The DUer Meets the Congressman – FogerRox-Rush Holt, D,NJ – Election Forum News 2/11

[autorank]

Election Fraud, Research & Discussion News 11 Feb 2007

1) Comment: Special Feature on user FogerRox, his blog, and the visit and comment by Congressman Rush Holt, D, NJ (Below and also Here)
2) Weekly Summary of Links

FogerRox has been a loyal and productive user on DU for a good while. He’s very active in the Elections Forum and blogs on Blue Jersey in his home state. He wrote an analysis of HR 811, Congressman Rush Holt’s bill on voting systems. Holt is a Democrat form NJ and heads the committee considering this critical bill.

GUESS WHO SHOWED UP? CONGRESSMAN RUSH HOLT. Rep. Holt responded to concerns about the bill and assured Foger and the BlueJersey folks:

“You may rest assured that if there were any language in the bill that required clarification, I will address it in the ordinary course as the bill moves through the Committee process. “ Rep Holt to blogger FogerRox
Fri Feb 09, 2007 at 16:13:51 PM EST

FogerRox began with the positives about the bill. The blog was placed on the front page almost immediately since it was about voting and also the sponsor, Holt is form NJ:]

Foger went along with more commentary, mostly positive.

...and then offered a critique...

He then summarized some draw backs on HR 811

…and guess who who showed up? Congressman Rush Holt, D, NJ himself.
Here’s what Cong. Holt had to say.

How often does a Congressman show up at a blog. Great job FogerRox and even handed, skillled treatment of your guest.

[flyarm]

1. That is the Rush Holt i Know!! and have assured people here at DU he

is a man of incredible integrity!

Understand this..Holt had a race where he held out for a very long time when he demanded a recount of an election that he supposedly lost..he was sure he hadn't lost and he demanded a recount ..and after the recount ..he had won!

Holt is a man who understands recounts! and the value of recounts and audits!!

He is no one's fool!

fly

[Febble]

5. Thanks for that

it is good to know.

[Bill Bored]

24. Maybe he's just been getting a lot of bad advice lately then huh? nt

[FogerRox]

2. My response

Thank you Congressman Holt for stopping by. (4.00 / 1)
Your reputation of accessibility precedes you.
I out was protesting the Iraq war here in Montclair, just got back home and I see by your answers I have not effectively voiced my concerns or the concerns I have read about on the email lists I subscribe to. Forgive me. I do hope, here, to better clarify those concerns.

Congressman Holt wrote

First, the post states that my legislation does not require optical scan machines to produce a voter-verified paper ballot.

Roger wrote:

HR 811 has a specific opscanner exemption, no audit of paper

1) Congressman I think you missed the direction of stated concern. I understand that HR 811 labels Paper scanned ballots as VVPB. Thats good, very good. But does HR 811 mandate a VVPB audit for paper ballots scanned by optical scanners?

SEC. 2. PROMOTING ACCURACY, INTEGRITY, AND SECURITY THROUGH VOTER-VERIFIED PERMANENT PAPER BALLOT.
(a)(1)(A)`(i) The voting system shall require the use of or produce an individual voter-verified paper ballot of the voter's vote that shall be created by or made available for inspection and verification by the voter before the voter's vote is cast and counted. For purposes of this clause, examples of such a ballot include a paper ballot marked by the voter for the purpose of being counted by hand or read by an optical scanner...

I see inspection and verification. I don't see the word audit, I would offer this:

available for inspection and verification by the voter before the voter's vote is cast, counted and audited as described in section 322.

Invoking the full weight of the improved "Holt Audit" for all VVPB's, as in section 322, leaves no wiggle room. I would prefer to see this in Section 2 clearly stated.

2)Washington attorney Paul Lehto who sued Sequoia, voiced his concern on the undisclosed software issue thusly.

