Why is Comelec still dealing with Smartmatic?

13

DURING the Joint Congressional Oversight Committee hearing on August 6, 2015, I read a statement which I am reproducing below, with some minor editing:

For our elections to be credible, three things must be present: 1) accuracy of the counting; 2) transparency of the process; and 3) speed. All three must be present.

How does Smartmatic’s PCOS fare with respect to these requirements?

Accuracy.
• No mock election conducted by Smartmatic has produced the required accuracy rate of 99.995% or better (1 error in 20,000 marks)


• The 2010 Random Manual Audit (RMA) resulted in only a 99.6% accuracy rate (80 errors in 20,000)

• The July 24-25, 2012 mock elections in Congress resulted in only a 97.215%accuracy rate (557 errors in 20,000)

• The 2013 RMA resulted in only a 98.86% accuracy rate (228 errors in 20,000)

• In 2010, 9% of the PCOS units failed to transmit the Election Returns (ERs) to the Transparency Server; in 2013, 23% of them failed.

• The huge differences between the PCOS and manual counts in the election cases in General Tinio, Nueva Ecija, Isulan, Sultan Kudarat, and Dinalupihan, Bataan are clear proofs of PCOS’ inaccurate counting and/or vulnerability to internal tampering. There surely would have been more had the protest process not been seriously impaired

• Smartmatic’s counts and canvassing results were so arbitrary. This was exemplified by the progress of Senator Poe’s total votes. In the May 18, 2013 report, with only 42% of the votes canvassed, she already garnered 20,147,423 votes. Extrapolating the total with the remaining 58% would have brought her total to more than the voter turn-out of 39 million. But instead, in the June 7 report, she only gained 189,904 votes. Some suspect that the results were massaged. And then, in the June 11 report, her total votes went down to 16,340,333. No explanation that I know of.

Transparency. Any system that automates precinct-counting loses transparency. And it follows therefore that all succeeding steps after precinct-counting would also be questionable. Loss of transparency is the common reason why Germany, Netherlands, Ireland, Australia, and 14 other countries out of the 30 that automated their elections, went back to manual precinct-counting.

Speed. If the Smartmatic system had run properly, the final results would have been released at half the time it took, especially in 2013.

Since Smartmatic’s PCOS machines are inaccurate and non-transparent, and only the speed requirement was somewhat met, why is Comelec still entertaining bids from this vendor? It just doesn’t make sense! And it would make our elections a sham.

Further proof of the unreliable Smartmatic system is the very recent fiasco (in June 2015) that occurred in Tabasco, Mexico, where their system failed miserably.

Should we allow Comelec to continue dealing with this vendor and putting the 2016 elections at risk? The Comelec recently awarded a contract to Smartmatic for the supply of 23,000 more PCOS machines and will bid out another 71,000, which, based on the existing liaison between Comelec and this vendor, will most likely be won again by Smartmatic. Comelec is digging a deeper hole for us that we will soon find difficulty extricating ourselves from.

Comelec cannot say that there are no other alternatives. There are two … and they are both Filipino-developed. If it still prefers an OMR (Optical Mark Recognition) system, there’s TAPAT, which, apart from being accurate, has the VVPAT feature (Voter Verifiable Paper Audit Trail), which PCOS does not have.

If it doesn’t mind a system that may take some 10 hours longer, there’s PATaS, which, apart from being accurate, does not only bring back transparency into the system, but enhances it even more. It includes a public website that is updated every 30 minutes, showing the progress of counting in each precinct. This website can be accessed by anybody anywhere in the world. I know of no system today that can match that.

And, by the way, it is not true that PATaS will cost P36 billion, as Comelec has been maliciously announcing. That was a work of fiction worthy of inclusion in Ripley’s Believe It or Not. It must have been computed by the same group, within Comelec, that came up with the cost of P700 million for ballot-secrecy folders for the 2010 elections.

