Tuesday, July 27, 2010

Revelation of the War


After the publishing of more than 90,000 obviously compromising documents on the wikileak one would wonder what else can be done in Afghanistan. No matter what strategies the US, Brussels and Kabul come up with their credibility has been severely damaged. There was already much distrist between the three parties especially between Kabul and Brussel and Washington; this takes it to a new level. It will help undermine the trust between the coalition partners fighting in Afghanistan - as well as increasing public anxiety.

I have been monitoring some media today to see the response and I think they got it wrong again. Instead of dealing with what has been leaked they are only blaming the whistleblowers. Children or Generals? Here are my questions: How are they supposed to explain that a special task force has been hunting down the Taliban for years now without success? How can they justify publicly praising cooperation with the Pakistani authorities when it turns out that the Pakistani secret service is 'probably the Taliban's most important non-Afghan helper'? … Governments should start reading them. This also shows how risky this operation is no matter what NATO is trying to do about it. For Nato, which since the end of the Cold War has seen it's role as that of the global policeman, this raises the question of its legitimacy.

The parallels with Vietnam are becoming obvious: In 1971 a US court forced the publishing of secret documents on the situation in Vietnam, nowadays the Internet ensures transparency. Is the same fate waiting for the US? many has argued 'Yes' but I disagree. The war in Afghanistan is winnable; only if Washington, Brussels and Kabul stop making mistakes; only if Washington and Brussels could stifle corruption in Kabul administration. These documents show how widespread corruption in Afghanistan is and how far it goes behind the crippled Kabul administration.

Sunday, July 25, 2010

SMS government: District Government registers client via cellphone

We are trying to promote the idea that district level government in Afghanistan need to engage population more proactively in the face of general disillusionment.

We work with a few district governments in southern Afghanistan to facilitate constituencies who are visiting the district offices to use government services. We have recruited and trained a few constituency caseworkers in the provinces of Wardak, Paktika, Nangarhar and Kunar; we are placing them inside the district governor offices. The candidate was briefed that the role is not to initiate a broad array of actions resulting in a speedy, favourable outcome. When performing casework, the scribe cannot force an agency to expedite a case or act in favour of a client. Scribe role is to intervene to facilitate the appropriate administrative processes. The scribes have a form which they fill out with details of each client/constituent visiting district government office. The form has three copies and one goes to the client.

Local governments (both district and provincial) should try to stay engaged with the public. Keeping continuous and systematic communication with them is a tool to gauge their mood and needs. The challenge in Afghanistan is lack of security, geographically dispersed villages, corruption and public reluctance to engage with the government; under such circumstances it is impossible for local governments to remained informed about public need and raise awareness of government efforts.

Using cellphone local governments can easily gather citizen input via voice platforms and SMS, grouping constituent interests by keywords. Then, based on those keywords, informing them of government actions in their interests.

we are working to use frontline sms to automate the system. this idea came to me when working with telephone operators; there is alot of neglect on the part of staff in a place like Afghanistan. they are not serious about details and cannot be bothered to be accurate.

Project Description and Technical Workflow

The idea is to make the inprocess form available via cellphone, so scribe can use client – or his cellphone when they don’t have – to directly input their detail into the database. we can automate the process this way. Client will be promoted to give regular feedback. District governor and related staff will be reminded about client status and scribe will be tasked to follow up on issues when hitting a barrier. The digitization of bureaucracy on an easy to use medium not only streamlines process but also makes up for lack of physical interaction as a result of worsening security situation.

We can make the form for data entry available via phones without using java enabled features. That we it can be available on all phones. The telephone which receives the form can send a simple hand-written SMS and our dB sms software can receive it. the data is not interpreted, validated or written in a database by FrontlineSMS. For this purpose we can develop a MySQL or MS Access database system to host the SMS data and to provide validation on the data received and to send an automatic reply to follow up with the sender, relate the problem to other stakeholders. For instance a client sends in form with a petition for Tazkira (birth certificate), the sender will receive an sms back, giving him the contact information for the Tazkira manager and working hours of the dept. the dB will also send a message to Tazkira manager informing him about the petition and any unusual notes.

For the functioning of the system we need two dBs, one at the front end and another at the back. The front-end database is the system that performs the validation and the queries would require adaptation for porting to another system.

We have received some money and currently working on a short-codes to toll-free SMS lines, to lower the barrier to entry for ordinary citizens who may not be able to afford a multiple SMS ‘conversation’ based on a menu tree.

There are a few other features that we could add to the casket: I am interested to look into geo-tagging, to give additional geographic context and trends analysis of citizen feedback and information requests.

So far we are only building this text based but we would like to bring in the voice; by using text to speech technology we can trigger recorded audio call-backs for those instances where audio is more appropriate than text.