My take is that the attempt even to escrow the source code will also fail. Analysis on that below (that's separate from open source development)
The Holt bill, while seeming to have a very broad duty on the part of state and local government officials to "disclose" source code and object code (etc) to the federal government, has very weak langage on how the local or state government officials are going to get their paws on all this source code in all jurisdictions in the first place. The relevant text simply states that manufacturers shall provide the appropriate election official with the information necessary for the official to provide information to the Commission . It does not say "all" information, and it does not specify source code, just the amount of "information" necessary for the official to provide some "information" to the Commission.

If the source code is to be disclosed then that trade secret "property" is being lost because it will be disclosed (these trade secrets) to all the world via public request and therefore no longer a secret. So the claim will be made by the manufacturers that (A) because the law abhors a forfeiture, the language should not be interpreted to require them to "forfeit" millions in investment in their view via losing their source code in the first place, and alternatively (B) if the language IS sufficient to command a forefeiture of their trade secret property, then their trade secret "property" worth millions has been subject to a "regulatory taking" such that they should be compensated. Their investment, they will argue, has been nationalized in effect, and they should be compensated handsomely for that.

I will be very surprised if the states and locals get their hands on all source codes for free. Instead, it appears that the local officials obligations to disclose are so broad (they arguably even include the source code on firmware embedded on hardware that the vendors literally don't control) that the state and local officials will be making at least a partial defense of impossibility, and they will probably also add that it was impossible for them to get a hold of the source code from the vendor like Diebold or Sequoia, in addition to the firmware or what have you on all the other hardware parts.

Here again, you'd think Holt II was providing us with source code that would be open, but it is unlikely to happen that way. Sequoia filed a declaration in my case claiming their trade secret investment was around $6M without considering lost profits from their purported 'investment."

Note also that in section 12, the state and local officials are required to document the secure chain of custody of the trade secret software, among other things. While chain of custody is a good thing for the election, this also has the effect of forcing the local government to document that secrecy has been preserved as to the trade secret software, which is the primary argument to destroy a trade secret once it is admitted to exist. This helps make the trade secrecy claims of the vendor stronger as well when they sue for damages or an injunction to stop the disclosure of their trade secrets to anyone not under a confidentiality agreement with them.

------------------------------------------

Compare Section (9)'s broad language for local officials:

9) PROHIBITION OF USE OF UNDISCLOSED SOFTWARE IN VOTING SYSTEMS.-No voting system used in an election for Federal office shall at any time contain or use any software not certified by the State for use in the election or any software undisclosed to the State in the certification process. The appropriate election official shall disclose, in electronic form, the source code, object code, and executable representation of the voting system software and firmware to the Commission, including ballot programming files, and the Commission shall make that source code, object code, executable representation, and ballot programming files available for inspection promptly upon request to any person.

with section 12's weak language for vendors:

(i) The manufacturer and the election officials shall document the secure chain of custody for the handling of all software, hardware, vote storage media, and ballots used in connection with voting systems, and shall make the information available upon request to the Commission.

(ii) (ed. note: Only criminals not allowed in elections per this section are election, accounting, or computer security fraud criminals, but con men, national security felons, embezzlers and many others are very welcome).

(iii) The manufacturer shall provide the appropriate election official with the information necessary for the official to provide information to the Commission under paragraph (9).

Congressman on your 3rd point, please correct me if I am wrong. Since all DRE's would generate a VVPB, and opscan paper ballots are themselves VVPB. If a state holds a recount because of a close race, said recount will be counting said VVPB. Based on this sentence in Section 327:

Nothing in the previous sentence may be construed to waive the application of any other provision of this Act to any election (including the ballot verification and audit capacity requirements of section 301(a)(2))

Am I citing the correct section in describing why no "close race" machine recount will be allowed under HR 811, that all "close race" recounts will be using VVPB as mandated by HR 811?
Previously I had just scanned this, but have now read this fully, sections 322 & 323 seem quite good, these sections are the "Guts" of the "Holt Audit", having a 10% audit rate in a close race is very powerful language. IIRC this is much stronger than HR 550. The "Holt Audit" is what garnered the term "The Gold standard". Now in HR 811 it might be the "Holt Audit" might be the "Gold Standard with decorative Platinum inlays and Diamonds".

