«

»

Aug 24

The Explainer: Automation as silver bullet

Ever since the birds and the bees, the flowers and the trees, not to mention mythical aswang, voted in the 1949 presidential elections, having clean and honest elections has been the holy grail of our democracy. In 1953 Magsaysay mambo’d to Malacanang promising clean elections. In 1986, as the nation protested the rigged Snap Elections, the Mambo Magsaysay made a come back.

Will 2010 be disreputable like 1949 and 1986 and even 2004, or will it be like 1953 and restore our faith in one man, one vote, and a presidency truly elected by the people? It may surprise you that the burden lies on the shoulders, not of the Comelec as you might think, but on the Supreme Court.

Tonight, we’ll look at the solution approved by Comelec for counting our votes; and the implications of the questions that have been raised before the Supreme Court. I’m Manolo Quezon. The Explainer.

 

I.

 

Courtesy of the Philippines Free Press, which turned 101 years old this month, let’s go back in time to early 70s, before martial law.

Back then, the Free Press was criticizing the headlong rush for Charter Change, because, as this editorial cartoon by the late E.Z. Izon pointed out, Charter Change was being approached as a kind of voodoo medicine to cure the nation’s ills.

And this brings us to this man, the late Raul S. Manglapus. He was a delegate to the 1971 Constitutional Convention.

In a speech he made in the Convention, Manglapus revealed three reasons why Congress had finally passed a Constitutional Convention law.

He said that by the 1960s, legislators wanted Charter Change because of three reasons. First, they were sick of presidents running for re-election and wanted to restore the single, six-year term for presidents; second, that the membership of the House should be increased; and third, for elections to be synchronized to save time and money.

This third reason is one that has found itself enshrined in law, but which has also made the computerization Garcia advocated way back in 1970 a logistical nightmare.

http://www.comelec.gov.ph/laws/republic_acts/ra_7166.html

With Republic Act 7166, passed in 1991, we have what are called synchronized elections.

http://www.comelec.gov.ph/laws/toc.html

This is the latest in efforts to establish the dream of politicians since the 1960s and even earlier, to just have all our elections at one time, supposedly to make things cheaper and more efficient in the long run.

The early 70s was a time of yearning for the cures to the nation’s many ills, and one such cure was proposed by another Constitutional Convention delegate, in fact, its first president, except he died soon after becoming Convention President -this man, former President Carlos P. Garcia.

In 1970 –that’s 39 years ago, and before the ConCon was convened- he delivered one of the Gregorio Araneta Memorial Lectures and proposed computerizing elections. Let me read you three brief paragraphs from his lecture.

First, he posed a practical problem. He asked,

How can we minimize… fraud? The answer is definitely, mechanize and electronify voting. We should use the voting machine which [is] as simple as an adding machine or we should use the different electronic devices and gadgets for voting and totaling as we go along with the voting. We should use computers for fast and accurate computations of votes in cities, provinces, and regions.

He viewed computerization –or, to be precise, he called it electronification- in terms of the hole-punching system then practiced in the United States. As former President Garcia described it,

By the mechanization and electronification of voting, all that voters will do is punch on the single line ballot the name of his choice… The taxing and long-winded discussions between election inspectors or watchers on whether to to count badly-written votes will be a thing of the past.

But Garcia went beyond how ballots would be filled out, or punched with holes, and discussed the counting and how machines might remove human error:

The preparation  of election returns should be done publicly by simply exposing to the public the figures of the automatic totalizer. It should be opened to cameras, telecameras, television, telephotos, taperecorders, radio and to all electronic recording machines so the public will have immediate knowledge of election results ten minutes after the last voter voted. Computers strategically located will get accurate totals of results in cities, provinces, and even regions…

Such was his dream. Such was his analysis. And, in a basic sense, he laid out, in 1970, the case for automating elections: first, to eliminate the old means of electoral fraud; second, to do so by making ballots countable, not by people, but by impersonal machines; and third, by doing so, in a manner understandable, and verifiable, by the public, to make the results of elections known in a timely manner.

In 1986, for the Snap Elections, computers reentered the political scene, when the Comelec tried to use computers to tabulate the results. We know how that ended up: the computer operators, instructed to cheat for Marcos, walked out of the counting going on in the PICC.

It’s taken twenty years to make the idea of using computers in counting ballots even imaginable again. But things have changed a lot from 1970, when Garcia dreamed of automation, or even 1986, when the Comelec discredited computers counting votes.

Among other things, the reputation of the Comelec, rehabilitated in the 1990s, is back to where it was during the Marcos years.

And internationally, voting machines have become controversial even in countries not previously known for dagdag-bawas, or vote padding and vote shaving.

If you remember the controversies surrounding the 1950s era voting machines in Florida in the 2000 US presidential elections, voters punched holes in ballots and when the holes weren’t punched properly, a recount became a mess.

