Re: Small Linux platforms

From: Richard C. Johnson <dick_at_iwwco_dot_com>
Date: Sun Jul 17 2005 - 10:16:11 CDT

Hi Folks!
 
Well, why not go for it?
 
It would be relatively simple and inexpensive to organize a response to RFPs from a state for voting machines. This would mean more than 1,000 machines and would mean that, if the bid were accepted, the contract could even be used as collateral for the $$$ necessary to produce the system. More likely, a contract would mean periodic payments and collaboration with state election officials.
 
A strong, protected, open source system could be developed and tested/certified in a reasonable time frame. Why not simply seek a contract to do exactly that and go for it? Let the contract pay for development and testing.
 
Why would a state grant such a contract? Simply because it is the least expensive and most effective way to get the job done. Simply because it insures that the state will have its particular requirements met without question, rather than forcing adaptation to the inadequacies or ideosycracies of existing equipment offered for sale.
 
Yes, most state procurement is (hopelessly?) corrupt and liable to Diebold lobbying. Yes, most states are political snakepits and unwholesome environments for us simple engineers. Yes, we would have to swallow our engineering sectarianism and quickly agree on a common set of requirements. Yes, 2006 approaches rapidly. Yes, I know all that.
 
But...while we fiddle around, Rome (or its US analog) burns.
 
Rather, I would propose the Little Red Hen project--who will help bake the voting loaf? We all want to eat it (vote in a fair election on an honest machine), but who will get to decide whether it will be rye or 12 grain or whole wheat? The state contracting authorities, that's who. It will not be the OVC design or the Little Red Hen design. It will be the state's design and the hen and helpers will bake it to the state's specs.
 
Obviously, many of us know how to do this job in its basic essentials; why have not a committee of volunteers draft a set of requirements which could be quickly implemented with OpenSource, gather helpers (chiefly technical sales and marketing types who will go after the necessary contract), and then go for the gold ring? Once there is a contract, we can all volunteer to help.
 
No other approach will likely work. No endless squabbling over what is the "best" design; instead, we would execute under the contracted input of state voting officials. Let's just seize on our commonalities, our common engineering sense, and just do it. We believe in OpenSource--let's act on it.
 
Otherwise, we are all doomed to go out to McDiebold's for some fast junk voting in our respective states. IMHO.
 
-- Dick

Edward Cherlin <cherlin@pacbell.net> wrote:
On Saturday 16 July 2005 11:37, Karl Auerbach wrote:
> > Cameron L. Spitzer wrote:
> >> if I had to pick a platform for The Product I might start
> >> with Strongbox or Smoothwall or something simple and
> >> security oriented like that.
>
> It is not all that hard to build a very tiny, very focused
> Linux system that contains only the stuff needed to get the
> machine up and to do its job.
>
> I've been working on a project and my entire linux system -
> from bootloader to networking - is on the order of 25
> megabytes, and that includes a running Python environment. (I
> boot from a compact-flash card that came with a digital camera
> but was really too small for photographic work.)

I have been making this point since we started. The Simputer
holds a substantial Linux system in 32M of on-board Flash
memory. There are several Linux distributions, commercial and
free of cost, specifically for embedded systems (single-function
devices with special hardware). Encore Technologies would be
happy to build low-cost voting machines if we can demonstrate a
market for a thousand or more.

> I'm running a 2.6.11 kernel. Much of the non-kernel software
> comes via the busybox and buildroot projects -
> http://www.busybox.net/about.html and
> http://buildroot.uclibc.org/
>
> (My hardware platform is the three-ethernet wisp-router board
> - it draws something on the order of 3 to 5 watts, i.e. it
> could run all day off of a battery. I don't need video
> output, and it doesn't have any, so its probably not an
> adequate platform for a voting system.)

The monochrome Simputer runs with the screen on for six hours, on
two AA rechargeable batteries, if you don't need the backlight.
The color version, and the larger Mobilis tablet version, run
for proportionately shorter times.

> http://www.wisp-router.com/product_info.php?cPath=39&products_
>id=269&osCsid=2dbb898284cf7fbdfec87775e3f62e97
>
> --karl--

There are dozens, probably hundreds of such boards. I worked on
documentation for a number of them at LynuxWorks.

-- 
Edward Cherlin
Generalist & activist--Linux, languages, literacy and more
"A knot! Oh, do let me help to undo it!"
--Alice in Wonderland
http://cherlin.blogspot.com
_______________________________________________
OVC discuss mailing lists
Send requests to subscribe or unsubscribe to arthur@openvotingconsortium.org

_______________________________________________
OVC discuss mailing lists
Send requests to subscribe or unsubscribe to arthur@openvotingconsortium.org
==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Sun Jul 31 23:17:17 2005

This archive was generated by hypermail 2.1.8 : Thu Sep 15 2005 - 11:43:09 CDT