Brad, as you know I have been a crusader for secure and verifiable elections for nearly 15 years now, especially in California. And I know you and I have agreed over the years on most issues regarding verifiable elections. You and I have also talked about SB 360. I honestly think you are overblowing your complaints about it. It is not a perfect bill, but I believe on balance it greatly improves the security of elections over the previous situation.

You are right that the bill gives the SoS authority to certify new voting systems for use in pilot projects to test them out in real elections. But you fail to mention that any such new voting system MUST produce a voter-verifiable paper record, and such pilot elections MUST be protected by a risk-limiting audit of the paper records. The notion of a risk limiting audit is the most powerful new security idea in election administration in the last century I believe. It can detect and correct almost all kinds of error, including fraud, regardless of how buggy or malware-infected the voting machines might be, as long as the paper trail records or paper ballots themselves are reliably preserved. This is a huge, sea-change in election administration thinking!

Now the law does not require risk limiting audits for all elections, mostly because the current voting equipment is not designed to make such audits easy. But in a pilot for new voting systems that support is required, and so the pilot process is a step toward moving all elections over time to support of risk-limiting audit procedures. It is important to understand that risk limit audits are fundamentally a much more powerful protection of election integrity than any kind pre-election certification testing or disclosed source or security review, however thorough. So as paranoid as I am about voting system security, I am not worried about the use of untested systems in limited pilots (the limitations are themselves in SB 360) because of these protections.

It is true that SB 360 does away with reliance on the federal EAC certification process. But you fail to mention that that process has essentially totally collapsed anyway. All four EAC commissioner seats are vacant and they will likely never be filled. Almost certainly the entire agency will be eliminated in one of the next few federal budget fights. If that happens, then California would be in a legal limbo with a requirement for voting systems to be certified by an agency that does not exist! You also fail to mention that SB 360 requires that SoS certify that any new voting systems used in CA satisfy new standards that must be at least as strong as those currently promulgated by the EAC. In other words, CA must essentially duplicate and improve upon the federal EAC process. The EAC refused to adopt any kind of software independence standard when it had the chance, and now, with no commissioners, it cannot do anything to strengthen standards. Frankly, knowing how the EAC was been more or less neutered by vendor interests from the start, I don’t see California’s disconnection from it to be a problem.

SB 360 allows counties (like L.A. County) to develop their own voting systems and submit them for certification to the SoS. This is huge because it opens a major crack in the historical monopoly relationship between vendors and their captive county customers. If the L.A. system is successful, or the Austin system, then they might be licensed to other jurisdictions, especially large metropolitan jurisdictions, and the stranglehold that proprietary vendors have over the elections process will be correspondingly reduced. Have you noticed that the vendors have been much less in evidence in recent years, and their influence substantially reduced already?

SB 360 does not mandate nonproprietary or disclosed source voting systems. But it does explicitly ALLOW jurisdictions to adopt them. This is a very important step because it effectively denies the canard that vendors have been using forever that keeping their source code proprietary is an important guarantee of the security of the system, which it is not. As far as I know, this is the first time the term “disclosed source” has ever been used in election legislation anywhere. It represents a new consciousness about the role of software and software development processes in security. I would like to see Los Angeles adopt disclosed source as a principle; we’ll see if they can be persuaded to do so.

Now as I said, there are imperfections in any comprehensive new law. But in this case I think you are failing to even mention the major positive themes in SB 360 that make it a genuine broad reform and major improvement in the security outlook for California elections. Go ahead and point to imperfections as you see them, but I think a more balanced reporting acknowledging the positive features of the would be helpful.

David Jefferson said @ 1:

It is not a perfect bill, but I believe on balance it greatly improves the security of elections over the previous situation.

First, thanks for sharing your detailed thoughts here. As you may have guessed, I disagree with a number of them, and will try to hit some of those areas of disagreement below.

you fail to mention that any such new voting system MUST produce a voter-verifiable paper record, and such pilot elections MUST be protected by a risk-limiting audit of the paper records. The notion of a risk limiting audit is the most powerful new security idea in election administration in the last century I believe.

Pre-SB 360 law already required, as I recall, that all voting systems in CA "MUST producer a voter-verifiable paper record". But, in any case, the "risk-limiting audit" you reference, as you know, can do little, if anything, to detect errors or fraud on the computer-printed ballots that are being planned as a part of LA County's new, 100% unverifiable touch-screen voting system. Such an audit could not detect misprinted ballots of the type that you'll recall was created when I attempted to vote on LA County's current computer-printed ballots system back in 2008 (when the system misprinted 4 out of 12 of my own votes.)

We discussed that fact previously, as I recall (which I'm only mentioning since you yourself had disclosed that you and I had discussed SB 360 during the course of my reporting on it.)

It can detect and correct almost all kinds of error, including fraud, regardless of how buggy or malware-infected the voting machines might be, as long as the paper trail records or paper ballots themselves are reliably preserved.

That's simply incorrect. Even if the ballots are reliably preserved (which they may or may not be), misprinted ballots on a computer-printed ballots system would most likely never be detected in a post-election spotcheck of the kind your describe and laud as reason to celebrate SB 360.

It is true that SB 360 does away with reliance on the federal EAC certification process. But you fail to mention that that process has essentially totally collapsed anyway.

That is simply not true either, as I reported in my very first story that touched on SB 360 back in April. That article, detailing my interview with LA County Registrar-Recorder/County Clerk Dean Logan, included both his comments to the effect you offer above, and detailed responses to them by several EAC officials (the ones who are actively, on a daily basis, carrying out the EAC certification process you describe as "totally collapsed"). You know very well, I suspect, that I'm no fan of the EAC. I've probably been as hard on them publicly over the years as anyone in the nation, and there are many flaws in their processes, but let's stick to the demonstrable facts here. Or, at least, the facts as offered by several different named EAC officials, on record, in direct response to Logan when he offered nearly the same criticism of the EAC you offer above.

