Home / News

Another Look at the Broadband Gap in US

Saul Hansell of the New York Times is writing a series of posts examining the state of broadband in US and what the country can learn from broadband deployment in other nations. Hansell writes: "In Japan, broadband service running at 150 megabits per second (Mbps) costs $60 a month. The fastest service available now in the United States is 50 Mbps at a price of $90 to $150 a month. In London, $9 a month buys 8 Mbps service. In New York, broadband starts at $20 per month, for 1 Mbps. In Iceland, 83 percent of the households are connected to broadband. In the United States, the adoption rate is 59 percent."

Read full story: New York Times

By CircleID Reporter – CircleID's internal staff reporting on news tips and developing stories. Do you have information the professional Internet community should be aware of? Contact us. Visit Page

CircleID Newsletter The Weekly Wrap

More and more professionals are choosing to publish critical posts on CircleID from all corners of the Internet industry. If you find it hard to keep up daily, consider subscribing to our weekly digest. We will provide you a convenient summary report once a week sent directly to your inbox. It's a quick and easy read.

I make a point of reading CircleID. There is no getting around the utility of knowing what thoughtful people are thinking and saying about our industry.

VINTON CERF
Co-designer of the TCP/IP Protocols & the Architecture of the Internet

Comments

More analysis yields more questions By Dan Campbell  –  Mar 11, 2009 5:32 pm PST

This is an interesting article and analysis.  What is more interesting is many of the comments that provide either excellent rebuttals or first hand evidence that things are not exactly as they seem.  Read some of the comments - there is some pretty good material there.  This is one of many broadband analysis articles that puts the U.S. somewhere from 1st to last in broadband penetration or uptake, and all it seems to prove is that you can't prove anything.  It is too complex to evalute fully and mostly boils down to interesting conjecture and arguments.  My questions / comments are…

1. When comparing broadband pricing across countries, how is the comparison done?  Is it simply just translating the price in one country via the currency exchange as it was around the time the article was written?  If so, that's hardly telling.  A $US may not be the same as a Yen translated to the dollar.  How do you factor in the average salaries and income in a country, which is necessary to understand if something is more or less expensive elsewhere.  The exchange rate in the UK to the $US a year about was an attrocious 2.1 or so, making travel to the UK from the US a very expensive endeavor.  Late last year it came down to a very reasonable 1.35 or so, and has since gone back up a bit.  In the meantime, salaries in the UK and the US didn't change much, nor did broadband bills.  So how do these analyses account for this?

2. As pointed out in the comments, how are they measuring consistency in performance testing?  If they are using a tool on the Internet, where is it located?  Indeed, latency impacts TCP (and thus HTTP) as much as congestion and almost as much as line speed (when line speed is low enough to be relevant).  And indeed many web sites and services are still hosted in the US.  So testing from the Far East or Australia may add in an extra 150-200ms RTT.  And indeed, as someone pointed out, when you get to a certain line speed and upstream/downstream speed, it doesn't matter much anymore for what most people do on the Internet and only matters if you are watching HD video or downloading alot of HD files.  Once you get to a certain bandwidth, latency, web server performance and traffic and a myriad of other factors come into play.  Your throughput on a single transaction may be well below what your broadband link is capable of because of the other factors.

3. How do you account for inconsistencies and exaggerations in what is advertised versus what is actually true in terms of bandwidth?  And how do you reconcile the differences in packages when some are capped by MB or GB per month or metered according to sustained Mbps over time, whereas some are unlimited (like most in the U.S.) It is not very easy to come up with a model that fairly compares pricing.  It may not be as bad as apples to oranges comparisons but it is defintely different types of apples.

4. Some comments pointed out that while some Far East countries boast of better fiber to the "home" than the US, the reality is that much of the FTTH in the Far East is fiber to a multi-dwelling unit where the last "mile" (or few hundred feet from the pole/conduit/basement to each condo/apartment) is DSL or Ethernet or whatever.  Not to split hairs, because that is still a very valid and fast technique to bring high speed broadband to each unit, but in such cases did they count each apartment unit as having "fiber to the home" or did they just count the single multi-dwelling unit once?  I'm actually not sure what is right or fair, but if they count each unit as getting fiber when it really has DSL or Ethernet, than comparing true FTTH numbers is at least muddied if not just out and out wrong.

5.  As someone commented, how can you make comparisons from country to country when the demographics and geographies are so different?  How can you compare the broad and geographically diverse 300M population US to Iceland that has about 300,000 residents of which half live in Reykjavik?

6.  How do these analyses factor in the relative interest in broadband so as to not "penalize" the US or anyone else simply because some residents choose not to have broadband for whatever reasons?  Do these analyses typically use homes passed or that which could have broadband, or do they measure the actual number of subcribers?  Probably a bit of both.  But in the latter - the number of subscribers - that doesn't tell you what is possible.  My 77 year old mom and 90 year old aunt live near me and have as many (at least 3) options for broadband service, all at reasonable prices, but they choose not to have it because they just don't need or want it.  The law of large numbers should take care of this, but it still goes to show you that any analysis can be spun in any direction.  There are enough ways to evaluate and measure broadband service to support or negate any argument.

Again, I think this article is pretty good analyses.  But this may be about the 10th article I've read recently on this topic, and the results always seem to vary, which just seems to show that whatever your theory is it will be tough to prove.

Add Your Comments

 To post your comments, please login or create an account.

Related

Topics

IP Addressing

Sponsored byIPv4.Global

Cybercrime

Sponsored byThreat Intelligence Platform

DNS Security

Sponsored byAfilias

Cybersecurity

Sponsored byVerisign

Whois

Sponsored byWhoisXML API

New TLDs

Sponsored byAfilias

Domain Names

Sponsored byVerisign

Brand Protection

Sponsored byAppdetex