Internal Memos: Diebold Doing End-Runs Around Certification Friday, 12 September 2003 (PDT) By Bev Harris – blackboxvoting.org http://www.blackboxvoting.com ******** If certification isn't being done properly, the whole house of cards falls. Below are actual copies of internal Diebold memos which show that uncertified software is being used in elections, and that Diebold programmers intentionally end-run the system. Quick backgrounder first, scroll down to see the memos. BACKGROUND Our voting system, which is part of the public commons has recently been privatized. When this happened, the counting of the votes, which must be a public process, subjected to the scrutiny of many eyes of plain old citizens, became a secret. The computerized systems that register voters, will soon sign voters into the polling place using a digital smart card, record the vote we cast, and tally it are now so secret they are not allowed to be examined by any citizens group, or even by academics like the computer scientists at major universities. The corporate justification for this secrecy is that these systems adhere to a list of "standards" put out by the Federal Election Commission, and that an "ITA" (Independent Testing Authority) carefully examines the voting system, which is then provided to states for their own certification. As it turns out, the states typically do not examine the computer code at all, relying instead on a "Logic and Accuracy" test which will not catch fraud and has frequently missed software programming errors that cause the machines to miscount. A Diebold message board has been used since 1999 to help technicians in the field interact with programmers to solve problems. The contents of this message board were quietly sent to reporters and activists around the world, most likely by a Diebold employee. In a letter to WiredNews, Diebold has acknowledged that these memos are from its own staff message boards. Without further commentary, judge for yourself whether Diebold has been following certification requirements: From Nel Finberg, Technical Writer, Diebold Election Systems (Note: Metamor/Ciber is the ITA assigned to certify the software) alteration of Audit Log in Access To: "support" Subject: alteration of Audit Log in Access From: "Nel Finberg" Date: Tue, 16 Oct 2001 23:31:30 -0700 Importance: Normal Jennifer Price at Metamor (about to be Ciber) has indicated that she can access the GEMS Access database and alter the Audit log without entering a password. What is the position of our development staff on this issue? Can we justify this? Or should this be anathema? Nel Reply from Ken Clark, principal engineer for Diebold Election Systems RE: alteration of Audit Log in Access To: Subject: RE: alteration of Audit Log in Access From: "Ken Clark" Date: Thu, 18 Oct 2001 09:55:02 -0700 Importance: Normal In-reply-to: Its a tough question, and it has a lot to do with perception. Of course everyone knows perception is reality. Right now you can open GEMS' .mdb file with MS-Access, and alter its contents. That includes the audit log. This isn't anything new. In VTS, you can open the database with progress and do the same. The same would go for anyone else's system using whatever database they are using. Hard drives are read-write entities. You can change their contents. Now, where the perception comes in is that its right now very *easy* to change the contents. Double click the .mdb file. Even technical wizards at Metamor (or Ciber, or whatever) can figure that one out. It is possible to put a secret password on the .mdb file to prevent Metamor from opening it with Access. I've threatened to put a password on the .mdb before when dealers/customers/support have done stupid things with the GEMS database structure using Access. Being able to end-run the database has admittedly got people out of a bind though. Jane (I think it was Jane) did some fancy footwork on the .mdb file in Gaston recently. I know our dealers do it. King County is famous for it. That's why we've never put a password on the file before. Note however that even if we put a password on the file, it doesn't really prove much. Someone has to know the password, else how would GEMS open it. So this technically brings us back to square one: the audit log is modifiable by that person at least (read, me). Back to perception though, if you don't bring this up you might skate through