Unique ID for citizens, voters' list, and German registration system

German registration system for citizens and aliens consists of a single data-base which is used for bringing people into the electoral rolls, tax net, into the compulsory education and vaccination systems, insurance systems, Births and deaths, etc. The following is the e-discussion that took place between "VOTE BENGALURU" members, amongst others, over the past month on the subject.

Kathyayini Chamaraj (CIVIC) to Mr M N Vidyashankar, CEO, Karnataka
     
Greetings and a very good evening to you!  I wish to draw your kind attention to an article in today's Business Line (15 December 2008) on Page 9 by Mohan Murti which inter alia explains the German registration system for citizens and aliens.  The registration system consists of a single data-base which is used for bringing people into the electoral rolls, tax net, into the compulsory education and vaccination systems, insurance systems, etc. Births and deaths are also automatically conveyed to this registry.  The onus is on the citizen to register himself and inform about any change in his/her address.  Without registering at the local registration office, no one can rent a house, get employment or get enrolled in a school.  I'm giving the link to the article and the relevant extracts below.  I believe we need to move towards such a unitary data-base managed by the local body or local police to eliminate the duplication of door-to-door surveys in the same localities by various departments for different purposes.
     

http://www.thehindubusinessline.com/2008/12/15/stories/2008121550290900.htm

     
"A truly unique, brilliant practice employed in Switzerland, Germany and Austria is the Meldewesen or registration system, by which each community, large and small, keeps track of the movements of its residents.

All inhabitants of a community, of whatever nationality, must report their coming and going to the local community office (panchayat) and the police. All citizens and foreigners living in the continent have ID cards that must be carried at all times.

No law in Germany is more rigidly enforced than that relating to the Meldewesen. Evasion is difficult and when detected is severely punished. The policemen on their beats are continually checking up new arrivals and special squads of detectives cover hotels and lodging-houses.

In most cities, hotel registers must be sent to police headquarters every day for examination and comparison. It is difficult for anyone to successfully conceal his identity.

The Meldewesen, maintained at great cost and often great personal inconvenience, is used by the officials for many purposes. Without the registration, no house can be rented, no company will employ, no school admission can be sought.

It is the local community office that notifies the tax board of new arrivals subject to taxation; it informs the school board of children who should be in school under the compulsory system of education; it informs medical authorities of unlicensed practitioners; it makes up lists of children to be vaccinated under the compulsory vaccination law; it compiles the election lists and furnishes information for the purpose of the state insurance act; it works in close co-operation with the post office and has charge of making up the annual city directory.

It is simply indicative of the German passion to have everything in its place. Indeed, to a German, the failure of a State to keep an accurate record of every citizen implies disorder.

Utter amazement was expressed by one German friend when informed by the writer that India has no such system. "How do you Indians keep track of each other?

How would you find a person whose address you had lost? How do the police know where anybody is living?" — these were the invariable questions."

MNV to KC
The German system of registration of citizen is the ideal scenario.  Unfortunately each wing of the Government thinks in a different way and is possessive of its own turf.  What makes matters worse is that we refuse to learn from some of the best practices available elsewhere in the World.

I mooted the idea identical to the one that you have cited, quite some time ago.  It fell on deaf ears and has not seen the light of the day.  I am in the same boat that you are in.  I feel rather helpless and suffocating.  Knowing fully that better solutions are available, but cannot implement them due to the archaic laws.

The only solution that will satisfy the general public in Additions, Deletions, Modifications in rolls an across the counter remedy which technology permits but law does not.

I am with you in your thoughts and suggestions.

I also take this opportunity to wish you and every one in CIVIC a happy, prosperous, peaceful and joyous year ahead.


I wrote to KC:
Very clearly the matter has to be pursued at P Chidambaram level. Can you with the help of Mr Vidyashankar figure out which are the specific laws coming in the way, and what needs to be done to facilitate introduction of the German kind of system. We can then perhaps prevail upon the likes of Rajeev Chandrasekhar, and others, to take it up with PC.


