Re: mechanicals for an open voting system

From: Cameron L. Spitzer <cls_at_truffula_dot_sj_dot_ca_dot_us>
Date: Tue Jun 27 2006 - 10:51:05 CDT

>Date: Tue, 27 Jun 2006 00:19:36 -0700
>From: Tom McCarty <>
>To: Open Voting Consortium discussion list <>
>Subject: Re: [OVC-discuss] mechanicals for an open voting system

>Cameron L. Spitzer wrote:

>>National emissions and safety
>>regulations were just the beginning. We met the CE Mark, but there
>>were also a bunch of government purchasing requirements.
>>In particular, the County of Los Angeles required either a UL label
>>or its own safety cert that was harder to get so everybody did UL.
>Hmm, I can't say I'm familiar with government purchasing regulations. I
>spoke to two different UL/CE testing agencies, and they both said it
>wasn't required.. then again, this was for another project, so I didn't
>say anything about government purchasing.

Perhaps I should elaborate. It makes economic sense to design
electronic hardware "for the world" rather than doing different
models for different markets. 3Com designed to an "international
envelope" which was basically the logical OR of all the requirements
we knew about. That envelope was more or less the shape of
the CE Mark requirements. Other things stuck out from under CE
but they were pretty small. A few frequencies where FCC B and
the Japanese equivalent were a little tighter than CE's CISPR B.

A few governments had their own safety rules tighter than
UL's, not regulations, just requirements for their own purchasing.
(Underwriters' Labs was not exactly known for in-depth scientific
understanding.) Certain large customers did their own testing for
IEEE compliance, and IEEE's safety-related specs were much tougher
than UL's. In particular, Ethernet is high-voltage isolated
to break up ground loops in real-world installations. Dell and HP
tested for that, while UL was oblivious to the need.

In an office building, the spike between two "grounds" created
when big electrical equipment goes on or off or lightning strikes
nearby can kill you if you're straddling those "grounds."
Or fry your computer if it's completing that circuit. The old
"RS232 standard" has this problem. If UL knew what they were
doing, nothing with a "serial port" and a PC-style "ground"
circuit would get a UL label.

Government purchasing drives a lot of design decisions in
that business. The Reagan Administration kept unix alive
by buying generic POSIX systems during a time when the unix vendors had
formed a circular firing squad. Reagan's Dept of Defense regarded
the emerging Microsoft monopoly as a threat to national security.
There's an amazing Rand study from that era that was ignored
by the pointy-haired bosses of the private sector.

>Regardless of whether it's required, I'd get a UL certification for
>anything we come up with.

I think you'll find the easiest way to do that is buying an
external power supply that has it. That's why cheap printers
and PC speakers do it that way.


OVC-discuss mailing list
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Fri Jun 30 23:17:12 2006

This archive was generated by hypermail 2.1.8 : Fri Jun 30 2006 - 23:17:12 CDT