Archive for the ‘Automation’ Category

July 24, 2008

For many people, it’s easy to reduce the question of ARMM elections to a simple yes-or-no proposition. Its not that simple.

Let’s face it: the proposal to postpone the elections has generated enough knee-jerk reactions harping on the same theme: postponement is a blow to democracy. The truth is, there appear to be enough valid grounds to support a postponement; you just have to be able to look at the issue from a different perspective than the promotion of representative democracy. Look at it, for instance, from the point of view of the peace process, and postponement makes some kind of sense.

This multiplicity of perspectives – and the patina of validity each one has – makes this a very difficult question to resolve. Worse, it becomes easy even for people with absolutely no sense of perspective to pick a stand and pretend to know what they’re talking about. Some don’t even bother to pretend; content in the knowledge that they are parroting some official line – pro or anti-.

All this, of course, does a grave injustice to everyone who is truly concerned with the elections in the ARMM. More bluntly stated, it would be far better if the people who get to decide whether the elections are a go or not were to actually ask the people who – despite not having the power to decide the question either way – are deeply invested.

COMELEC for instance, is about as deeply invested in this as possible.

First, it has already spent millions of pesos on the automation of the electoral process. Considering that the automated election system about to be rolled out on the 11th of August is leased, this might not be much of a problem, considering that the suppliers can probably just stop what they are doing and then simply re-start ops when a new election date has been settled on. Unfortunately, the talk going around has it that the postponement might be all the way up to 2010. It that happens, one is left wondering whether the leased machines units will still be ideal, or even whether the leased system is still relevant. At that point, we will have to grapple with the reality that we would have wasted hundreds of millions of taxpayer money for a project that would have borne great fruit had it not been cut down by political considerations.

Or, if as other whispers suggest, the postponement is only for a period of 90 days or less – what is the postponement for in the first place? And besides, additional costs will undoubtedly accrue due to the delay. Where will the money for  that come?

Read more?



Bid Bulletin No. 5
March 5, 2008

The COMELEC yesterday released Bid Bulletin No. 5, relative to the bidding for the Automation of the 11 August 2008 elections in the Autonomous Region in Muslim Mindanao.

Bid Bulletin No. 5, signed by COMELEC Executive Director Jose M. Tolentino Jr., amends those portions of the Request for Proposal which require that the automated election system (AES) to be procured by the COMELEC should have been successfully used in a prior public political election. In order to be considered in the bidding, an offered AES now only needs to have been successfully used in a prior election.

The requirement of successful use in a prior public political election was objected to by Filipino technology proponents who pointed out that local technology has never been used in any public political election. This, they said, would make the questioned requirement discriminatory as it would effectively disqualify local companies.

In authorizing the amendment, the Commission en banc acknowledged that many non-public and non-political elections – such as elections held by private organizations – have been successfully conducted by electronic means.

“Considering the widespread use of electronic voting systems in various non-political elections throughout the country, as well as the general acceptability of the results of those elections, it is abundantly clear to the Commission en banc that prior successful use in a public political election need not be the only applicable standard.”

Read Bid Bulletin No. 5 here.

February 20, 2008

Unfounded suspicions

It is remarkable how quickly some quarters move to discredit the work being done on automation. Less than a week after the release of the bid documents for the Automation of the ARMM elections in August, some “local IT companies” have gone straight to their favorite columnists – instead of properly bringing up the issue directly with the COMELEC  right away or at the scheduled pre-bid conference – raising the spectre of irregularities simply because there is a provision requiring that the automated election system being peddled should have a track record. According to these companies, this effectively shuts out potential local participants.

The problem with this approach is that it gives the pundits a one-sided appreciation of the issue, which is then published or broadcast in the guise of giving the COMELEC the opportunity to respond. Well, why not give the COMELEC the opportunity to respond before slamming the project? (Isn’t that how journalistic objectivity used to be practiced?) If the official explanation remains unsatisfactory then publish the complaint. That way, there is at least the semblance of fairness. Otherwise, airing one-sided complaints only give rise to unwarranted suspicions about the project.

Fortunately, the Advisory Council’s (and PPCRV’s) Amb. Henrietta de Villa was on talk radio this morning confirming that the provision (A-13 in the bid documents) was supported by the Advisory Council as an important safeguard.

My opinion

Now for my own take on the specific issue of whether the track record provision should be used.

The complaint basically points out that local IT companies are effectively prevented from participating in the project. Integral to this complaint – and I believe the only substantial claim that can be derived anyway – is the issue of whether Filipino tech should be promoted.