Anand Gupta added:
I shared a long flight with our CEO and we had extensive discussions on this matter. He proposed meeting couple of us to list down steps required for such a change. But he was very clear that most fundamental changes require Parliamentary vote (ordinance will not do). Biggest devil seems to be the law to have Voter Master List prepared only in states official language and any other list should be transliteration from the same. I intend to meet him in early Jan. If anyone here wants to contribute ideas in this discussion, do let me know and I'll coordinate.


I responded:
I can't see how any body could have stipulated that the master list should be in the official language of the state. Does anyone have access to this piece of legislation in order to re-confirm what the CEO, K is saying, and it is not just his interpretation? If this indeed is true, we have to move things at PC level, and for that, the CEO, K can possibly advise us as to what may be required.


I further added:
I read this in the latest Janaagraha Times (http://janaagraha.org/node/2328)

STEERS re-engineers the business processes involved in the preparation and maintenance of electoral rolls by leveraging the power of modern day technology – Relational Database Management Systems, web based network architecture with role based access control and Geographical Information Systems. A centralized electoral roll management system at the state level is envisaged, which will automate the workflows of every actor in the system – from the Booth Level Officers at the grassroots to the Chief Electoral Officer of the state. Under the new system, each actor is given a clear role that is monitored by the STEERS software application. The system facilitates ongoing maintenance of the voter list, doing away with the resource intensive comprehensive and summary revisions. The system also brings in transparency and accountability within the administration as well as increases accessibility to the citizen through the web.

STEERS has been approved by the Chief Election Commissioner, who has recommended each state to move into the new system after it has been piloted in Tamil Nadu.

With the advent of STEERS, it is only a matter of time before our voter lists boast of negligible error rates, heralding a climate for democracy to function in its true sense.

STEERS works towards reducing the error rates in the voter lists. The programme does so by reducing the barriers in the voter registration process for the citizen and standardizing workflow for preparation and maintenance of electoral rolls.

So, should I understand that the problems have been identified, and there's a solution in sight?

Jasmine Shah of Janaagraha responded:
I had shared brief info on STEERS in a mail to some members of this group around a month back. I had also offered to share more details on this in case anybody was interested. But I received no response.

STEERS has been catalyzed by Janaagraha almost two years back with the help of Microsoft. It is a very systemic response to all challenges in voter list management. It has also received an initial nod from ECI and some state CEOs. The pilot for the same is currently ON in Karnataka and not Tamil Nadu.

I am copying Rajesh to this mail, who is anchoring STEERS within Janaagraha. Any specific questions may be addressed to him.


Viswanath Srikantiah (CIVIC) responded:
I'll be damned if I want to be a number in my country, however unique it may be. Germany is NOT an example of a nation state which has carried itself creditably through history whatever be the current inclinations of the new republic.

The other countries which have done some system of registration for its citizens- virtual police states- and then have gone ahead to abrogate personal liberties drastically, halt migration of people,restrict free movement, recruit for the military et al - China, North Korea, Albania (then)

I am also amused by the association of Microsoft in this whole venture. Fascinating and what is their interest? I'm sure it is noble.

Where does our constitution stand on this issue? Where will this document reside? Who will have access to it ? Who will use it? and for what purpose?

I hope we also use the democratic institutions to debate this idea and bring consensus politically, hear all views before diving in. Using networks of powers and non-accountable to people institutions is hardly the right way. An article has come in a newspaper -great. A system has been thought about 2 years back -wonderful.

Let us but make haste slowly as we understand the implications of empowering an already powerful bureaucratic system and the notion of a state which is increasingly politically unaccountable.

What we need is democratic decentralization and political accountability at scales and at all levels.