Our estimate of the equipment cost of PATaS is only P4 billion and they can be donated to public schools after each election, thus saving on warehousing and maintenance expenses. This, plus the other expenses (bloated, still) that I picked up from the Comelec version, our total comes up to P13 billion, against PCOS’ P20 billion (plus P400 million a year for warehousing and P1.8 billion – or is it P3.1 billion? – for repairs.

It is also not true that it would take one year to develop the software. The Consolidation and Canvassing System (CCS) was developed in three months when I was still with the Comelec. It might need another one or two weeks for “stress” testing.

And for the nth time, I dare the Comelec lawyers and the pro-Smartmatic Congressmen to review R.A. 9369 and then show us what provision in that law is violated by PATaS.

As a Filipino saying goes, “Pag gusto, mayparaan; pag ayaw, may dahilan.” (If they want it, thereis a way. If they don’t, there’s always a reason.) Or how about this? “Mahirap daw gisingin ang nagtutulug-tulugan lamang.” (It’s said that Nothing is harder than to wake up someone who’s pretending to be asleep.

Share.
.
Loading...

Please follow our commenting guidelines.

13 Comments

  1. It’s a fact that having an automated election is by far faster and better. It’s a fact that it is far more accurate. However a computer/machine can only do what it is told to do.
    I’m sure most of the voters are concern are the trustworthiness of the people operating them. Being an automated election quite frankly speaking it is far easier to manipulate data, before you need to steal ballot boxes, now it’s just memory card.
    Do smartmatic uses file transfer checksum? Otherwise packet injection would be a possibility. And there are many other cases. At the end of the day it is down to the integrity of the people running the election.

  2. Nope, there is no need to estimate anything, a shade is a shade, reducing it from 50% to 20% means there was a problem in the first place. So you are actually implying that there were votes not counted in 2010 as they had to reduce the treshold to 20%, is that correct? If it was 100% correct in the first place, why change the treshold?

    How can you be so sure that what was in the ballot was correct in the first place? You cannot put a cost on the accurary of the election so justifying that printing a receipt is costly at the expense of the accuracy and confidence in the election is not a good justification.

    If the picture image are in the system, why was the recount not comparing also the picture images, would it have been correct for COMELEC to compare the 3, the actual ballots, the PCOS results, and the picture image? If you are not counting the picture image in the recount, what is the point in storing it? Where are you going to use it?

    • Obviously, you are ignorant of electoral protest procedures before the HRET and SET. The stored image of the paper ballot is actually the final basis of decision when compared to the paper ballots. There have been hundreds of protests filed before the HRET and SET and none of them ever prospered which leads to the conclusion that PCOS is accurate and there was no cheating.

  3. Kahit anong haba ng diskusyun, sa SMARTMATIC pa rin ang dating. Money talks…simple as that…

  4. Amnata Pundit on

    I suggest you challenge Smartmatic to go head to head with PATAS and TAPAT in a simulated election to be monitored on national TV. Lets see if Smartmatic will accept it. And can you please reply to Owen Sy’s claim that PATAS and TAPAT are vulnerable to hacking? That’s one thing that is nagging this computer illiterate. Thanks.

    • PATAS and TAPAT are only experimental prototypes. Without undergoing certification from an independent international certification body, it poses a lot of operational risks. Take the case of TAPAT which uses a run of the mill tablet, which it scan properly when there is a brownout, a really case scenario? It relies on proper lighting, so my guess is that it won’t. It is very easy to put together components of readily available technology to achieve a specific objective and call it an “invention” but it is another thing to make it function as it should do in real life conditions.

      Where did I say that they are vulnerable to hacking? But since you brought it up, yes it is vulnerable to hacking because it hs not been fully tested in real life elections and not even certified as industrial grade.

  5. Ang daming magagaling na IT Professionals sa atin kaya we don’t need SMARTMAATIK..but since kailangan ng extra income ng opisyales natin so ditto na tayo sa SMARTMAATIK :-)

  6. The government is going to rig the next election again to get whoever Aquino wants in there to pardon and save him and his friends.