Well, of course it should. But for the elections? Local IT companies trumpet their “innovations and inventions” which are supposedly suited for the “peculiar needs of Philippine elections.” The first question that comes to mind is: how sure are we that those innovations and inventions will actually work as advertised? Remember Namfrel’s much bally-hooed SMS-based reporting system? That was touted as a home-grown innovation as well and it crashed and burned nevertheless.

The point simply is this: do we want Philippine elections to be the guinea pig for these untested systems? While it is true that Filipino ideas and innovations should be supported, I doubt that we should do so at the risk of the electoral exercise. The COMELEC would be amiss in its mandate if it were to allow the use of technology that it knows is untested and carries with it a higher probability of failure.

It is true that Filipino developers must be helped; but not at the expense of elections.

About the Philippines being a unique electoral environment: true enough. Does this mean, however, that Filipinos have a monopoly of knowledge on how to deal with that unique environment? Technology is not about nationality.

Examples have been bruited about to bolster the complaints of these local IT companies:

Example 1: The ARMM elections in 1998 used machines that did have a track record. They did not work.  

False. The machines worked. The problem was with the ballot paper which was trimmed to the wrong size by an employee who had no training, much less authority, to do so. The machines themselves counted correctly.

Example 2: The ACMs for the 2004 elections. The argument goes if the South Koreans had imposed a “track record” requirement on their developers, they would never have built their own ACM.

Example 3: The Indian Electronic Voting Machine (EVM). Same argument; if the Indians had the same prohibition … so on and so forth.

Not quite the same thing. In South Korea and India – especially in India – they have a multitude of small elections nearly every year. We have one BIG election every three years. The stakes are much higher in this country because one failed election affects major elective offices like Senator and Member of the House of Representatives. In 2010, the stakes are even higher: President and Vice-President. Will the nationality of the technology used be any comfort when we fail to elect a President because we used an untested system?

In places like India, these machines were first rolled out in small elections, affecting small constituencies. This allowed them to field test the equipment with very limited adverse effects in case of failure.

So, does the track record guarantee success?

Of course not. What the track-record requirement does is it lowers the probability of failure.

When you get a system that has been used in other elections, what you get is a system that necessarily includes experience with unforeseen or unforeseeable  problems encountered in actual operation. When you get a virgin system, on the other hand, all you have are theoretical scenarios which do not necessarily cover 100% of all the things that might go wrong. If you have an untested system in place and an unforeseen problem crops up, then the support team for the system will not have a well-defined procedure for dealing with problem and they will likely be improvising as they go along. Again, you want to entrust the success of major elections to a bunch of people who, in essence, are just winging it? They may be geniuses – which they probably are in all fairness – but the stakes are simply too high.

Would you leave a nuclear power plant in the hands of a child prodigy?

In any case, the COMELEC reassures the public that this provision will be discussed fully during the pre-bid conference scheduled for the 26th of February (that’s tuesday next week). Everyone who thinks the track-record idea is wrong can go there and tell the COMELEC so. Of course, since this provision was already extensively discussed before it was even included in the bid docs, people should not expect the COMELEC to simply fold and take the provision out. Expect to be argued against. After all, it is through argument that we can ensure that the ensuing decision has the highest probability of being the right one.

Automate ARMM!
February 9, 2008

The Commission on Elections en banc has ordered the full automation of the 2008 elections in the Autonomous Region in Muslim Mindanao. COMELEC Resolution No. 8415, dated 06 February 2008, called for the use of automated election systems based on two different kinds of technology: Direct Recording Electronic (DRE) technology in the province of Maguindanao and Optical Mark Reader (OMR) technology for the rest of the region. DRE allows voting through a touch-screen or touch-pad, while OMR requires voters to fill up a paper ballot which is then counted with a specially designed machine.

“The use of different kinds of automated election systems will allow the COMELEC to evaluate the suitability of the two different technologies of the Philippine setting, preparatory to designing an Automated Election System for nation-wide use in 2010,” COMELEC spokesman James Jimenez Said. “This set-up closely adheres to the recommendation of the Advisory Council.”

The only difference, according to Jimenez, was in the scope of the use of DRE. “The Council’s recommendation called for the use of DRE-based automated election system in only two cities or municipalities in the ARMM.”

The COMELEC resolution, however, cited concerns regarding the small scope. “Among other things, it was pointed out that if the implementation area is too small, there won’t be enough basis to say whether a DRE based automated elections will work nationwide,” Jimenez clarified. Other reasons cited by the Resolution included the concern that the small scope of the project would discourage technology proponents from bidding for it, and that the results would not be enough to make a comparison between OMR and DRE technology.