Kathyayini responded:
Dear Vishwanath - You are already a 'numbered' citizen since I guess you have a voter ID card, PAN number, property tax payee code number, etc.  Others may be having ration cards numbers and other sundry numbers.  The state is the holder of all this data already.  All we are saying is that this numbering be done by one agency, instead of each department separately, and be shared and used commonly.    Germany may not have a great history but they are one of the best organised and efficient.  I think we could do with a little more efficiency. Germany is not the only country having it.  Several other European countries have it.  I believe even the US does.  Their social security number is the equivalent, I think.  I am all for decentralisation and political accountability.  But let's not forget that our politicians are the ones blocking decentralisation and all accountability mechanisms to people, currently.   But let them discuss this by all means and come to a consensus or take a vote, if they have the time or inclination for it!


Anand Gupta came back with:
Perhaps Vishwanath needs to realize that without a unique identity in the eyes of government, all of us are below the status of even a number. We are simply nobody!! Govt looks at us through large scale statistics where lakhs like us can be adjusted in the error margins. I would not waste efforts in finding who amongst states arms is the real culprit but I do know that all the organs - bureaucracy, politicians, executive, judiciary - are struggling to find some order in this chaos. Indexing recipients of states welfare and policies would be most basic step to create some sort of order. That would not rob you of your unique or special personal identity nor give you less chance of winning Bharat Ratna. Any report on status of any kind of people should be derived from actual data - not estimates cooked up by officials, as it happens now.
 
Jasmine/Rajesh - can you please provide pointers to details of STEERS. Let us see if this indeed would be right model to strive for.
 

I came back with:
Well said Kathyayini and Anand. The benefits of having a unique ID are far more than the problems that can arise out of its misuse. It's time such bogeys are dispelled (Nandan Nilekani has elaborated on it in his wonderful book 'Imagining India', which I would recommend as a must read for all citizens, particularly those engaging in public debate. It's a kind of a SWOT analysis of India. The opportunities are fantastic, and all ours to benefit from, if only we could build on our immense strengths).

Now, if Microsoft (alongwith Janaagraha's STEERS team) has been working on this from two years, and not been able to come up with a satisfactory process, there's apparently a policy issue in the way. So, why hasn't that been addressed?

Also, 'preparation and maintenance of the electoral rolls' across the country calls for a lot of work, which, I believe, can be undertaken competently only by professional organisations, like TCS (who have recently won the contract for front-ending the passport processing work), Wipro, Infosys, etc (I am leaving out Microsoft, since, being a non-Indian multi-national, some may consider it a security risk, though TCS and the like are already doing enough of 'sensitive' work for many countries). And, if they are involved right away, we can perhaps have near cent percent accurate voters' lists in a matter of months, leaving the NGOs to concentrate on other areas.


VS came back with:
Absolutely you are not alone Here is Shivraj Singh Chauhan CM of M.P. in support of a National identity number and the very interesting take people have on the issue

http://www.rediff.com/news/2009/jan/06mumterror-give-national-identity-cards-to-citizens-says-mp-cm.htm?zcc=rl

and a take by Ron Paul on the Social Security Number in the USA

http://www.lewrockwell.com/paul/paul71.html    

But our politics may be Democratic now with Obama and while we are at it why don't we hand over the Public Distribution System, Schools ,Police and Governance which is in a mess to the eminent companies enumerated by Muralidhar Rao not to forget transport :)

Onwards then and no time to waste on debates and discussions, we shall fix problems when they arise.

PS:  Thanks for reminding me that my government sees me below the status of a number How awful. Once I get it my status will rise. How nice. I want my number and can it please have a 7 in it? It's lucky for me ... and sure lets have blood samples,gene coding and biometric identification to make it absolutely fool proof. Cheers group and don't take yourself very seriously :)


Anand Gupta responded with:
Dear Vishwanath - Don't know what made you so cynical about this concept. One surely has freedom to look at the glass as half empty or half full. But chose your leaders carefully, don't just find quotes that suit your purpose. This whole concept of tracing terrorists using citizen Id is bogus.When we read stories of people in orissa eating rats and grass, while PDS food gets diverted to black market ; when we hear Medha Patkar  stating that compensation meant for displaced was siphoned off by few; when we read P Sainath reporting  how NREGS is helping "identified, enrolled" villagers - that is what one should look at.