You also fail to mention that SB 360 requires that SoS certify that any new voting systems used in CA satisfy new standards that must be at least as strong as those currently promulgated by the EAC. In other words, CA must essentially duplicate and improve upon the federal EAC process.

I spoke to that point somewhat in my 9/10/13 article just after the passage of the bill. Specifically, I (along with quotes from Jim Soper of the Voting Rights Task Force in CA, one of the few folks willing to go on record to discuss the bill) discussed that there is no standard for determining how/if state testing standards are "as strong" as the current EAC standards. That determination, under SB 360 is, supposedly, to be made by the Sec. of State. (Let's say he's a guy named Bruce McPherson, for example. The same guy who previously decided Diebold touch-screen systems, with "voter verifiable paper records" could be certified for use in CA despite warnings from his own testers --- which included you, as I recall --- warning that testers had confirmed both the very serious security flaws discovered in that system by Harri Hursti in Utah and Florida previously, as well as 16 other bugs you all described at the time as "a more dangerous family of vulnerabilities" which "go well beyond" what was discovered by Hursti. As you'll recall, SoS McPherson certified for those systems for use anyway, because he had the power to do so.)

Making matters worse, the security standards you believe will have to be met for certification in CA don't even have to be met at all for pilot systems used in real live elections, according to SB 360 as passed, signed and adopted for use as of Jan. 1, 2014. As I quoted Soper in the above referenced article, while some of the text in the bill suggests pilot systems must be "tested", before use: "it's not even specified if the [pilot system used in a live election] must pass the test before being used."

At least prior to SB 360, if you had an SoS as irresponsible and corrupted as McPherson, the system would had to have undergone a completely seperate, independent level of testing at the U.S. EAC (or other federal body). Now all power is vested in a single person in CA. Can't imagine how you would see that as a good idea, given your own personal experience with SoS' like McPherson and others, and knowing so well how a single person can game such a process.

SB 360 allows counties (like L.A. County) to develop their own voting systems and submit them for certification to the SoS. This is huge because it opens a major crack in the historical monopoly relationship between vendors and their captive county customers.

They could have done that before. Had the original February version of SB 360 been passed --- freeing up more money for development, but requiring any county-developed system to be federally certified, as per long-standing CA law --- we wouldn't be having this discussion.

Again, I have been as loud and critical of the corporate control of voting systems in this country as anybody in it. Yet, there are many ways around it under pre-SB 360 law. As you know, LA County already owns its own system and rewrites the code for that system regularly. Nothing keeps them from developing their own system, using a disclosed source system or an open source system or, for example, doing what Oklahoma did years ago --- essentially inviting ES&S in, buying their software and off-the-shelf hardware, and then kicking ES&S out of the state. As the largest voting jurisdiction in the nation, LA County easily offers the spending power to get such a deal from a vendor who would be delighted to have the contract.

Suggesting that SB 360 is needed to "crack the historical monopoly relationship between" vendors and customers seems misleading, at best, when used as a reason to support SB 360.

When its presented as necessary and the only reason for the bill, as state Sen. Alex Padilla (author/sponsor of the bill) has dishonestly and deceptively done since the beginning, it's far worse than simply "misleading".

SB 360 does not mandate nonproprietary or disclosed source voting systems. But it does explicitly ALLOW jurisdictions to adopt them.

Is there something in pre-SB 360 law that DISallows that? If so, I believe I'm unaware of it, but happy to admit I'm wrong if you're able to point me to the evidence for it.

Now as I said, there are imperfections in any comprehensive new law. But in this case I think you are failing to even mention the major positive themes in SB 360 that make it a genuine broad reform and major improvement in the security outlook for California elections. Go ahead and point to imperfections as you see them, but I think a more balanced reporting acknowledging the positive features of the would be helpful.

As you may have guessed, my feeling is that the dangers far outweigh the perceived positives you cite. I have noted, in my coverage, the ostensible reasons for the original legislation (that LA County needed the law to free up millions in federal HAVA and state Prop 41 money for use in development of their new, 100% unverifiable touch-screen voting system.) I have interviewed Dean Logan, live on air, to allow him to explain the system and his interest in SB 360, and quoted his advocacy for the bill at length.

I have quoted the advocacy for the bill by state Sen. Alex Padilla (author of the bill and leading candidate for SoS). I have invited him several times to respond to questions and/or address them in interviews. He declined to respond to any of those invitations. If more supporters of the bill had been willing to go on record in support of it, I suspect I would have quoted them as well. But, they weren't. Almost everyone I spoke to about the bill concurred with my assessment of it, even as they had other reasons to support the bill anyway, but they were not willing to go on record to that end.

That said, as far as I can tell, I am the ONLY person in the state to have done ANY real reporting on this bill, its text, its potential effects, its dangers, and its reason for being introduced. There has been NO OTHER REPORTING by anyone in either the state or nation about it, to my knowledge. Other than my own reporting, there has been only advocacy for it elsewhere --- cut and paste from press releases in support, essentially --- at least until the Elias column I cite above.

I'm sure my series of stories on the blog and discussion on air did not hit every aspect, but while there have been plenty of supporters, there has been no one, for the most part, other than me, willing to report publicly on the serious concerns in the bill. And, of course, that's my job as a journalist.

Beyond that, of course, I always welcome your thoughts and input, and very much appreciate your putting your opinions in support of the bill on record here so folks can have a broader understanding of both its dangers, as well the reasons folks like you and Verified Voting, etc. supported it (which I also noted in my coverage previously.)