The resolution expanded the scope of DRE use to include the entire province of Maguindanao, citing the province’s level of development, as well as the contiguous nature of its geography. “The en banc determined that, comparatively speaking, the province of Maguindanao is the most viable implementation area for DRE-based automated elections,” Jimenez said.

The Advisory Council is the body of expert advisers created by Republic Act 9369 that is tasked to recommend to the COMELEC which technology to use for the automation of elections. Republic Act 9369 also states that “nothing in the role of the Council or any outside intervention or influence shall be construed as an abdication or diminution of the Commission’s authority and responsibility for the effective development, management, and implementation on the Automated Election System.”


Couldn’t post this last night, but it’s been up on eCOMELEC since yesterday. Other mentions here, here, and here

Automation 2008
January 20, 2008

A quick update on automation: Last monday, 14 January 2008, the COMELEC received a draft terms of reference and request for proposal (TOR/RFP) from the Advisory Council. The document was vetted by the AC and found to be consistent with its recommendations – that the ARMM polls be automated; and that the solutions offered could be direct recording electronic (DRE) or based on optical mark reader (OMR) technology. The TOR/RFP has since been reviewed by the COMELEC for consistency with the governing law (RA 9369) and existing policy – mostly to do with period for doing things, especially in relation to the timetable for counting and canvassing.

Last friday, 18 January, the reviewed TOR/RFP was returned to the AC and the Technical Evaluation Committee (TEC) for its consideration of the suggestions made by the COMELEC senior staff. We expect that the AC will then finalize its recommendations and submit a completed final draft TOR/RFP to the Commission en banc. When the en banc approves it, the TOR/RFP gets sent out and will hopefully attract bidders.

The AC is composed of representatives from the Commission on Information Communication Technology  (CICT), the DOST, the DepEd, the UP, and various representatives from the private sector including people from the ICT industry and from civil society, such as PPCRV. The AC also sits in the Bids and Awards Committee as a non-voting member.

Advisory Council met today
August 15, 2007

Erwin Oliva has a report on today’s meeting of the Advisory Council, here.

The conclusion reported by Erwin is bound to cause a reaction, considering everybody and his mother has been egging the COMELEC to pilot automation for the barangay elections.

My take on it is that even if we could, there is still the question of consolidation. The key feature of automation is that it speeds up the consolidation of election results, so that is the feature we must be most keen to test: will consolidation be as quick as we want it to be. More to the point, will it be accurate? You just can’t test that within the framework of the barangay and SK elections.

Errata, re: SK Registration
August 2, 2007

This is to correct the announcement I made on radio station DZMM (630Khz) today, the 2nd of August 2007, in the matter of Sangguniang Kabataan registration tomorrow.

I said that applications for registration would be received tomorrow, the 3rd of August 2007, at the offices of Election Officers nationwide, and that – pursuant to the COMELEC arrangement with the National Youth Comission – applications for registration would be received on August 4 and 5 at the barangay level.

This is not correct.

There will be no reception of applications for registration tomorrow, despite the President’s declaration of a holiday for all students. It must be emphasized that the holiday was declared to give students the opportunity to travel to their home provinces, if they need to, in order to register at their respective barangays on the scheduled registration dates: August 4 and 5, 2007.

I apologize for any inconvenience my erroneous announcement may have caused.

July 19, 2007

Tomorrow, the 19th of July 2007, the COMELEC will be launching the internet voting – dubbed I-VOTE – pilot test for registered overseas absentee voters in Singapore.

Here’s the announcement:

In a major step forward along the path towards the full modernization
of Philippine elections, a non-binding remote electronic voting pilot
test will be conducted from the 20th  of July to the 8th of August
2007 in Singapore – host to over 15,000 registered Filipino overseas
absentee voters.

The pilot test aims to evaluate the advantages, usability, security,
and reliability of internet voting, in order to assess its potential
use in future elections, both overseas and domestic.

The pilot test will be jointly managed by the Commission on Elections,
the Department of Foreign Affairs – through the Philippine Embassy in
Singapore – and Scytl, the solution provider.

During the pilot test period, voters can cast their votes from their
homes, offices, and cyber cafés, in addition to the voting stations
that will be set up at the Philippine Embassy in Singapore.

Commissioner Florentino A. Tuason Jr., Chairman of the COMELEC
Committee on Overseas Absentee Voting (COAV) enjoins all overseas
Filipinos registered in Singapore, wherever they may be in the world,
to participate in the exercise, anytime from 8:00 AM of July 20 to
3:00 PM of August 8, 2007.