Forget about personal status. We will all die soon and there's no point talking about "status" when you are dead. But as long as we are alive, we must realize that we are all connected to each other. If one has to do something, it should be to make positive changes in upgrading the "social fabric" that we all share. Bringing people up from starvation and penury, making them equal participant in nation building and plugging leakages in public system - all these can happen only when each participant/citizen is identified. Of course there will be criminals who take advantage of any progress - on-line financial frauds are now 1000 times more than armed bank robberies. Does it stop banks from computerizing accounts and transactions? Just look around you - anything, just anything available has dual use. Even your cell phone can be used to hit someone on the head and kill him/her. You can drive your car into a crowd and kill several in a moment. Why don't you do it? What stops you? Once you get answer to that, you will stop being cynical.


I further added:
Well, whatever Viswanath may want to say, it looks like even Mr Vidyashankar has also seen the writing on the wall, and is perhaps negotiating with one of the 'eminent companies' a package whereby they start a new division, with him as the CEO, and make an offer to the government (perhaps a Swiss Challenge approach) to revamp the entire 'electoral rolls preparation and maintenance' contract. Actually, a welcome idea, I would like to think.

 

Rajesh Choudhary, STEERS coordinator, Janaagraha Centre for Citizenship & Democracy, wrote:

STEERS (State Enhanced Electoral Role System) is the ideal solution ‘coz the solution comes from the supply side of the system. It combines the technology and civic participation efficiently to maintain the electoral system. It is spearheaded by Janaagraha in partnership with Microsoft. Due the unscheduled Karnataka election, the STEERS pilot got shifted to tamil nadu and now last week we completed the pilot for processing 20,000 form in Hosur Tamilnadu. Once the report is generated, we will move with rolling out in entire state. Please understand we are talking about paradigm shift and it will take some time to get technology in place and roll out plan in place.

But the good news is that ECI itself agrees that this is the solution going forward.

I am copying the content of document which was prepared by Mr Ram Seshan Ex Ceo Karnataka which will help the group to understand what exactly is the problem and what is the solution which STEERS offers.


Pitfalls in the current process

While the roll has moved from being a manual document and has been computerised, the business process for preparing the roll suffers from a legacy of the past manual system and has not fully leveraged on advances in information technology. The process as followed at present suffers a few drawbacks.


Shortcomings in part numbers of the roll

In the absence of a detailed map depicting demarcation of ACs and part numbers, clearly showing details of houses in an area, inclusions of voters gets jumbled up and many a time leads to geographically discontinuous sections in a part number of the roll. Exclusion of eligible citizens, in this scenario cannot be ruled out. This also results in an anomalous situation wherein a voter is attached to a distant polling station; criss crossing voter movement is a dampener for increased voter participation on the polling day.

Instances of discontinuous house numbering at the ground level, but a roll arranged with house numbers running serially causes difficulties in physical verification and enumeration. The problem gets compounded with partitioning of properties and/or individual houses giving way to apartments, especially in a fast growing metropolis like Bangalore.

Verification and/or enumeration of voters is quite an arduous task under these circumstances. Omission of dwellings is not uncommon. Inclusion of non existent dwellings is difficult, if not impossible, to detect. All this could lead to exclusion of eligible citizens and possible, if not probable, inflated rolls.


Poor availability of information

The above shortcomings in the roll can be partly managed if a rugged information system on the Assembly Constituency (AC) and the part number to which the citizen belongs to is readily available. However, such an information system is not available. This results in erroneous requests for inclusion, especially in case of electronic requests.

A limited search facility is available, wherein a citizen can ascertain whether his/her name has been included or not, if the AC to which one belongs to is known. Such a search facility is operationalised after migrating the data into Oracle database and hosting the details in the web site of the CEO or the ECI. Though a step forward in providing information to citizens, this involves manual intervention and manipulation of the database, which is undesirable. Further, keeping the migrated database current is another challenge.


