I love this product...

We purchased the squeezebox from Slim Devices and absolutely love it! This is exactly the kind of innovative product that WiFi is screaming for...

Within minutes I was able to connect this tiny box to my stereo and to my home wireless network. The box finds a computer on my network running server software provided by Slim Devices and I am immediately able to play any MP3 in my collection, over my home network and on the stereo in the living room.

Oh, and any device connected to the network with a web browser can control the jukebox...

Who wants a computer in the living room? Who needs extra wires around the stereo? One little box, using WiFi to connect my stereo to my home network and I am done. Can it get better than this?

Why hasn't anyone produced a WiFi baby monitor? WiFi walkie talkies for kids? And I am not talking about analog over unregulated bandwidth -- I am talking VoIP over WiFi. The technology is all there... Let me know if you have heard of other innovative products using WiFi.

The VoIP debate on interconnecting

A recent C Net Article reports on the FCC judgement that "pure VoIP" is no different from email or any other IP based application and thus should not be subject to federal regulation regarding telephone services.

However the FCC has left open the question of what happens when VoIP services interface with the PSTN (public switched telephone network). Perhaps then they should be regulated?

This issue is a red herring. Anyone who connects a VoIP service to the PSTN pays an interconnect charge -- in California at least $10 a month for a single phone line or a lot more for a T1 PRI bringing 23 lines in on one circuit. So the regulation is available at the interchange point... how do you make the argument that the regulation should then extend to other parts of the VoIP network if a "pure VoIP" network has no such regulation?

And how do you keep the "pure VoIP" networks from interconnecting to the blended networks?

This is a meager attempt by the phone companies to at least create some FUD for operators like Vonage while they line up legal arguments and lobbying dollars to slow down VoIP, which, left unchecked, will destroy their current business models.

Amazon Honor System

I really enjoy writing this blog, and I believe there are folks out there that enjoy reading... So I have decided to try out the "Amazon Honor System" -- a way of making voluntary donations to web publishers you'd like to support. My commitment to you is that the more donations I receive, the more time I will spend on my blog - bringing you ideas, information, analysis, and ponderings on life, technology and philosophy.

I'd love your comments -- positive or negative -- on this voluntary support system, since I am just trying it out for the first time.

The economics of VoIP

I spoke with two companies last week that are in the long distance VoIP telephony business and it made me start thinking about the economics of VoIP telephony. My conclusion -- it really doesn't take that much money or work! We should all build our own long distance companies.

Here is some back of the envelope economics:

Variable Costs

There are three kinds of variable costs -- bandwidth, co-location facilities, and the cost of the T1 PRI from the phone company, allowing you to interconnect with the PSTN (public switched telephone network -- the analog stuff).

Bandwidth will cost you about .0001 cents per minute.

Co-location space works out to be about .00004 a minute

Finally the monthly fee for the local T1/PRI is about .00035 per minute

Total cost at maximum efficiency for your VoIP telephone network -- approximately .0005 per minute

No one can reach maximum efficiency -- but if you assume a bell curve for utilization against the capacity you need to build the variable cost is still less than .001 per minute of talk time. There is a minimum scale you have to be at to get to this rate... about 7 million minutes a month.

Now that is a lot of time on the phone, and one person isn't going to talk that much. A good comparison point -- common nationwide long distance deals are available for about $0.05 a minute.

To get to that price you'd need to use your VoIP circuit at least 70,000 minutes a month... 2333 per day... 97 minutes per hour... If you think about the average customer as having 2 hours of long distance calls a week, this is less than 150 customers for your phone service... of course they all have to want to call the same single destination if you only install one circuit...

Equipment and software

You will need a little bit of hardware for this exercise

First you need to get Linux based Intel machines. You can get solid machines for $2,000.

Then you need to download and install Asterisk -- an open source hybrid TDM/VoIP PBX.

You'll also need a card for each PCs to manage the T1 PRIs -- A company called Digium sells cards for Asterisk at $1500 each.

And you'll need at least one of these machines on each end of your connection.

$8,000 in capital investment to be a phone company.

Of course there is your time, or the time of a technician that you hire.

Still, if you could build a clientele between two points -- say India and the US -- you could probably offer them $0.02 a minute phone service and make a solid profit. And save your customers an enormous amount.

The traditional phone companies have to be worried.