After reading VVPB law from 13 states I get the feeling that parts of HR 811 are very well crafted, other parts are not, some are even vague and seem to be teeming with loopholes.

Impeach, Indict.


--------------------------------------------------------------------------------

by: FogerRox @ Fri Feb 09, 2007 at 20:36:14 PM EST

< Parent | Reply >


Link to blog, to get the whole picture:

http://www.bluejersey.net/showDiary.do?diaryId=3976

[Ellipsis]

4. You have to admit the fact that he posted on your site is pretty "effin" cool in itself.

A response within in 8 hours ...and dialog.

What a dude! We're sending the dixie chicks to your door.

:toast:

[FogerRox]

19. Well, not my site, but my diary. Bluejersey is like the D-KOS of Jersey

[Ellipsis]

23. My bad. Still you'd think you would be able to finagle a sit-down with the man.

While one by default must respect Holts contribution to 811 (even if parts of it do suck) He's a physician not a technologist. And while I would surely wouldn't want you to remove my appendix, ...election technology oozes from you pores (not to blow smoke up your skirt). You are from the same state, both Dems, it is to his benefit to have a private meeting with you to discuss how one can take a *good bill and make it a bullet proof excellent. To compromise this bill because of an installed base of vulnerable voting equipment as a set metric ain't going to fix the problem. They rushed to market and need come to terms with "they done got bamboozled" by carpet baggers and snake oil salesman. They need make these guys liable for providing equipment they (the vendors) HAD TO KNOW was substandard when it came to security.

While I know the likelihood of this is nil I still had to say it.



*positive marketing statement.

[Bill Bored]

25. Holt is NOT a physician. He's a physicist and he should know better!

He can probably do the fricking auditing math in his head for God's sake, yet his auditing protocol sucks!

A lot of serious people have been working on auditing over the last year or two, and some for much longer than that. Yet Holt didn't consult with any of them.

And for some reason he's got a bug up his ass about allowing GEMS servers on the Internet. I'm still looking for a computer scientist that thinks this is a good idea. Haven't found one yet.

They might cover for him on the audits, not having studied the problem using actual election results. (Too much work that is.) But no one has said they think GEMS servers belong on the Internet.

I'd say see if Feinstein's Senate bill fixes some of the bugs in Holt's and then decide which bill if any to support or not support.

And there are still other members of Congress to be heard from.

[Ellipsis]

29. Jumped to the bottom of hi bio and misread.

His wife is the physician. ( Now I feel dumb) But you do get my point (luckily) of outside experts giving perspective . Yes... he should know better shouldn't he...

"And for some reason he's got a bug up his ass about allowing GEMS servers on the Internet. I'm still looking for a computer scientist that thinks this is a good idea. Haven't found one yet.

They might cover for him on the audits, not having studied the problem using actual election results. (Too much work that is.) But no one has said they think GEMS servers belong on the Internet."

This I get full well.

Outside of the obvious vulnerableities while on the net ..the logistical issues of having to get "collected" votes to tabulation... In Cuyahoga county they hired what... 70 cabs to run memory cards around, makes the case for votes to be tabulated where they are cast ...which is as it should be. Decentralizing tabulation makes it difficult to manipulate the vote in a meaningful way.

[Bill Bored]

30. Dial-up can be used for all that tabulation stuff.

The Internet is raising the risk to whole other level. Besides if the precinct totals are posted on election night, the tabulator can be checked. It's more the pre-election stuff I'm worried about. Ballot definition programming can be easily hacked.

[Ellipsis]

31. ... just read up on Ballot definition files a bit

First thanks, your reply gives the epiphany.