High barrier for citizen involvement

The processes for voter inclusion/deletion/modification requires physical visit to selected locations that are not well advertised. This coupled with lack of easy access to information about the status of the request brings down the motivation of ordinary citizens to voluntarily participate in electoral roll management.

The citizen is kept in the dark on the outcome of his/her request. One needs to put in extra effort if the outcome of the request has to be known. Invariably, this would involve a visit to the office of the ERO/AERO, which very few can undertake. A more interactive process would generate enhanced interest from the citizen.

Inadequate process control

A consolidated list of requests received physically is not available with the ERO/AERO, rendering it difficult to monitor progress. Misplacement of forms, inadvertent omission in deciding on the request, etc., cannot be ruled out. Delays in the decision making and/or omissions severely dampen the involvement of the citizen, leading to exclusion of voters.

In case of electronic forms, the request is kept pending if the applicant is not available at the time of verification. Given the work environment in a city like Bangalore, giving prior intimation before the verification personnel visit the applicant is critical. The process adopted at present does not provide for such planning. There have been instances wherein such applications are lost sight of, resulting in the citizen losing confidence in the process.

Multiple inclusions (the same name being included more than once, either in the same AC or in different ACs) can be avoided if, before including a name the authorities concerned can ascertain whether the name is already included or not. Similarly, inclusions rejected at a certain point of time should not get included subsequently for want of harmonization of past decisions. Such cross AC decisions or decisions spanning various time horizons are severely constrained by the database structure adopted.

Requests for transposition of entries throw up a different challenge. While the inclusion may be accomplished, the associated deletion if transmitted manually to the concerned ERO, may be lost in the maze of papers handled. In a city like Bangalore where intra city migration is quite common, the roll gets obsolete in no time, with scores of ineligible entries.

There is no prescribed process for identification and correction of errors and anomalies found in the voter list. A one shot updating of the rolls every year, without a prescriptive process and continuous monitoring of the process, leads to a gradual deterioration of the quality of the voter list over a period of time.

The procedure breaks down in case of continuous updation. Requests received during the year are either kept pending for verification, or after verification the data entry process is clubbed with the next revision, defeating the very purpose for which continuous updation was conceived.

Manual intermediation at various stages – ascertaining the AC and part number, submission of forms, printing of electronic forms and obtaining signature, preparation of manuscripts, data entry, etc. – leads to concentration and duplication of work and does not lend itself to close monitoring. Consequently, one cannot be certain that any request can reach its logical end in a given time frame.


Poor data management

There is very little control over the data management. Partly due to lack of expertise to supervise this work and mainly due to little interest envisaged by the authorities concerned, this crucial work is left to the mercy of data entry vendors, leading to non standard and unacceptable procedures. Typical of these is bye passing the front end designed for data entry and accessing the database directly, thereby circumventing the modest controls put in place to prevent data entry errors. Consequently, junk characters in various fields, empty fields, inconsistent data, etc. are not uncommon. More dangerous is deletion or addition of names, as indicated by differences noticed between the number of voters in the mother roll and supplementary roll before integration and after integration.

If one can consider the mother roll as the master data and the changes that occur – during revisions and continuous updations – as transactions, then the care taken in any standard database management procedure to manage transactions data and master data is absent in the procedures laid down. Except for verifying control tables and other associated particulars and partial verification for data entry errors, the integrity of the data collected is not verified before inclusion in the supplementary roll. Typical of such instances include repeating of names already in the roll, inconsistent relationships like children having multiple parents, individuals having multiple spouses, etc. That the practice can have pernicious consequences may be appreciated by an instance noticed wherein an odd request for inclusion was rejected, but got included within a couple months during revision!

No value added analysis of the captured data is undertaken. Typical of these could be an analysis of the sex ratio to detect abnormal inclusions/exclusions, correlation between the growth in the number of voters and increase in population, multiple inclusions, rejected requests, etc. Such blind inclusion leaves the quality of the roll at the mercy of the lower most official collecting data and gives room for manipulations. A quick verification has indicated that such instances could be around 8 per cent of the existing voters in the state.