“Filipinos in Singapore are presented with a rare opportunity to make
history,” Commissioner Tuason said. “Although it is a non-binding
exercise, the historic opportunity to be the first to tread along the
path of modernized Philippine elections is a chance that should not be
missed,” he added.

And here’s the PDI article.

Another PDI article reports an interesting development: a shame campaign.

We’ve been proposing a shame campaign as well, one directed at candidates in elections. That way, or so the thinking goes, we get to them before they’re able to entrench themselves in power.

I’m not quite sure how it fits into the ‘shame manifesto,’ but the article seems to indicate that the following passage is indeed part of that document:

“We propose one count one canvass where the two middle canvasses are eliminated and the precinct level vote count is transmitted electronically to the national canvass center for electronic tabulation and from which local and national election results are announced. The goal is a 48-hour wait from the close of the precincts to the announcement of results.”

I agree in part. Our Law Department even tells me that their used to be only two canvasses: one at the provincial level, and the other at the national level. In practical terms, this meant that political parties needed to watch far fewer canvassing proceedings than they have to under the present set-up. Easier to watch, means easier to secure. Although we need legislative action to make this happen, we can at least do it even without automation.

The manifesto’s proposal however, takes it a step further and proposes a set-up that mirrors a Namfrel quick-count, i.e., there is a direct link (preferably electronic) between precincts and the national board of canvassers. This will definitely speed up the count, but it will also mean that the initial capital outlay will be large.

Another factor that  must be considered is the entry point of candidates’ counsels: where will they be given a chance to question the results? Just of the top of my head, I’d say that, first, the national canvass must remain open to challenge from counsels – which means a 48-hour waiting time for proclamation will probably not happen; and second, counsels must be present at the counting level, there to challenge everything they want to. Perhaps a better solution would be to restore the old system where the BEIs are composed of, not three teachers, but one teacher to act as chairman, and representatives from the dominant majority and the dominant minority as 2nd and 3rd members. That way, the counter checking is built in into the counting process.

I hope these changes can be made in time for 2010.

The Advisory Council Meeting 02.14.07
February 14, 2007

The Advisory Council has been convened by Secretary Ramon P. Sales, Chairman of the Commission on Information Communication Technology.

The members:

  • DOST – Usec. Fortunato T. dela Pena
  • Representative of the Academe – Manuel C. Ramos
  • Representatives of ICT Professional Organizations – Renato B. Garcia; Lilia C. Guillermo; Ivan John E. Uy
  • Representatives of Non-government Electoral Reform Organizations – Gonzalo O. Catan Jr.; Andie C. Lasala


The Advisory Council, through its spokesman and legal counsel Lorenzo Formoso III, announced its recommendation on whether the COMELEC ought to push through with automation under the AES law.

The long and short of it is what Formoso said “… the Advisory Council as a body recommended that the automated election system not be implemented for the May 14, 2007 elections …”

Excerpts from the Resolution of the Advisory Council, as signed by Messrs. Ramon Sales (CICT), Fortunato dela Pena (DOST), Renato Garcia (PETEF), Gonzalo Catan (PPCRV), Andie Lasala (CER), and Ivan Uy (PCS):

“Solutions Provider Readiness: Preliminarily, it must be stated that COMELEC does not have the option to simply buy the most suitable packaged or off-the-shelf AES and simply adapt or customize COMELEC processes and procedures to the purchased AES. While Section 31, paragraph 2 of RA 8436, as amended, authorizes the COMELEC to prescribe other mannor or procedure for the canvassing and consolidation of votes as technology evolves, such authority is not extended to other aspects of the electoral process and is further subject to the minimum capability requirements of the AES specified in Section 6 of RA 8436.

“COMELEC must, therefore, require the AES package that will be procured to be adapted or customized by the solutions provider to the existing electoral processes and procedures of the COMELEC as required by law, whether through changes in configurable parameters or through programming.”

This explanation, to me, speaks to the often repeated claims of ‘automation-NOW’ advocates that there are many available election solutions just waiting to be plucked, so to speak, from the commercial vine. It is not enough that the technology exists; that technology must be tailor-fitted (or tweaked or re-tooled, if you prefer) to the unique requirements of Philippine laws.

The AC struck this point merely a glancing blow, true enough, when it said that the COMELEC’s rule making authority did not extend to other aspects of the electoral process – other aspects that would inevitably be affected by automation – but it did expose the greater truth: while the COMELEC has rule making powers, these powers are not a substitute for the legislature. If we were to try to wrap existing rules and procedures around off-the-shelf tech designed for other election systems, the COMELEC would probably end up re-writing laws. If we did that, Congress would be up in arms, our work would be struck down by the Supreme Court, and we would, once again, be tarred and feathered. Thanks. But no thanks.

