Eminent Domain. Nonetheless, you may not use your property in an anti-competitive manner. That is not a property right you have. The government can set appropriate constraints on the market place. Indeed, most telecommunications networks are built using public rights of way. Are you suggesting that this public domain has been transferred to private ownership and cannot be undone? Unbundling is an effective tool used (with different levels of success) by regulators in nearly all industrialized countries.

Re: Solving Network Neutrality

georgeou — Wed, 18 Aug 2010 06:45:36 -0000

The Net Neutrality movement as a whole is devoid of logic and goes against the wishes of the end user. http://bit.ly/aefl4O

Unbundling isn't all that it's cracked up to be because it's very difficult to achieve scale as a small ISP which means you're going to underprovision for backhaul bandwidth while paying a lot due to lack of quantity. This problem is especially true in the UK and a lot of those DSL-based ISPs can't

The unbundling vision of Net Neutrality is one of the more extreme views and it is often seen as an alternative to the other extreme forms of Net Neutrality (http://bit.ly/9JG8LA). Unbundling is essentially the government grabbing private property with full blown Title II but it's clear that the FCC doesn't even have the authority to mandate Title II lite which leaves out the unbundling provisions. The whole unbundling movement is incoherent http://bit.ly/bVy5bI.

Re: My iPhone vs. My iGo

Kenneth R. Carter — Wed, 28 Jul 2010 14:31:55 -0000

UPDATE: I upgraded to OS 4.0. I am pleased to report that I was able to connect my iGo keyboard. In fact, I am typing on it now. Once I turned on the Bluetooth, it took about 90 seconds to pair the keyboard and iPhone. Why this innovative feature had to wait until the fourth generation of the software is beyond me. My Apple boycott is over.

Re: My iPhone vs. My iGo

Kenneth R. Carter — Fri, 07 May 2010 14:10:34 -0000

UPDATE: After two weeks with my new iPhone, I have found a few other things you cannot so with it. For one, it will not sync with MS Outlook 2002. The first time you sync it downloads contacts and appointments, but will not update them. The iTunes does not give an error message that it has not updated the iPhone, so the only way you know is when you start missing appointments. I guess that means I do not need to bring cable to the office. The iPhone will not let you sync book marks from Firefox. You cannot offload photos without additional software or sending them by MMS. You cannot load PDFs to the iPhone read except by email. Finally, something really cool I used to do with my old Nokia which I cannot do with the iPhone is to tether the onboard camera with a laptop and use it as a webcam.

I conclude that the iPhone is a terrific media player, a decent mobile phone/Internet appliance, however, as a computing device or computer peripheral, it is a total wrote off. In fact, I took my iPhone back and tried to exchange it today. “Kein Chance.” Hopefully, the Bluetooth will work in OS 4.0.

Re: One Word Oxymorons

Kenneth R. Carter — Fri, 07 May 2010 13:50:11 -0000

I have thought of two more: AstroTurf (star-ground) and spendthrift (spending and frugal). They are both, however, compound words.

Re: The Spoon

Mark Essel — Thu, 06 May 2010 13:56:31 -0000

Howdy Kenneth,
Enjoyed a few of comments on AVC, so I dropped in and have to say I'm a fan of your blog mojo. I use mine as a way to offload thoughts, some of which are far out. Dig your analogies, keep it coming :D

Mark

Re: My iPhone vs. My iGo

Kenneth R. Carter — Thu, 29 Apr 2010 03:18:38 -0000

Steve,

Thanks for your comment and helpful information. I don't mind so much that my iPhone did not come with a firewire (at least it did not come with a serial cable), but the fact that I cannot use my old ones annoys me.

I am glad to hear that OS 4.0 will resolve this problem. I will hold my breath and will hopefully be able to end my boycott.

Re: My iPhone vs. My iGo

Steve Stroh — Wed, 28 Apr 2010 20:14:38 -0000

Re: the change from Firewire on iPods to USB on iPhones. Understand your frustration, but I think that was Apple just giving in to the reality that Firewire has become a legacy port and going forward USB was where the momentum is... like when they gave up on LocalTalk in favor of Ethernet.

Agreed WHOLEHEARTEDLY on the idiocy of not allowing the Apple Bluetooth keyboard to work with the iPhone. Fortunately, you won't have to wait long. iPhone OS 4.0 will add this, and it's due later this year. I've been waiting two years for this feature on my iPhone 3G.

Re: Every Ash Cloud Has a Silver Lining (for some)

Flo — Tue, 20 Apr 2010 05:57:26 -0000

Hi Ken, this could be an interesting article for you too:
From a cloud to the cloud: How ash kills airmail (by Jeff Jarvis)

http://www.buzzmachine.com/201...

The way of communication will change faster.

Re: The Uncommon Unlicensed – A Licensed Commons

Andrew Harms — Mon, 19 Apr 2010 14:58:55 -0000

Thank you Ken, this is very helpful. Harmful interference is now clear from Margie's article.

Maybe a technical point, but if current unlicensed Wi-Fi hotspots are limited to smaller areas, how are they able to receive signals/information from beyond their range, and how are the outside signals transfered into the wifi hotspot? How might that be limiting?

And if white spaces in urban areas were to be unlicensed, what sort of commercial models would develop? Could this mean that a larger section of a city could be covered with wifi, perhaps offered through a company charging a fee, or maybe being used as public access? (Free wifi throughout central park...) - Is it possible that if White Spaces are unlicensed for a time, and certain technological innovations are developed and established, that there might be a later need for licensing in White Spaces?

Finally, would there be different policy created for use of white spaces in rural areas compared to urban areas?

Re: Keeping up with the Jitsuzumis

Kenneth R. Carter — Wed, 14 Apr 2010 16:15:42 -0000

TJ, thanks for your comment. 25 Mbps is still about 5 times what the average American gets for his broadband Internet connection and more than 100 times the FCC's legal definition of broadband.

Re: Keeping up with the Jitsuzumis

TJ — Sat, 10 Apr 2010 22:54:37 -0000

FCC's ambitious plan is to achieve "actual" 100Mbps, which is much higher than the current Japanese status. According to my recent survey, Japanese broadband users can enjoy around 25% of their "up to" speed on average.

Re: The Uncommon Unlicensed – A Licensed Commons

Kenneth R. Carter — Wed, 31 Mar 2010 16:43:29 -0000

Andrew, thanks for your questions. To begin with, there is no really clear definition of what exactly constitutes harmful interference. (http://cli.gs/ST34qd) However, the FCC’s definition is not generally with in the Part 15 rules. That said, it is my considered opinion that the Part 15 rules are in fact overly restrictive. FCC could certainly make the rules less restrictive before harmful interference became a major problem. When was at the FCC, I developed some data for my 2004 TPRC paper which supports that conclusion: In the first 4 years of the Enforcement Bureau, from 2000 to 2003, 9.7% of the EB’s case load related to interference problems. Of those nine percentage points, only 1.7% related to Part 15. In other words, a mere 65 cases out of the full 36,000 case load (0.16% of the total) in four years were related to unlicensed. While these numbers are a bit out of date, I cannot imagine that the percentages have changed very much. To your second question, it is not so important where the White Spaces rules are codified. I believe that they will be codified at 47 U.S.C. §15. However, what is important is that they offer more than the existing regime. As I allude to in the blog entry, the TV bands are better frequency ranges in terms of congestion and propagation characteristics than the 2.4 or 5.8 GHz ranges in which Wi-Fi currently operates. Unlicensed operation is, however, permitted in the 902 to 928 MHz band, just above the TV beach-front spectrum. I hope this answers your questions.