    Everyone knows whats going to happen the only question is.
    Are the people going to allow it again ?

  7. Here we go again, Gus or Guz or whatever you fancy at any give day! You are a master of twisting facts to discredit a proven and successful automated electoral system just to get your way come hell or high water. You’ve been pushing one thing after another since the 1990s when it comes to automated elections, unfortunately for you and fortunately for the Filipino voters, you have been unsuccessful. I really find it odd that you, a self proclaimed “computer guru”, is really pushing hard to revert back to manual.

    As for your twisted facts, consider these:

    1. Accuracy – a random manual audit with never match computerized counting due to the simple fact that the manual part of the random audit is prone to human error. Case in point shading thresholds. In 2010 this was set to 50% and in 2013 I believe it was lowered to 20%. The computer software will accurately determine whether the oval was shaded less that the threshold (50 or 20% a the case may be) while a human doing a random audit will not, thus the discrepancy. Given these facts, the discrepancies are within acceptable levels and therefore concluded as highly accurate. But you purposely made a big deal out of it to mislead people and discredit the current provider for your self gain.

    2. Transmission – you know fully well that this is highly dependent on the telcos and prevailing conditions on any given day just like the internet connections. Fact is no telco in the Philippines have 100% coverage that is why Smartmatic provides BGAN satellite modems in far flung areas. There are also Comelec contingency plans that’s in place for occurrences of non transmittal of ERs for whatever reason , part of the whole systematic scheme and you know that. So my question to you is, what in your proposed alternatives will make transmission better when you know fully well that this component is not in or beyond your control? Again your twisted facts are self serving!

    3. Transparency – as a “systems guy” you know fully well that once you computerize or automate, the meaning of transparency changes. In manual, transparency is “you can see” while in automated it is ” you can prove”. And proof is achieved in several ways, international certification of systems software by an independent body, random manual audit, controlled source code review, voter acceptance post elections and protests mechanisms.

    Just a word on controlled source code review because I know you will pounced on the use of that term. It has to be controlled because these source code are protected by intellectual property and besides it is literally a map of its inner workings and will expose it to would be cheaters and hackers possible ways to penetrate it. Comelec will never give in to your demand to make copies of the source code for you to take home and review at your leisurely pace with or without malicious intentions.

    Lastly and with all due respect, please put the country above your self interests and personal motivations. Your actions are very disruptive to the process of trying to achieve a clean and honest elections because you chose to put your ego first rather that opt for the common good.

    • i don’t agree with your statement a random audit is less accurate as it is prone to error due to a manual part. it might be the other way around. Given the facts of the manual recounts from the different parts around the county and the PCOS machine giving less than accurate results, i would agree that the PCOS is dangerous for the Philippines. You might be on the side of the administration and say that Gus should do what is good for the common good but take note that if the same happens and it is perpetuated by the anti-admin group, i am sure you would also complain.
      Being also an IT guy, i don’t trust any systems that just tells me that your votes or any transactions has been received and process without any confirmation. You would want to deposit or pay to a bank and not get any confirmation receipts? How about later them telling you that they have no records of the transaction or that you only deposited 10% of what you actually gave them?

    • @Julius

      As in the case of determining whether the oval was shaded 20% to pass the threshold test, a computer can be programmed to determine this with 100% accuracy while a human eye can only estimate, therefore it is prone to errors and thus discrepancies. Very elementary, Watson!

      The Supreme Court has already ruled way back in 2010 that the paper ballot itself is equivalent to the receipt therefore no need to print a receipt because this is just redundancy that add to the cost and time of the voting process. Since the PCOS also takes and store a picture image of the paper ballot, any attempts to tamper the paper ballots post elections will not prosper because the SC also ruled that the picture image will be the final basis.

      BTW, I too am an IT guy and I don’t buy your arguments.