“For the procurement and deployment of the AES, the following activities need to be undertaken:

“For the Advisory Council:

  1. Prequalification of Vendors
  2. Deliberate on system technical requirements; systems integrity and accountability; and compliance with the law
  3. Shortlist highly recommended providers
  4. Officially transmit to the COMELEC the resolution

“For the COMELEC:

  1. Issue terms of reference
  2. Conduct bidding process
  3. Contract awarding
  4. Testing of the AES
  5. Acceptance testing
  6. Secure the certification of the AES

“As (the law) requires that the AES be certified by the Technical Evaluation Committee thru an established International Certification Entity eight weeks prior to the May 14 (polls) or not later than March 19, 2007, (all the steps except Securing the certification of the AES) have to be undertaken or completed between today (2/14/07) and that date, or a period of less than five weeks.

“The schedule is actually tighter because the certification process requires the completion of the following conditional events even prior to march 19, 2007:

  1. The successful conduct of a field testing process followed by a mock election event in one or more cities/municipalities;
  2. The succesful completion of an audit on the accuracy, functionality, and security controls of the AES software;
  3. The successful completion of a source code review;
  4. A certifiction that the source code is kept in escrow with the BSP;
  5. A certification that the source code reviewed is one and the same as that used by the equipment; and
  6. The development, provisioning, and operationalization of a continuity plan to cover risks to the AES.

“Given the above timelines and the legally mandated minimum capability requirements of the AES, it is very unlikely that an AES solutions provider can comply. And even assuming that an AES solutions provider can hurdle the above conditions, the issues of COMELEC readiness and voter -and other stakeholder- readiness remain.

Again, and at the risk of sounding smug, this is exactly what we’ve been saying all along. Alright I sound smug; deal with it.

Anyway, the AC resolution goes on to discuss the issue of COMELEC readiness to implement the AES. It talks about the various processes and procedures that will have to be re-engineered to accomodate the AES, the organizational and structural changes that will be needed, the training, the competence building …. and so on and so forth. None of which can be adequately achieved or met by the COMELEC in the time remaining.

And then the Resolution winds up with stakeholder readiness. This part I really appreciated because the AC resolution really fleshes out what it means to conduct a public acceptance and voter education program to support a pilot in 2 highly urbanized cities and 2 provinces in LuzViMin. We will have to educate and train:

  • 2,000,000 voters
  • 46,281 election officials
  • 22,968 local government officials
  • 309 COMELEC staff
  • all National politicial parties and affected Regional or Local parties
  • all National and affected Local candidates
  • the military and police in the areas affected; and
  • 15,000 members of the BEI (who are mandated by law to be IT-capable)

Rough, man.

All told, the AC Resolution paints a realistic picture of the challenges that the COMELEC will face if it forges on ahead with this pilot. As the Advisory Council itself says:

  1. “Given the timelines and specific requirements of RA 8436, as amended, it is highly unlikely that an AES solutions provider will be able, ready, and willing to comply with all the conditions and provide COMELEC with an AES solution in time for the May 14, 2007 elections.
  2. “The COMELEC does not have enough time to change manage in time for the May 14, 2007 elections.
  3. “The voting public and other electoral stakeholders in the pilot areas will not be ready for the AES for the may 14, 2007 elections.
  4. “The AES cannot be implemented in the 12 pilot areas for the May 14, 2007 elections and (the Advisory Council) cannot recommend the use of the AES therefor.”

Well, I guess that wraps it up and 30.

From Singapore
February 6, 2007

From Singapore, I got the sense that the local Filipino community there is eager for the implementation of internet voting. Over the weekend, Commissioner Florentino A. Tuason Jr., met with community leaders there and discussed the possibility of conducting voting over the internet. The meeting was packed with Filipinos working in the IT industry and their questions showed a deep deliberation of the pros and cons, as well as a an honest assessment of where the whole thing can go wrong. The meeting lasted for almost two hours, most of the time consumed during a question and answer session where the pinoy IT professionals raised scenario after scenario where the system could go wrong or be compromised. Tuason and his team responded to all queries frankly and in the end, the IT pros were content.

But it wasn’t all IT either. There were representatives from other groups, like the Batangas Varsitarians who gave assurances that they would assist the education drive – a kind of bayanihan effort that mobilizes those in the community with the savvy to assist those without.

Good vibes all around. Let see what happens next.