Metamor. There might be some clever crypto techniques to make it even harder to change the log (for me, they guy with the password that is). We're talking big changes here though, and at the moment largely theoretical ones. I'd doubt that any of our competitors are that clever. By the way, all of this is why Texas gets its sh*t in a knot over the log printer. Log printers are not read-write, so you don't have the problem. Of course if I were Texas I would be more worried about modifications to our electronic ballots than to our electron logs, but that is another story I guess. Bottom line on Metamor is to find out what it is going to take to make them happy. You can try the old standard of the NT password gains access to the operating system, and that after that point all bets are off. You have to trust the person with the NT password at least. This is all about Florida, and we have had VTS certified in Florida under the status quo for nearly ten years. I sense a loosing battle here though. The changes to put a password on the .mdb file are not trivial and probably not even backward compatible, but we'll do it if that is what it is going to take. Ken Reply by Nel Finberg RE: alteration of Audit Log in Access To: Subject: RE: alteration of Audit Log in Access From: "Nel Finberg" Date: Wed, 17 Oct 2001 14:48:16 -0700 Importance: Normal Thanks for the response, Ken. For now Metamor accepts the requirement to restrict the server password to authorized staff in the jurisdiction, and that it should be the responsibility of the jurisdiction to restrict knowledge of this password. So no action is necessary in this matter, at this time. Nel From Tyler to Ken Clark, Diebold Election Systems Re: Nichols Lab · To: >, · Subject: Re: Nichols Lab · From: "Tyler" · Date: Mon, 15 Feb 1999 14:04:19 -0600 In point of fact, the user documentation MUST be completed before attempting certification. It is my understanding that the documentation is a certification requirement. I don't know how closely Nichols will scrutinize the documentation, but I wouldn't feel comfortable going forward with certification with what we have for GEMS. Ostensibly, the documentation we submit to Nichols will become the "certified" documentation and we ostensibly shouldn't provide anything but that to customers. But then again, with regards to the entire NASED certification process, I can never quite get a handle on the relationship between "ostensible" and "reality."... :-) From Ken Clark RE: AVTS - Diagnostics & Installation · To: "Support Team (E-mail)" · Subject: RE: AVTS - Diagnostics & Installation · From: "Ken Clark" · Date: Tue, 6 Jul 1999 16:41:56 -0500 · Importance: Normal > From: owner-support@gesn.com On Behalf Of > Juan Rivera > > I do not feel that it is necessary or desired to do this on each and every > election. We, the manufacturer, are supposed to set the > procedures to follow > for this equipment since we build it. I hate more than anyone else in the company to bring up a certification issue with this, but a number of jurisdictions require a "system test" before every election. I just helped Knecht yesterday with an RFP from Riverside that required this. That is why the AccuVote displayes the silly ***System Test Passed*** message on boot up instead of "memory test passed", which is all it actually tests. No argument from me that it is pointless. You could probably get away with a batch file that prints "system test passed" for all I know. We will do something along those lines with the new unit after a memory test or whatever. Ken From Ken Clark RE: Testing sb100 database 1.14.2 (asap please) · To: "SUPPORT (E-mail)" · Subject: RE: Testing sb100 database 1.14.2 (asap please) · From: "Ken Clark" · Date: Fri, 7 Jan 2000 09:38:52 -0600 · Importance: Normal > Do you all think it would be a good idea to get Jeff Dean to send us 10 or > so precincts by eight parties with pre-printed test decks from one of the > California sites for Jane to test AccuVote and CC???? If so, > I'll call Jeff > Dean and set up asap. *Any* testing we can do on 1.14 is a good idea. With the risk of sounding alarmist, 1.14 really needs more testing. Even though much of GEMS looks the same from the outside, the guts changed substantially between 1.11 and 1.14. That's why you see all kinds of things completely unrelated to shadow races broken in the early 1.14 releases. Hats off to