The current system does not have a prescribed process for managing the database. The data is copied manually many times over, copies are distributed across different systems, data represented in different formats and there is no prescribed control over how the data is modified in the database that is under the control of the CEO. Manual data duplication leads to data entry errors, duplication errors etc., which in turn requires costly correction process to clean up the data.

The situation is compounded by multiple language electoral rolls being maintained as separate databases. The high level of manual intervention involved in transliteration of the roll leaves one wondering whether the roll across different languages is identical or not.

Lack of security

Even with the above limitations, storage of the captured data requires much more attention. Many a time, vendors continue to hold the data that they have supposedly parted with. Due to limitations in the availability of technical expertise in the offices of the CEO and the DEOs, little attention is given to systematic storage of data. With multiple copies of the data in various places (with the DEOs, the CEO, the SLA and the vendors) what exactly is the current data that has to be relied upon is a moot question. In this scenario, capturing details of continuous updation and ensuring that the same finds entry in the roll seems to be well nigh impossible.

For maintaining control over any process, it is important to keep a log of various activities performed in the system (such as modification of voter information, etc.). This information can be helpful later on during the auditing process as well as for monitoring performance of the system. Since the current system does not keep this information, it is difficult to monitor and audit what happened during various processes.

This being the status of storage of data, issues like access control (to ensure that the data is not tampered with), back up of data and disaster recovery has received little attention.

 
The STEERS proposal

Time tested procedures in the management of the roll, namely, enumeration, draft publication, deciding on inclusions, deletions and alterations and the final publication of the roll have to be in accordance with the procedures prescribed in the Representation of People Act, 1950 and the rules made there under and hence few changes can be proposed.

This proposal primarily, seeks to redesign the procedure for recording and maintenance of voter particulars. Given the overlap between preparation and maintenance of the roll, the redesigned procedure would aid the preparation process to a large extent. Though the procedure draws upon the experience gained in Karnataka, given the administrative practices in the country, it would be relevant in every other state.


Comprehensive framework for better electoral roll management

Any electoral roll management process should address the issues raised in the previous section, namely –
 

• Rationalise part numbers of the roll to avoid exclusion and improve verification;
• Make information easily available and facilitate citizen involvement; and
• Adopt better process control, data management and security features.
•  

Rationalising part numbers to improve verification

Every part number in an AC should be geographically continuous, with the polling station centrally located in that area. No cross movement of voters, namely, voters of one area crossing a polling station and reaching another station to cast their vote should be allowed. All details of the area covered in a part number, namely, streets/localities in that area, houses/structures in every street indicated with prominent landmarks, etc., should be available as a GIS based map with drill down features.

Verification procedures vary widely, with dwellings identified in a particular manner in a year changing entirely in the subsequent year. Thus, correlating past verification with the current one becomes impossible. It is therefore necessary to standardise the verification protocol, which would also prevent exclusions and create a high barrier for inclusion of ineligible/fictitious voters.

Since there could be multiple dwellings in a single location and the number and character of dwellings in a physical location could change with time (an individual house could give way to multiple flats), as against marking every dwelling with a temporary number during every verification, which could change the next year, every location may be assigned a ULN on an as is where is basis, which would be permanent. A ULN may be based on a systematic numbering protocol or may adopt the municipal property (called khata) number. It may have multiple dwellings, which may then be numbered based on municipal/property numbers. Cross reference between ULN and municipal numbers, the character of the structure (residential, commercial, etc.), the number of floors, etc., can be maintained. If the number of dwellings in a ULN changes, one simply adds that many number of dwellings, without changing the ULN.

This approach would call for coordination with other agencies, like various Urban local bodies, the National Remote Sensing Agency and the NIC to leverage on the initiatives already launched. Further details in managing such an initiative on an ongoing basis would have to be worked out.

Till a larger initiative with a ULN is launched, section details and existing maps of the polling station area may be corrected and made more exhaustive, which can be a precursor to the more exhaustive system subsequently.

