[SATLUG] OT-Time Warner Download Cap
bruce.dubbs at gmail.com
Fri Jun 6 11:51:22 CDT 2008
Mark Spieth wrote:
> "Bruce, you are "Right On!". This is a "Red Herring" like oil
> demandexceeds supply or Iraqi WMD's."
> I would be happy to debate the point of this being, as you put it, a
> "Red Herring" as the subject of this thread is partially related to part
> of my Doctoral Thesis in Cultural Anthropology in regards to how
> Internet usage effects and changes the cultures where internet is widely
> available and the ramifications of such a change in society. So, since
> you feel this is a "Red Herring" would you care to elaborate on your
> Since they pay way more per Mb for
> bandwidth than they could ever sell it for, they hope that the average
> user will not use most of their available bandwidth.
I think you need to document this assertion. The cost for a network consists of
connections (generally copper or fiber, but could include cell or wifi) and the
termination equipment. Operating costs include electricity, supoprt,
These costs are generally shared with phone and TV service so the costs are
leveraged with different income streams.
Also, by far, the largest cost is the installation of the "last mile" of "wire".
That has a very low maintenance cost. Once that is in and paid for, any
additional revenue is profit. The cost of termination equipment is really
pretty low. Probably less than $100 per connection.
> So let’s look at the realities here.. Today you can purchase a 7mb link
> from Time Warner for $46.95 per month. That equals $6.70 per month per Mb.
> Now let’s look at the other side, about the cheapest you can get a DS3
> (45Mbs) for is around $8000.00 per month which equates to $177.78 per
> month per Mb, or you can get a T1 for $600.00 per month which equals out
> to be $400.00 per month per Mb.
That's what they charge, not what it costs them.
>> “There are 10 other countries that beat the US cost per megabit… On top
>> of that, the US is 14th in speed.”
> While this may be true, this is a prime example of one using a statistic
> to prove ones point, without looking at all the other factors involved.
> Yes your right per mb they are cheaper, and statistically they do have
> faster speeds” Having been to the UK many times, I can tell you why this
> is. First off, the UK like many other European countries is small, the
> demands of putting in the infrastructure is significantly less than it
> is here in the United States. In fact Great Britain as a whole is 1/3
> the size of Texas, let alone the rest of the country.
And their population is generally proportional too. We are not talking about
electrifying rural communities here. The areas we are discussing have generally
similar population densities. Your personal situation may be different.
I live in the sticks too... 18 miles from AT&T World Headquarters. I can only
get 1.2 Mbps download from AT&T.
> That’s like saying
> my office has a gigabyte network in it, so why doesn’t my entire city?
> There is a big difference in cost in putting in high speed networks
> across 93,000 square miles than there is putting it in 3,537,000 square
But they can start. Right now, it's really not available anywhere in the US. I
heard about Comcast offering 50Mbps in Minneapolis-St. Paul for $150.
That's about $3/megabit and half the cost of must US costs.
I don't think that they will be losing money on that over the long term.
You also have to take into account that the big phone companies have been given
tax breaks to bring Internet to the country, but they haven't done that.
Let me give you another cost/price point. I once tried to get a static IP
address. The bottom line is that the price to me was about $40/month to do
that. I checked the costs and 8K IP address blocks from IANA cost about 16
*cents* ber *year* per IP address. That means if I got a block of 8, (network,
broadcast, router, 5 usable), it would cost them $1.28 per year and they charge
me $480.00. A 37500% markup is a great way to make money. On top of that,
after the fixed costs of setup, their maintenance cost for that is virtually zero.
More information about the SATLUG