The 2000 elections, in a sense, stampeded the Americans into adopting more hi-tech machines for the counting of ballots, but some of these machines proved even more controversial.

http://www.rollingstone.com/news/story/10432334/was_the_2004_election_stolen

 

Back in 2004, Robert F. Kennedy Jr. wrote a controversial article in Rolling Stone Magazine, where he alleged that the Republicans did two things to swing the election in Bush’s favor.

1. Purge the list of registered voters, so those likely to vote Democrat  couldn’t vote. That’s the bawas.

2. Count Kerry votes as Bush votes. And that’s the yung dagdag, yung dagdag, portion.

http://www.washingtonpost.com/wp-dyn/content/graphic/2006/03/16/GR2006031600213.html

In 2006, the Washington Post pointed out that your average electronic Las Vegas slot machine was actually more reliable, and more trustworthy, than many electronic voting machines, particularly the kind that tried to eliminate the paper ballot.

 

They identified five main weaknesses of electronic voting machines at the time:

1. Software was a trade secret.

2. No checks of the computer chips used was required, and spot checks made useless because inspectors lacked samples to compare the computer chips to.

3. The public had no way of finding out if programmers had been convicted of fraud.

4. Equipment certification was done by for-profit companies, not independent bodies, and the public wasn’t informed of how testing was done.

5. Complaints, if filed, may nor may not be investigated, once citizens complained by calling a number that may or may not work.

 

Critics of electronic voting machines therefore often point to the United States as an example of computerization actually diminishing, and not enhancing, the credibility of elections.

http://www.dw-world.de/dw/article/0,,4069101,00.html

Organizations like CenPeg, for example, of which you’ll hear more later, point to Germany, which is Federal country, and thus, like the United States, has states supervising elections, as recently having disallowed voting machines, when its Federal Supreme Court insisting citizens need to be able to see the counting of their ballots; Holland, too, in 2008, abandoned paperless voting systems.

Critics of automated and in particular, paperless balloting also point to to Japan, which, like Germany, is under the parliamentary system, that still uses a manual system –just like France, which is unitary and presidential, like us, and which does its elections manually, too.

So if there’s one emerging consensus, it’s that there’s no substitute for a paper trail, even when machines are involved.

http://www.nytimes.com/2009/06/22/opinion/22mon2.html

On June 21, the New York Times published an editorial titled “How to trust electronic voting.”

It asserted that,

Electronic voting machines that do not produce a paper record of every vote cast cannot be trusted.

And that a system in which voters directly record their votes on paper is the best system of all.

The machines and the questions they’ve raised, is what we’ll tackle when we return.

 

II.

 

That was a satirical look by ScottRose.tv of how a computerized voting machine company, Diebold, allegedly swung the 2004 elections in George W. Bush’s favor in 2004. Critics of the automation approved by the Comelec, fear a similar situation in 2010.

In last year’s ARMM elections, a company called Smartmatic participated by providing a set of machines.

Here’s a video:

http://www.youtube.com/watch?v=xI5ZbxGXHx0

 

Notice that the system Smartmatic tried out in the ARMM was essentially paperless, making it very similar to the Diebold machines criticized in the USA. Voters cast their votes by pressing buttons and these choices were then recorded in the machine. But noticeably absent is a paper ballot.

When the Comelec finally decided on Smartmatic and its local partner, TMI, the system they put forward relies on a paper ballot, which is then scanned.

Smartmatic will provide the Comelec with Precinct Count Optical Scan (PCOS) machines…

And the software required to tabulate the contents of the ballots that those machines capture.

Here’s how it will supposedly work, based on Smartmatic’s own website.

Pretty straightforward: shade in circles, like your standard scholastic tests, feed it in the machine, and go home.

The machine in your precint will produce eight copies of the Election Rreturn, that are printed and transmitted according to RA-9369.

 

But let’s focus on two aspects, which emerged a demonstration to the editors and reporters of the Philippine Daily Inquirer last July.

First, the way the machine will actually try to count your ballot.

http://www.youtube.com/watch?v=EMV3KoU3RTQ

 

It will only count the positions for which you voted, and it won’t count positions for which you either didn’t vote at all, or over-voted.

 

This is all about the security controls on who can operate the machines.

According to Pablo Manalastas’ analysis of this system, what you saw is where the BEI insert their CF cards (or, Manalastas suggests, it could be USB thumbdrives) into the PCOS machines, so that they can digitally sign the ER with their secret keys in their CF cards. When the BEI digitally sign the ER, their digital signatures will prove that the ER came from a valid source, and so that the data in the ER cannot be changed without being detected.

The BEI will then attach a GSM/3G/HSPA cellular modem to the PCOS machine, and then the ER is transmitted to the provincial and national computers for canvassing at those levels.

Finally, the ballot images, audit and system logs, and ER are all saved into a 1GB CF card.  Manalastas says that ideally, all political parties should get a copy of this CF card which contains everything about the precinct.

Up to an estimated 50 million voters will be trying this new system in May, 2010.

Where there used to be 250,000 precincts nationwide, the number will be reduced to 80,000 clustered precincts with each cluster composed of five precincts with a maximum of 1,000 voters each. Each clustered precinct will have a machine.