everyone posting 1.14 bug reports. Ken (The above memo is important because it documents that the "guts" of GEMS 1.14 are substantially changed from the certified version, 1.11 -- it was then used in elections, but according to Diebold's own chart of which versions were certified, version 1.14 was never certified.) From Steve Knecht 1.14 vs. 1.15 GEMS versions (uncertified versions used.) · To: "Global Support" · Subject: 1.14 vs. 1.15 GEMS versions · From: "Steve Knecht" · Date: Fri, 14 Jan 2000 17:10:34 -0800 Is it the intention of development staff that California March election will be run on some version of 1.14 or will we end up in the 1.15 range. Can you comment on the following: Are the changes being made now to 1.15 GEMS things that are in the ballot layout realm, and will not impact ballot processing, or tabulation issues?? In other words, is it possible that changes made from now on will break things we're starting to test, such as memory card up/download, central count, etc. We are beginning testing of 1.14.4 this week. Should we be testing with something else? I guess a little summary picture of what you expect over the next 3 weeks of testing would be helpful. I'd say we will have to lock down GEMS by mid - February, AVTS ballot station is to go on-line, along with a pollbook function by Feb. 7, but we are supposed to do testing and L&A prior to this. No panic yet, just wondering where we're going to lock some of this down for the March primary. Here is the related memo from Ken Clark: Needless to say, the changes were extensive. The paint is still wet, and I expect people will want some tweaks in functionality as well as the obligatory bug fixes. We'll treat the early 1.15 series as "prereleases" for LHS testing so California does not have to suffer. Once 1.15 looks at least as solid as 1.14 though, we'll end the 1.14 branch. 1.14 and earlier Databases will upgrade to 1.15 without harm as usual. People testing 1.14 are encouraged to try out 1.15 to avoid any surprises when they are forced to upgrade. Ken (Here is a whole series of odd memos pertaining to how they should handle the inconvenience of an uncertified version number popping up on the screen in Florida) From Greg Forsythe Florida Certified Versions · To: "Support" · Subject: Florida Certified Versions · From: "Greg Forsythe" · Date: Thu, 17 Feb 2000 11:12:02 -0500 · Organization: Global Election Systems, Inc Just received a call from Beverly Hill, Alachua County. She has a survey form from the state regarding versions and things. She is at the SA screen and the version is 1.92-15. Saturday, Feb. 12 she created a screen test database. This copy has 1.92-14. 1.92-14 is certified, 1.92-15 is not. SOLUTION REQUIRED! Greg Forsythe Re: Florida Certified Versions From Nel Finberg · To: · Subject: Re: Florida Certified Versions · From: "Nel Finberg" · Date: Thu, 17 Feb 2000 09:51:10 -0800 I am currently looking into the problem with Beverly. Nel From Greg Forsythe Re: Florida Certified Versions To: Subject: Re: Florida Certified Versions From: "Greg Forsythe" Date: Thu, 17 Feb 2000 12:55:09 -0500 Organization: Global Election Systems, Inc References: <007301bf7961$cda9d030$cdbb2fd1@greg Hernando's original database installed with Gunzip shows 1.92-09. Their copy has 1.92-14. It appears that when the database is gunzipped from the original diskette it carries the version from the source. When a copy is made on the customer's computer the version relates to the version the customer's computer programmed for. Solution might be to make the copy the official database showing the correct version. Comments ..... From Nel Finberg Re: Florida Certified Versions To: Subject: Re: Florida Certified Versions From: "Nel Finberg" Date: Thu, 17 Feb 2000 10:24:37 -0800 References: <007301bf7961$cda9d030$cdbb2fd1@greg The problem has been fixed. Nel From Greg Forsythe Re: Florida Certified Versions To: Subject: Re: Florida Certified Versions From: "Greg Forsythe" Date: Thu, 17 Feb 2000 13:33:55 -0500 Organization: Global Election Systems, Inc Well done Nel! How did you fix it? Did you delete the original and use the copy? If the diskettes had been sent in an unzipped format using Number 10, the Restore function in the System Administration Menu, would the database have come up with the version the customer's machine was running the first time and cause no problems? Greg From Nel Finberg