Re: Diebold on eBay

From: Joseph Lorenzo Hall <joehall_at_gmail_dot_com>
Date: Fri Jun 16 2006 - 17:07:03 CDT

Jim... the trick is in the 6.4.1(e) exception where it says no "source
code", "compilers" or "assemblers" be resident after testing.
Bytecode is none of those... however, the physically removable aspect
of the pcmcia cards are clearly within the intent of 4.2.2. Some of
you might be interested in a short document I wrote when Doug Jones,
myself and others from NCVI did a briefing a week ago on capitol hill
for congressmembers and their staff:

<http://josephhall.org/papers/DESI_vulns_background_briefing-20060607.pdf>

Also, I've finally published my paper on transparency and access to
source code... feel free to critique it:

Joseph Lorenzo Hall, Transparency and Access to Source Code in
E-Voting (forthcoming) in USENIX/ACCURATE Electronic Voting Technology
Workshop (2006). available at:
<http://josephhall.org/papers/jhall_evt06.pdf>

On 6/16/06, Jim March <jmarch@prodigy.net> wrote:
> David Jefferson wrote:
>
> >For the record, as we clearly stated, we did not have access to a TSx in the
> >writing of that report, or an AV-OS either (the primary subject of the report).
> >We had nothing but source code--no hardware.
> >
> >David
> >
> >
> >----- Original Message ----
> >From: Jim March <jmarch@prodigy.net>
> >
> >One primary thought is to photograph the innards for oddball data ports and
> >connections. Failure to do that was the single most glaring error from
> >the "Berkeley Report" of Febuary '06 on the TSx.
> >
>
> Right. So you had no idea the back pops off with eight standard
> phillips screws making a joke of the "keylocks".
>
> Did you know there was an active SD memory card slot? One long enough
> to hold a WiFi adapter meant for a Palm Pilot or whatever? Or that the
> internal modem socket had dual sets of interface pins and could hold a
> wide array of comm gear?
>
> I'm curious: why didn't you demand physical access to a TSx? You must
> have known the possibility for oddball ports existed?
>
> While we're at it...Black Box Voting obtained the Wyle report for the
> TSx version 4.6.4 (latest rev):
>
> http://www.bbvforums.org/cgi-bin/forums/board-auth.cgi?file=/2197/32746.html
>
> There's two noteworthy things in it:
>
> 1) The "ABinterp" files were analysed. That's the Accubasic
> interpreter. The interpreter is banned, but they examined it and passed
> it. In the Bowen hearing of 3-29-06 the Wyle reps denied this - they
> claimed that Accubasic had mistakenly been declared "COTS". Why did
> they lie to Bowen? My guess is, they didn't want to admit to having
> examined Accubasic and not spotting it as illegal. I think rather than
> doing source code review with human eyeballs they threw automated code
> review tools at it that were OK at checking some basic syntax issues but
> blew it completely when looking at the "big picture". I think they
> ended up declaring the structure of the code legal but missed that the
> whole thing was top-to-bottom illegal. Which means they had no clue how
> ANY of this stuff really worked.
>
> 2) What's not in there is any mention of examining customized Windows CE
> code. Bev Harris, myself and Doug Jones and God knows how many others
> have been screaming about Windows CE since 2003...yet here it is late
> 2005 and Wyle isn't checking CE? In the Bowen hearing Wyle said they
> rely on the vendor to declare items "COTS" and once they do, the labs
> don't check them at all. The FEC2002 specs says the ITA checks to see
> what is COTS and whether or not the COTS is unmodified.
>
> At the Bowen hearings Wyle and Systest reps claimed there was no way to
> make sure COTS files are actually unmodified. They've never heard of
> file compares? Load the same alleged "commercial off the shelf" program
> from a trusted source other than the vendor on the same box, see if the
> files are the same as the vendor's. They're not doing it.
>
> According to the Wyle people at the Bowen hearing, the CE files were
> withheld from scrutiny by Diebold. That's fraud.
>
> What the hell kind of bad joke is this?
>
> If custom code in the TSx hasn't been examined by anybody outside
> Diebold, the entire certification system at the state and fed levels has
> broken. If the certification system was defrauded by Diebold, their
> "NASED number" isn't legally worth a rusted dime.
>
> This is top to bottom sick, twisted, diseased.
>
> Professor Jefferson, did you even try and get access to a TSx?
>
> Jim
> _______________________________________________
> OVC-discuss mailing list
> OVC-discuss@listman.sonic.net
> http://lists.sonic.net/mailman/listinfo/ovc-discuss
>

-- 
Joseph Lorenzo Hall
PhD Student, UC Berkeley, School of Information
<http://josephhall.org/>
_______________________________________________
OVC-discuss mailing list
OVC-discuss@listman.sonic.net
http://lists.sonic.net/mailman/listinfo/ovc-discuss
==================================================================
= 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 23:17:07 2006

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