Ballot definition file, The phrase is self defining. In Google reads No "certification" and "non auditable" pop up quite often. One can see how simple it would be to make simple changes in a file in an encrypted or secret working environment.

Coming from the non-political (until 2000) print work-flow side of life... the transition to voting equipment and it's procedures has had it's learning curve, especially when one adds potential deceit into the formula.

Ballot definition file -- to me we are talking about page description Geometry (sorta) ...postscript.

The more I read up on this industry the more pissed off I get. They take the concept of a ballot, voting, collecting and counting votes and exponentially complicate it in order to an create a new vertical market so a couple of brothers can grab majority market share. As a vehicle, HAVA is created under the guise of helping the handicapped and disabled with an intension of forced deployment (premature) in four years. They hang a carrot of federal money for implementation, largely fostered by a now convicted congressman from Ohio where the election of 2004 was called into question. "Cronyism city" everywhere you look, with politicians in the lead.

Segue:

Your IEEE post was quite cool, though "rush" to change or set a spec. is not a traditinally how those guys work. Still, It will be intriguing to see what comes from it.

Peace

[OnTheOtherHand]

6. actually, there's a problem with the 10/5/3, although

obviously it is stronger than a flat 2%. For instance, take New Jersey 7, where Linda Stender lost by about 1.7 percentage points (indicating a 5% audit under HR 811). I haven't looked at the actual precinct-level returns from NJ-07, but I have a little algorithm for estimating how many precincts would have to be miscounted in order to alter the outcome of an election, and how likely an audit of a given size is to find at least one of those precincts. NJ-07 came out at 37%. Now, that's pretty much a worst-case percentage (for instance, it assumes that Stender 'should have' won by a handful of votes), but still, it isn't very impressive. And there were other House races where the HR 811 audit would do even worse.

The frustrating thing about this is that with the same percentage margin in a New Jersey statewide race, a 5% audit is actually much larger than one would need. A 2% audit would give about a 99.6% probability of detecting outcome-altering miscount in that case. So, the basic characteristic of fixed-tier audits that don't take account of size is that they tend to be 'too big' in statewide races (not that I ever mind counting ballots) and too small in House races. Treating statewide and House races equivalently imposes a tradeoff that doesn't necessarily exist: as far as I can tell, we could get confidence levels well over 90% in every federal race even without auditing any more ballots!

It's not that the HR 811 percentages open the door to massive undetected miscount. Suppose, for the sake of illustration, that there are 10 House races in each of which the HR 811 audit would have a 50% chance of detecting the minimum outcome-altering miscount. And further suppose that every one of those races actually has outcome-altering miscount (in fact, minimum outcome-altering miscount). Then there would be about a 99.9% probability of detecting the miscount in at least one of those races -- most likely in 5 of the 10. For folks who think that the Republicans stole 3% or more of the 2006 House vote, be advised that statistically speaking (assuming the ballots are protected, etc.) the HR 811 audit would detect the problem, even if not in every race. The problem is subtler than that, but it still bothers me: the HR 811 audit will typically offer the least confidence in relatively close races, where people are most skeptical to begin with.

I guess some smart people think this is the best they can do, but so far I don't see why it is. HR 811 does allow states to implement stricter audit standards -- although it isn't clear to me whether it would allow states to, say, settle for 99% confidence instead of 99.999999% confidence in some races in order to help offset the cost of attaining the same high confidence in other races.

Don't get me wrong, though. If I were staring at paperless DREs with 0% manual audits, 10%/5%/3% would be pretty darn good. Some people seem to be committed to making this bill sound much worse than it is; I'm not going to worry about feeding them talking points, because they obviously don't need me for that role anyway.

[FogerRox]

9. Thats another point, all my local math heads say the same

In a Municiple race with 8k votes cast, a larger sample (maybe 10%)is needed to gain the same level of confidence as:


A state wide race with 2 million votes could gain the same level of confidence with a smaller sample (Maybe 2%).


