SPECIAL REPORT-AUTOMATED ELECTION First of a Series- May 4, 2010
SPECIAL REPORT-AUTOMATED ELECTION
First of a Series- May 4, 2010
By Ike Seneres
An anonymous caller alerted me last night that in the testing and sealing procedure conducted by the COMELEC in Makati City as part of the installation of the automation project, the results were unusually slanted in favor one mayoral candidate in the optical count, but the results were different in the manual count, recording more votes for the other mayoral candidates. Clearly it was a signal that something was wrong, so I decided to go to one of the Makati precincts to find out what was going on.
I arrived at the Ramon Magsaysay High School in Barangay Olympia just in time to interview Mr. Neal Aquino, who introduced himself as a technician working for Smartmatic. He explained that the problem might have been caused by the thin paper stock that was used in the testing, suggesting that the paper stock used was not the same as the stock used for the actual ballots. I wondered about the explanation of Mr. Aquino, since I could not see a connection between the thickness of the paper and the accuracy of the scanning process.
From Olympia, I proceeded to interview Councilor Junjun Binay, one of the mayoral candidates in Makati. He showed me video footages of the actual testing, and in the videos I saw the differences between the results of the manual count and the optical count. The video also showed footages of Board of Election Inspectors (BEI) who appeared to be bewildered about what happened.
On the telephone, I interviewed Mr. Lito Averia, an IT expert who has been monitoring the COMELEC automation from the start of the bidding until now. He explained to me that under COMELEC Resolution 8785, Smartmatic is required to conduct mock elections in every precinct, using ten random voters who are present at the testing site. This is the same testing and sealing procedure that was conducted by the COMELEC in several locations. Apparently, the purpose is to test first the Precinct Counting Optical Scanning (PCOS) Machine, before it is sealed for future use on May 10, 2010.
This morning, I received reports from other IT experts who said that the COMELEC has already explained that the discrepancy in the counting was caused by the ink bleeding over to the other side, thus disabling the capability of the scanner to read the marks on the other side. Mr. Leo Quirubin, an IT expert refuted this argument, saying that if indeed there was bleeding of the ink, then it should have also disabled the capability of the scanner to read the marks at the other side.
Going back to the explanation of Mr. Aquino, it would not seem right for COMELEC to use a thinner paper stock to test the scanner, because the point in an actual test is to find out whether the same machine could read the marks in an actual situation, meaning to say that it should have been an actual simulation using the same inputs.
Based on the interviews that I have conducted so far, and based on what I saw on the video footages, I am now under the impression that some political operators might have succeeded in inserting malicious codes in the PCOS machines either at the warehouse in Laguna, in transit or upon delivery at the precinct level. This observation leads me to speculate that the results that came out in the testing might have already been pre-programmed to favor certain candidates.
Under the bidding rules, the error rate of Smartmatic should not exceed 1% of the results. By its own admission, the COMELEC has already admitted that the error rate could exceed 30%. Obet Verzola, another IT expert has predicted that the error rate could exceed 75%. Based on what happened yesterday, it would appear that Verzola is right, and that Smartmatic has already violated the terms of the supply contract.
In text messages that I sent to other IT experts, I suggested that we now change the terminology from “automated counting” to “optical counting” because I think that this would be a more accurate description. After all, the counting procedure using the PCOS is not actually automated, because it is manually fed.
In a private meeting that I held with some election lawyers, I suggested that at this point, candidates could already demand for manual counting as a logical sequel to manual voting. I argued that in the first place, the voting process as it is now is not really automated, because it is paper based, and it is manually marked by the voters. There is now a reason to set aside “optical counting” because the machines already appear to be corrupted.
Copyright © 2010, publication by newspapers is authorized.
First of a Series- May 4, 2010
By Ike Seneres
An anonymous caller alerted me last night that in the testing and sealing procedure conducted by the COMELEC in Makati City as part of the installation of the automation project, the results were unusually slanted in favor one mayoral candidate in the optical count, but the results were different in the manual count, recording more votes for the other mayoral candidates. Clearly it was a signal that something was wrong, so I decided to go to one of the Makati precincts to find out what was going on.
I arrived at the Ramon Magsaysay High School in Barangay Olympia just in time to interview Mr. Neal Aquino, who introduced himself as a technician working for Smartmatic. He explained that the problem might have been caused by the thin paper stock that was used in the testing, suggesting that the paper stock used was not the same as the stock used for the actual ballots. I wondered about the explanation of Mr. Aquino, since I could not see a connection between the thickness of the paper and the accuracy of the scanning process.
From Olympia, I proceeded to interview Councilor Junjun Binay, one of the mayoral candidates in Makati. He showed me video footages of the actual testing, and in the videos I saw the differences between the results of the manual count and the optical count. The video also showed footages of Board of Election Inspectors (BEI) who appeared to be bewildered about what happened.
On the telephone, I interviewed Mr. Lito Averia, an IT expert who has been monitoring the COMELEC automation from the start of the bidding until now. He explained to me that under COMELEC Resolution 8785, Smartmatic is required to conduct mock elections in every precinct, using ten random voters who are present at the testing site. This is the same testing and sealing procedure that was conducted by the COMELEC in several locations. Apparently, the purpose is to test first the Precinct Counting Optical Scanning (PCOS) Machine, before it is sealed for future use on May 10, 2010.
This morning, I received reports from other IT experts who said that the COMELEC has already explained that the discrepancy in the counting was caused by the ink bleeding over to the other side, thus disabling the capability of the scanner to read the marks on the other side. Mr. Leo Quirubin, an IT expert refuted this argument, saying that if indeed there was bleeding of the ink, then it should have also disabled the capability of the scanner to read the marks at the other side.
Going back to the explanation of Mr. Aquino, it would not seem right for COMELEC to use a thinner paper stock to test the scanner, because the point in an actual test is to find out whether the same machine could read the marks in an actual situation, meaning to say that it should have been an actual simulation using the same inputs.
Based on the interviews that I have conducted so far, and based on what I saw on the video footages, I am now under the impression that some political operators might have succeeded in inserting malicious codes in the PCOS machines either at the warehouse in Laguna, in transit or upon delivery at the precinct level. This observation leads me to speculate that the results that came out in the testing might have already been pre-programmed to favor certain candidates.
Under the bidding rules, the error rate of Smartmatic should not exceed 1% of the results. By its own admission, the COMELEC has already admitted that the error rate could exceed 30%. Obet Verzola, another IT expert has predicted that the error rate could exceed 75%. Based on what happened yesterday, it would appear that Verzola is right, and that Smartmatic has already violated the terms of the supply contract.
In text messages that I sent to other IT experts, I suggested that we now change the terminology from “automated counting” to “optical counting” because I think that this would be a more accurate description. After all, the counting procedure using the PCOS is not actually automated, because it is manually fed.
In a private meeting that I held with some election lawyers, I suggested that at this point, candidates could already demand for manual counting as a logical sequel to manual voting. I argued that in the first place, the voting process as it is now is not really automated, because it is paper based, and it is manually marked by the voters. There is now a reason to set aside “optical counting” because the machines already appear to be corrupted.
Copyright © 2010, publication by newspapers is authorized.
posted by Ike Señeres at 7:59 PM
0 Comments:
Post a Comment
<< Home