Easy access to information and citizen involvement

Citizens and citizen groups must be able to obtain electoral roll related information with ease and actively participate in the process, without the involvement of any intermediary. Availability of the following information, though illustrative, could be a starting point for citizen involvement.
 

• List of ACs and PCs in the state;
• Prominent areas and localities included in each AC;
• An AC locator system – given the postal address, the AC and the part number should be available;
• Status of inclusion in the roll – if included the AC and the part number in which the name has been included;
• Status of inclusion request – whether under process or if rejected reasons therefor or if included the AC and the part number, etc.

While a list of ACs and PCs can be made available easily, section details included in every part number should be expanded and stored in a manner that can support the AC locator system. The database tables containing state and constituency particulars, geographic details like cities, municipalities, etc., and details of polling stations, etc., may have to be suitably modified to support this function.

Status of inclusion in the roll possible at present through a limited search facility is grossly inadequate. To display the most current position any time and from any location, the electoral roll database must be stored centrally, without the need for any manual intervention. It is therefore, imperative to redesign the database structure to store the roll for all ACs centrally and provide for hardware and networking architecture with other attendant systems that go with such architecutre. This would also facilitate supporting status of inclusion requests.

Citizen involvement is greatly enhanced if intermediation by the official machinery is reduced/eliminated. Given the infrastructure available and the prevailing digital divide, the strategy would differ across geographical locations and the profile of the citizen involved.

A visit to the concerned office should be completely eliminated for willing citizens in urban areas. For instance, a citizen can visit a self-service portal (such as a website) or to convenient centres like Bangalore One (a one stop service centre for all citizen services like bill payment, tax remittance, etc.) or a RDS (a rural e – service centre at the sub taluka level for copies of land records, income certificates, etc.) and should be able to access information, initiate requests for inclusion/modification, etc., and print the request made for submission during official verification. Apart from disintermediation, this eager digitization would facilitate performance monitoring of the system and support status of inclusion requests.

However, in non urban areas and citizens without access to convenient centres, intermediation by the official machinery is inevitable. The BLO in such cases would play a crucial part in obtaining requests for inclusion/modification, etc., all through the year, as against periodically during revisions. Here too, eager digitization by entering the data at the nearest RDS before the application is handed over to the concerned AERO would bring benefits as above, apart from facilitating monitoring the performance of the BLO.

Better process control, data management and security

Improving the process adopted for managing the roll is another key factor for better quality electoral roll. Currently, within an overall framework, every district adopts a process that is subtly different from the other. Thus the quality of the roll critically depends on the process adopted and the rigour with which it is monitored, which to a very large extent, is person dependent. It is therefore, imperative to standardise the work flow and the process adopted to achieve a uniform quality of the roll.

The process adopted should address certain critical features –
 

@ Claims for inclusion and objections or transposition requests for modification, when initiated from a self-service portal or accepted by the BLO and digitized at the RDS should be accepted only after the details captured are confirmed;

@ The request made from self-service portals should be printable by the citizen for handing over to the official conducting the verification;

@ Every request, whether self initiated or initiated through the BLO shall have a unique RIN, which can be used for monitoring by the CEO/DEO and tracing supporting documents later;

@ The request should be transmitted to a central database and stored separately and individual EROs/AEROs should be able to view the applications coming under their purview;

@ A verification of the request should be taken up after preferably informing the applicant through email, wherever available;

@ Decision support systems to prevent duplication of entries, contradicting entries, etc. should be available which can support field verification;

@ A citizen should be able to ascertain the status of his/her request by referring to the RIN;

@ The decision on the request to be communicated to the applicant through email, if available, and in case of rejection, the reasons therefor should be available online;

@ Requests accepted should be appended in the electoral roll database, as a part of the supplementary roll, with necessary controls applicable in managing a database;

@ Correction of errors and anomalies, which could creep in despite a well defined and rigorous process. Value added analysis such as a skewed sex ratio, compensating additions (number male additions being equal to the number of female deletions), etc, should be possible for improving the integrity of the roll; and

@ Consolidation of the supplementary roll with the mother roll should be carried out regularly on the 1st of January every year, with adequate controls to ensure that electors belonging to a family are retained together.