I think this issue was discussed a year ago, in conjunction with the Mercuri ping pong ball- lotto style precinct selection scenario.

[kster]

27. Second round goes to FogerRox

Will Holt get back up?

Stay Tuned!

[Liberty Belle]

3. how about amending it to include a cap on recount costs?

San Diego's Registrar charged around 10 times as much per vote as a Registrar in a neighboring county.

You also need to change the rules so that no Congressman can be officially sworn in if results have not yet been certified by the local registrar, as a San Diego court ruled that it's virtually impossible to UNSEAT a member of Congress once they are sworn in, unless Congress itself decides to do so.

Finally, we must make sure that NO ballots are counted on electronic central tabulators. I would much prefer to see hand-counting, preferably at polling places first.

I've also seen games played on certification dates, timing them to hit holiday weekends, so the clock runs out for requesting recounts. Perhaps this could be addressed as well.

I am a journalist who has interviewed a Diebold whistleblower and covered election fraud both nationally and in San Diego's various disputed elections (Busby/Bilbray, Donna Fry, etc.)

[FogerRox]

12. Then you need to consider fraud or sumsuch charges in SAN Diego

[kpete]

7. WOW!

Terrific work!

[truckin]

8. Atta boy Foger! With dedicated activists like you and sincere &

accessible politicians,like Rush Holt, there is a chance we will end up with good legislation.

[FogerRox]

13. truckin, you've been here for yrs, and yet your post count never changes

WTF is up with DAT ?

Good to see ya buddy.

[Bill Bored]

26. It's cause he's just truckin'! (And lurkin'!) nt

[Mnemosyne]

10. Alright Foger! You Rox! Thanks for the update auto!

:hi:

[autorank]

14. Foger does Rox, eh. He had the perfect stance.

As a citizen, he discussed an important issue. He was gracious but not intimidated by the Congressman and he continued the dialogue. I really got a kick out of it!

:hi:

[doublethink]

11. K&R.

Awesome FogerRox !!! :thumbsup: :applause: Peace.

[galloglas]

15. K and R

KandR--Har!

Way to Go, Rajah!!!

Ask him how much he needs to be pushed to advocate

Paper Ballots, Hand-Counted? I've got a Show Me Bulldozer!

[Kurovski]

16. Oy! I'm kvelling here!

A K&R for Foger's great activism and Sir auto's post bringing it to the fore. Many thanks.

[autorank]

18. Foger is a power broker! Cool stuff.

:hi:

[Kurovski]

28. I'm going to harangue him for a lobbyist position.

Just in case.:hi:

[DemReadingDU]

17. Awesome!

[nicknameless]

20. It's very good to hear Holt's response

I hope he continues addressing concerns this way.

Does 811 call for the dumping of DREs, or just an addition of printers to them?
ALL DREs have GOT to go. The souped-up TSx test in CA had a near-33% failure rate.

(R'ed earlier)

[FogerRox]

22. 811 is not very well written, parts of it suck.

[FogerRox]

21. Heres the face behind the computer

Thanks eveyone, I can't help but think of that saying:

"Sometimes its better to be lucky than good". I think the word went around that somebody posted up a good diary on hr 811 on bluejersey and the word got to someone in Congressman Holts office.

The folks at bluejersey knew I put a lot of effort into that diary, & based on all the good work these people have done



Just a little group I started.

I think the powers that be @ blue jersey then decided to promote my dairy to the front page. Of course Autorank found out, and hes like a dog with a rag doll, he won't let go.

ARRRrr, ARrrrrr


Thanks.

Roger Fox.

[BeFree]

32. Way to go Foger

This is an example of democracy in action. One person CAN make a difference.

[FogerRox]

33. BeFree, you have only one heart? I made it 2.

Hey Auto, kpete, you folks got a spare heart for befree?