Editorial endorsing Rep. Holt's Election Bill Misleads, Gets Facts Wrong...
By Brad Friedman on 6/23/2009, 3:28pm PT  

The New York Times just doesn't get it. You'd think, by now, they would. But they don't. And they should print a correction immediately.

In a brief, unbylined editorial yesterday, headlined "How to Trust Electronic Voting," the paper endorses this year's version of Rep. Rush Holt's election reform bill (H.R. 2894). The editorial is misleading and, even worse, blatantly (and inexcusably) inaccurate on at least one important point...

The first graf begins thusly:

Electronic voting machines that do not produce a paper record of every vote cast cannot be trusted. In 2008, more than one-third of the states, including New Jersey and Texas, still did not require all votes to be recorded on paper. Representative Rush Holt has introduced a good bill that would ban paperless electronic voting in all federal elections.

No need to read the piece much further, frankly. The reference to "paperless electronic voting," aside from being so very 2005, is also incredibly misleading. The idea that "electronic voting machines that...produce a paper record" are somehow more trustworthy than those that don't, has long ago been discredited by computer scientists and security experts. For just one such example, see this video produced by the UC Santa Barbara's Computer Security Group for CA SoS Debra Bowen's landmark "Top-to-Bottom Review" of electronic voting systems. The scientists demonstrate how to hack a Sequoia touch-screen voting machine --- one that produces "a paper record" --- in seconds. The same technique can be applied to virtually any other Direct Recording Electronic (DRE, usually touch-screen) voting system.

"Paper records" printed by such a voting machine are no more trustworthy or verifiable than votes stored on similar systems which don't print "paper records."

A similar system mis-printed 4 out of 12 of my own votes when I voted in California's June 2008 state primary. Luckily, I was able to notice the failure after very careful inspection of the printed ballot. But given that an MIT/Caltech study found that some 80% of voters don't check their computer printed "paper records," how many wouldn't have noticed --- especially as the system I was using was intended for the blind! And, even if all voters had noticed such failures, how would post-election auditors be able to know for certain that what they were auditing was actually the voters' intent? Answer: they wouldn't and couldn't.

Unfortunately, Holt's bill, while it offers several needed improvements to our electoral system, fails to ban DRE machines until 2016, and would allow the type of machine that misprinted my votes to be used forever. Nonetheless, the NYTimes goes on to write:

Mr. Holt's bill would require paper ballots to be used for every vote cast in November 2010.

They are entirely wrong and should publish a correction on that point immediately. The bill would only require paper records (the hackable type, as seen in the video noted above) by 2010, not "paper ballots" as the Times incorrectly argues. If passed as currently written, it would allow for the use of wholly unverifiable DRE machines for three two more general elections, including the next Presidential election in 2012. Paper ballots would not be required until 2016 2014 [see correction at end of article], and even then, they could be printed by the same type of systems that misprinted by own ballot, as opposed to far more reliable hand-marked ballots.

There remain other problems in Holt's bill (for example, it federally institutionalizes the protection of proprietary trade secret software for use in public elections), even as it offers other measures of much needed reform. See The BRAD BLOG's most recent analysis of the bill here.

There are legitimate reasons that the "paper of record" could have found to support its endorsement of the bill. There are no legitimate reasons why they'd be so recklessly inaccurate, misleading, and irresponsible when editorializing about it.


For a broader and more fervent rebuttal to both the Times editorial and the "concealed vote counting" that would be allowed forever vis a vis optical-scan systems in Holt's bill (that's a point we don't get into in the comments above), see New Hampshire election integrity advocate Nancy Tobi's take right here...

CORRECTION 6/24/09: As VotersUnite.org's Ellen Theisen points out in comments below, the Holt bill, as introduced, changed the year when a paper ballot would be required for all votes cast from from 2016, as it had been in their "final" version before introduction, to 2014. The final version of the bill, as introduced, is posted here. We regret the error in our original piece, and have noted the correction above.

UPDATE 7/1/09: A week and a half later, and the Times has still failed to publish a correction. And now we learn why they might have made the "error" in the first place. Holt is making the same erroneous claims about the requirement for paper ballots as of 2010! Details here...