All operations on the roll such as inclusions, corrections, deletions, etc., shall go through the ERMS, which would codify the workflow related to a process, route the activities to appropriate stakeholders at the right time and capture the progress at every stage. This information can in turn be used to monitor the performance of the process. This will also ensure the presence of only one master copy of the data in the system and would lead to a robust control and security policy around access rights to the electoral roll.

Another critical feature would be to maintain a history of all inclusions, modifications and rejections, creating an audit trail for enhanced control and having service level agreements to define the performance metrics for the process at every stage.

It is also important to take regular back-ups of the primary database on a periodic basis in a tertiary disk storage and store in a remote location as a part of disaster risk management. This will ensure that data can be restored in case of any failure.

Security of the database is primarily achieved by restricting direct access by stakeholders. Access to information shall be available only through the ERMS portal application. Audit trail, access controls to the database, physical controls at the data centre, etc., shall be made rigourous to improve security.

Hardware and network architecture

It is proposed that the databases of electoral roll and other tables be centrally stored in the office of the CEO. A stack of servers (a main server, a firewall, a web server and a back up server) with power back up, air conditioning and other facilities may be created.

The main server would host the database and all other tables. The firewall would restrict access to the servers. The web server would provide electoral roll information to the public, either directly, or if the telecommunication infrastructure cannot handle the expected traffic, this server may be connected to the NIC, which can then host the information.

Since data would reside in a centralised database, it is important to ensure the database is available almost all the time as without the database, the rest of the system would come to a grinding halt. The database systems should survive the following failures in order to be available continuously.

@ Server failure – this can be addressed by creating a cluster of server nodes that would share the storage sub-system. In this case, if one server node fails, another takes over the database and provides services to application thus ensuring availability of the system; and

@ Disk failure – this can be addressed in two ways. The server is taken offline, the disk repaired and the database restored, after which the server is brought online. This approach would result in unavailability of the database till the server is brought online. Alternatively, mirroring, involving three servers – primary, mirror and a witness may be used. The primary database is replicated in the mirror so as to ensure that the mirror holds an exact to near exact copy of the primary database. When the primary database goes down, the witness server switches the roles of primary and mirror – in this case mirror becomes primary and primary becomes the mirror. This approach can be used for disaster recovery mechanism as well.

A combination of both the approaches is recommended in order to provide high availability of the database system. However, the final configuration of the servers may be decided by a group of experts to be constituted for this purpose.

The server in the office of the CEO would be connected through a leased line of appropriate capacity to the telephone exchange to provide access to the internet. Alternatively, the office of the CEO may be connected to the NIC which can then route the traffic. This configuration may provide an additional layer of security to the database. This would receive data on claims, objections and transpositions requests initiated by individual citizens and digitized data of requests received by the BLO and support information needs of citizens through the internet cloud.

All AEROs can be connected to the server through the internet cloud in two ways – through any designated RDS located in the taluk headquarters or through a dedicated computer with a broad band connection located in the office. While the former is a solution specific to Karnataka (though other states are expected to catch up soon) and is cost effective, the latter can be applied to any situation. However, as the RDS in the taluk headquarters may not be within the office premises, performance could be hampered to some extent.

All DEOs have access to the internet through the NIC network and can hence have access to the ERMS for monitoring performance.

Hope this note gives detailed insight about STEERS to all of you.


I responded:

Oh Ok! Looks like some real thinking has gone into the matter. But, what is the arrangement between Janaagraha (STEERS), Microsoft and ECI? Is this some kind of a pilot study for evolving the parameters, after fixing which some kind of a professional contract will be signed up? I am asking this because how does one ensure accountability without a professional contract?

Would someone like to enlighten me?

Muralidhar Rao