Re: Florida Certified Versions To: Subject: Re: Florida Certified Versions From: "Nel Finberg" Date: Thu, 17 Feb 2000 13:13:25 -0800 References: What we failed to do at the time the date was repaired in Alachua's database (which showed question marks in the date field as a result of being prepared in patch 15) was set the database release file to patch 14. This is what I did this morning as well as set the release files for all of the remaining databases on their system. It should be noted that it could be that a lot of databases were initially set up with earlier versions of VTS, which we should be attentive to, given the stringency of certification in the state. I will clean up release files on the new Florida accounts in the next few days. Nel From Nel Finberg Re: Florida Certified Versions · To: · Subject: Re: Florida Certified Versions · From: "Nel Finberg" · Date: Thu, 17 Feb 2000 10:09:56 -0800 (uncertified versions used.) You are correct. However, Hernando's database should technically have been compiled using patch 14, not patch 9. We do want to make sure that ballots have been successfully tested and memory cards uploaded, particulary given the initial version conflict. It would be a good idea to get rid of the original diskette in order to remove the perception of version conflicts. Nel From Nel Finberg Re: Florida Certified Versions · To: · Subject: Re: Florida Certified Versions · From: "Nel Finberg" · Date: Thu, 17 Feb 2000 10:24:37 -0800 The problem has been fixed. Nel From Cathi Smothers GEMS Versions From: Cathi Smothers [mailto: csglobal@earthlink.net] Sent: Monday, June 05, 2000 5:02 PM To: Ken Clark Okay. Here's a "stupid new employee" question. I need to get the MN accounts upgraded to 1.16. How do I know which version of GEMS (i.e. 1.16.3, 1.16.4, etc.) to use? From Ken Clark RE: GEMS Versions To: Subject: RE: GEMS Versions From: "Ken Clark" Date: Mon, 5 Jun 2000 18:00:49 -0500 Importance: Normal In-reply-to: From: Cathi Smothers [mailto: csglobal@earthlink.net] Sent: Monday, June 05, 2000 5:02 PM To: Ken Clark Hope you don't mind the support list follow up. Certainly not a stupid question, and its worth a post on this topic every once in a while. Baring any certification issues, the latest stable release is what you want to upgrade accounts to. We always let people know when a release is for testing only in the release announcements. Testing releases are usually 1.X.1.Y releases. For example 1.16.1.1-6 were all testing releases. At some point we conclude that testing is going well, and declare the branch stable. A new testing branch is then opened, and only bug fixes go into the stable branch. Right now 1.16.latest is considered stable, 1.16.4 being the current release by my mail. How stable the stable branch really is has everything to do with how much testing by support it receives. Its fair to say the nature of this company and business make this process fairly informal, perhaps more so than I would like. Testing releases go out to customers when they shouldn't, and new features get added to stable branches when they shouldn't. It is not entirely undisciplined either though. Obviously you need to keep an eye on the support and bugtrack lists. Sometimes a bug slips into a stable branch, in which case its better to ship a version you trust, or wait for it to get corrected. Secondly, does the upgrade simply consist of installing the new executable file or are there other components that need to be installed as well? They are currently using 1.11.8. There are several components. The GEMS exe The ABasic directory and abasic.ini The Reports directory and reports.ini Locale.ini The DLL files shipped on the GEMS CD get updated from time-to-time as well, though not often. Is usually a good idea to order the CD for a long-haul upgrade. Its not really clear whether 1.11->1.14 qualifies as long haul or not. That really depends on your comfort level. There is never any harm in ordering a CD. Other frequently asked questions while I am here: Features are always propagated forward. I suppose one day we might remove a feature, but I've never seen it happen. Baring bugs, artwork and memory cards are still compatible after GEMS upgrade unless there is a big announcement to the contrary. Its only happened once that artwork was incompatible after upgrade, and memory cards have never been incompatible. The database changes between