In the past, precincts were open 8 hours on election day. Then they’d close, and precincts would tally their votes within anywhere from 5 to 24 hours. Then, the canvassing would begin, with three stages: the city or municipality, the province, and then, the national canvass. This would take up to 40 days.

Under the new system, the winners of municipal, city, and provincial positions will be proclaimed on election day itself, May 11, 2010.

National positions –party-list representatives and senators, to be proclaimed by the Comelec, and Congress to canvass and proclaim the Vice-President and President- will all be proclaimed between May 12 and 13, 2010.

A minimum of 25 and maximum of 40 days of electoral drama, beginning with the frantic and exhausting manual counting of votes on the precinct level, would therefore be reduced to three days.

automated election system (or AES).

So you’d think we’d all be celebrating. Instead, the whole scheme’s been questioned before the Supreme Court.

It has to do with lawyer Harry Roque of the Concerned Citizens Movement, took the Comelec to court.

Roque believes that the government’s own bidding regulations were ignored by the Comelec, and that Comelec hasn’t complied with the requirement to pilot test the Smartmatic system before rolling it out nationwide. Furthermore, the Comelec, according to Roque, has illegally outsourced managing the elections to Smartmatic.

And there’s the Center for People Empowerment in Governance ( orenPeg). A Marxist-oriented UP-based think tank that’s been studying the AES in partnership with the UP College of Law. CenPeg opposes the Smartmatic system on the following grounds:

The Precinct Count Optical Scan-Optical Mark Reader (PCOS-OMR) technology chosen by the Comelec goes against the basic democratic principle of “secret voting and public counting.” This is because the OMR system makes the counting, canvassing and consolidation of election results hidden from public eye and, hence, lacks any transparency as the Constitution and RA 9369 require. The proclamation of winners will be done in 2-3 days making it extremely impossible to file any election protest which is expected to be widespread – and poll watching almost futile.

CenPeg claims it has identified at least 30 plus vulnerable spots in the AES.

CenPEG argues that these vulnerabilities exist throughout the whole system: from ballot printing, to the warehousing of the counting machine4s to hardware and software deficiencies, and potential flaws in the manner of voting, counting, and the electronic transmission of votes to canvassing and proclamation of winners.

Not to mention potential weak spots in the infrastructure system such as telecommunications, phone and electric lines, and cell sites.

An interesting account of how the Supreme Court is going about hearing Roque’s case, can be found in Pablo Manalastas’ blog:

http://pmana.multiply.com/journal/item/68/Harry_Roque_vs_COMELEC-Smartmatic

Basically, from Manalastas’ observations at the hearing last August 7, it seems the Supreme Court’s evenly divided, 6-6, right now.

The debate between advocates of the Smartmatic system and its critics has become fierce and highly technical. It’s impossible to summarize them here so let me point you in the direction of three specific places online to visit:

First, there’s the blog of our very own Dean Jorge Bocobo, philippinecommentary.blogspot.com:

http://philippinecommentary.blogspot.com/2009/08/automation-debate-between-cenpeg-and.html

Where he tries to debunk the critics.

For the critics, start with, of course, CenPeg itself:

http://www.cenpeg.org/

That way, you can get a sense of the organization’s general political leanings and positions, as well as its specific criticisms of automation.

Then visit the Multiply site and blog of Pablo Manalastas:

http://pmana.multiply.com/journal

In particular, see his July 8 entry:

http://pmana.multiply.com/journal/item/53/James_Jimenez_Reacts_to_CenPEGs_list_of_Vulnerabilities_of_AES_2010_Jul_08_09

In which he responds to Comelec Spokesman James Jimenezes’ response to CenPeg;

And his July 11 entry,

http://pmana.multiply.com/journal/item/54/Comments_on_Dean_Jorge_Bocobos_Reactions_to_CenPEG_July_12_09

In which he responds to our very own Dean Jorge Bocobo’s criticisms of CenPeg.

http://filipinovoices.com/trading-the-ballot-box-for-a-black-box

And for something in between, visit FilipinoVoices.com, in particular this August 15 entry, titled”trading the ballot box for a black box.”

But regardless of the legal merits and technical aspects, the political reality is this.

On June 30, 2010, the terms of the President, Vice-President, Senate President, Speaker, indeed, of half the Senate, the entire House of Representatives, all local government officials from governor down to councilor, would end.

If automation fails, we may not have any or some of these officials proclaimed; if, to prevent the possibility automation fails, automation is cancelled entirely, then we run another risk –of a repeat of the 2004 elections.

This, in a very real sense, is the great burden placed on the Supreme Court’s shoulders, even if it is only supposed to be deciding questions of contracts and provisions of law.

So later tonight, join us as those supporting Smartmatic and who question it’s Comelec contract, debate whether the country can risk automating –or not automating.

I’m Manolo Quezon. This has been the Explainer.

 

 

 

 

 

 

Leave a Reply