Bear Flag Republic Radio Weblog

Living out on the left coast

Last modified:
11/1/04; 5:43:00 PM

Feeds:

LIVE webcam Cumbres & Toltec rail yard in Chama, New Mexico.

Current BlogRoll.

Click here to send an email to the editor of this weblog.

Click to see the XML version of this web page.

Subscribe to "Bear Flag Republic Radio Weblog" in Radio UserLand.

Click on the coffee mug to add Steve Brune's Instant Outline to your Radio UserLand buddy list.

[Macro error: The server, api.google.com, returned a SOAP-ENV:Server fault: Exception from service object: Invalid authorization key:]

Click here to visit the Radio UserLand website.

 Saturday, October 9, 2004
Microbatteries Built on a Bed of Nails [Slashdot:]
comments < 12:09:38 PM        >

Broadband over Power Line communications appears to interfere with "RFID" tags used to identify inventory and other items.

Next week, on October 14th, the FCC will also announce its Report & Order regarding power line communication systems. Recently, the National Telecommunications and Information Adminstration (NTIA) has issued statements now generally supportive of BPL, even though the NTIA itself documented radio interference problems in the past.

In spite of its positive spin on the interference, the NTIA made a filing with the FCC on September 13, 2004, (PDF document) in which it requests a complete prohibition on BPL on frequencies allocated for aeronautical radio in the 1.7 to 30 Mhz band and radio navigation in the 74.8 to 75.2 Mhz band (see page 16). The NTIA found that when many BPL access devices operate simultaneously (as would be seen from an overhead aircraft), significant communication degradation occurs.

The NTIA also requests that BPL be prohibited within 1 km of all coastal (e.g. ship and aircraft) communications stations (page 18).

NTIA requests that BPL be prohibited within 80 km of radio astronomy facilities.

BPL systems within 4 km of FCC field offices and other "protected" locations must be coordinated between the power company and the organization, and should coordinate BPL within 80 km of certain radar receiver systems.

This document represents a rather stunning confirmation of the interference problems associated with BPL. Yet the FCC is hell bent on approving BPL on the basis that all interference must either be tolerated or can be magically "coordinated".

Meanwhile, yet another utility has backed away from BPL.

Also next week, my own local electric utility has scheduled a meeting with the Amateur Radio community to discuss BPL options in the Spokane, WA/Couer D'Alene, Idaho markets. They say they wish to meet with Amateur Radio operators before hand and work to resolve issues before they begin a BPL test system in this area. The meeting will be held Oct 14th at 7pm in the Avista headquarters building at Mission and Upriver, in Spokane, WA. (Noting that the date is the same as the FCC's R&O date, one must wonder if the utility has been tipped off as to the content of the R&O?) I hope to attend, as representative of the Red Cross. We use both amateur and non-amateur communications, including systems on frequencies that would likely not be protected from RF interference generated by BPL systems.
[Edward Mitchell: Common Sense Technology]
comments < 12:08:52 PM        >

Associated Press caught making up the news, again. As I noted last week, many election polls indicate a close election, so close that the only valid statistical conclusion is that neither candidate can be judged ahead of the other. Early this week, MSNBC got it right and called the race a "statisical dead heat". But other reporters are, unfortunatley, either somewhat dimwitted on elementary statistics, or they are outright making up the news stories.

The last couple of weeks, Newsweek boasted of a "tightening" race, when in fact, the confidence interval of the survey was so wide they could not conclude anything. Now, in AP's latest poll, Kerry is proclaimed as having a "slight lead". In fact, the data does not support that conclusion either - AP made up the news report to say that.

Once again, here is a simple explanation of a poll or survey. A sample of potential voters is surveyed to learn their voting preference. (Only 7% of those contacted even agree to participate in a poll, I've read. I was actually called a week ago and declined to participate.)

This produces an estimate of the population mean. For example, 50% for one candidate and 48% for the other. This however, is only a guess of how the entire population will make their selection.

When we expand the sample to an estimate for the entire population of all voters, we introduce a margin for error. That is, because we have only sampled a small number (relative to the population), we might have sampled a biased group and our survey results could be wrong.

From statistical theory, a confidence interval can be calculated giving us an idea of a likely error range for the estimate we make for the entire population, based on a small sample.

For example, we may have calculated a standard error of + or - 4% in our survey result. What that means is that - if we had measured the entire population instead of a small sample, the genuine population result will fall within the standard error or confidence interval or, for this example, 46% to 54% and 44% to 52%.

Since the actual population result could be anywhere within these ranges, we have no way of knowing who is ahead.

Every time a "journalist" reports yet another survey result with overlapping confidence intervals and tells us the race is tightening, they are publicly proclaiming that they have made up the news. This is very unfortunate and leads to genuine mistrust of news reporters and editors.

I won't write much about this during this election cycle - except to point further examples. Its just so frustrating to see such shodd work being passed off in news and headlines, daily.
[Edward Mitchell: Common Sense Technology]
comments < 9:05:22 AM        >

Wow! CNN writes "an initial poll indicates a statistical tie in Friday night's face-off".  Read it. CNN gets it exactly right and explains why. Finally!
[Edward Mitchell: Common Sense Technology]
comments < 9:04:00 AM        >


Archive:

October 2004
Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31            
Sep   Nov


Links:
428 CobraJet Registry
ARRL
ARRL Contest Branch
Book Pool
Chama NM Web Cam
Cumbres&Toltec Scenic RR
Dilbert
Digital Photography Review
eHam Radio on the Net
Home Power Magazine
Internet Radio Linking Project
John Robb
Mini Usa
Motoring File
NASA Human SpaceFlight
Open Secrets
Palo Alto Amateur Radio Association
Radio Amateur Satellite Corporation
Railroad News
Scripting News
SF Giants
South County Amateur Radio Emergency Service
Southern Pacific Historical and Technical Society
Space Weather
The Factory - Go back to 1965, the Shelby Mustang Factory, where on a quite night you can here Chevy RUSTING!
The Shifted Librarian
Through the Looking Glass
Tom's Hardware
William Shatner

Click to see the XML version of this web page.  Click here to send an email to the editor of this weblog.
Last Update: 11/1/04; 5:43:00 PM Copyright 2004 Steve Brune, All Rights Reserved.
Click here to visit the Radio UserLand website. Subscribe to "Bear Flag Republic Radio Weblog" in Radio UserLand.