major releases (1.15->1.16) but not minor releases (1.16.1->1.16.2). You can downgrade out of trouble between minor releases, but a major release upgrade is a one way trip. Ken From Jeff Hintz Software for Los Angelas, CA To: "Support Team (E-mail)" Subject: Software for Los Angelas, CA From: "Jeff Hintz" Date: Thu, 31 Aug 2000 12:31:06 -0500 Importance: Normal I am going out to LA next week, and I would like to know what software version of Gems & AVTS is being sent out on their equipement. Thanks, Jeff Hintz (uncertified versions used.) From Rodney D Turner RE: Software for Los Angelas, CA To: Subject: Re: Software for Los Angelas, CA From: "Rodney D Turner" Date: Thu, 31 Aug 2000 13:31:55 -0500 References: <000201c01371$42bb57a0$0903a8c0@Jeff Hi Jeff, I have completed the computer for LA and Alameda. The computer for LA has GEMS 1-16-9 and the AVTS units have 3-13-1-4. The computer for Alameda has GEMS 1-16-10 and GEMS 1-16-9 ( there is a short-cut on the desktop for GEMS 1-16-9) the AVTS units have 3-13-1-4. All of the AVTS units including VIBS, have an OS of Windows NT. Because of NT, you have to remove the floppy from the drive during start-up. If you do not, NT changes the Imation drive from "A" to "D". If you forget to take out the disk from the drive, you have to restart without the disk in the drive to get it back to drive "A". Drop me a line Jeff, if you have any questions, or concerns. Rodney rodney@gesn.com From Talbot Iredale RE: Software for Los Angelas, CA To: Subject: Re: Software for Los Angelas, CA From: "Talbot Iredale" Date: Thu, 31 Aug 2000 12:03:50 -0700 (uncertified versions used.) Jeff and Rodney, LA and Alameda will need a revised version of GEMS and maybe BallotStation to support the import/export that they require. I am working on it now but I am certain there will be more changes. From Larry Dix RE: Software for Los Angelas, CA To: Subject: RE: Software for Los Angelas, CA From: "Larry Dix" Date: Thu, 31 Aug 2000 15:41:58 -0500 Importance: Normal Tab – Would you be willing to venture an outside guess as to when the revised GEMS version will be ready. This really becomes an issue since I need to coordinate staff to be onsite. Is this also the case for Alameda? Coordination of time and staff is everything on these 2 installs. Larry J. Dix Global Election Systems From Ken Clark Re: Gems-1-17-1 To: Subject: RE: GEMS-1-17-1 From: "Ken Clark" Date: Thu, 31 Aug 2000 17:33:04 -0500 Importance: Normal In-reply-to: <058c01c0138a$f76e27e0$ 1204a8c0@gesn.com> Is this a "testing" release or not? (Ashamed to ask). I think the hallucinations ought to be resurfacing with Steve already. Ken From Talbot Iredale (uncertified versions used.) RE: GEMS-1-17-1 To: Subject: Re: GEMS-1-17-1 From: "Talbot Iredale" Date: Thu, 31 Aug 2000 16:14:23 -0700 References: This is no more of a test release than 1.16.9 was though I would not be surprised if we have to make more changes to fully support LA and Alameda. Tab Re: Software for Los Angelas, CA · To: · Subject: Re: Software for Los Angelas, CA · From: "Steve Knecht" · Date: Fri, 1 Sep 2000 09:10:49 -0700 · References: <000201c01371$42bb57a0$0903a8c0@Jeff> (uncertified versions used.) Jeff, I think my thread may be out of sync, but discussion with Tab yesterday indicated that you'd be at least at 1.17.1 or higher to provide you with the "import" capability with their database. I believe Rodney / Mike would have to tell you what they loaded onto AVTS. Tab is still working on several programs that may affect what AVTS Rev and GEMS rev we both end up with. From Tari Runyan 1-17-7-5 testing · To: "SUPPORT (E-mail)" · Subject: 1-17-7-5 testing · From: "Tari Runyan" · Date: Mon, 23 Oct 2000 08:21:16 -0600 (uncertified versions used.) I have tested this version to the extent I am able - twice even and unless anyone else has discovered anything - i think it can be released to the Ca Counties - Let me know if anyone else has any concerns as I would like to get this out this morning. Thank you (There are dozens more memos like this, and hundreds that document the use of uncertified versions of the voting system, spanning a period from 1999 to 2003.) # # # # # Bev Harris is author of Black Box Voting: Ballot Tampering In The 21st Century … See http://www.blackboxvoting.com/ and it's activist arm http://www.blackboxvoting.org/ homepage: http://www.scoop.co.nz/mason/stories/HL0309/S00150.htm