Dinesh


Biography

Computer Scientist and founder of SERVELOTS, a web service provider for Small to Medium Enterprises.

3rd December 2008, Bangalore

When I was traveling in 2005 with a group of computer scientists from the University of Amsterdam and the Hogeschool of Amsterdam to establish relationships with computer science departments in India, one day Dinesh stepped into our bus to guide us through Bangalore. It was a happy encounter since we found that he actually knew some of us because he had worked at the Centrum voor Wiskunde en Informatica in Amsterdam. Dinesh only uses one name 'Dinesh'. He is not fond of social conventions and is highly motivated by his curiosity and compassion to all things that happen around him. At the time it had been 4 months since the Tsunami happened and in his office in Bangalore about 15 people were working very hard to offer help with the use of technology to the rebuild activities that were happening in the areas of great destruction. Dinesh and friends were using Pantoto, an open source platform that they built with Servelots and that would facilitate rescue workers to know who was doing what, where and when. This time we meet at the house of Zainab Bawa and after dinner Dinesh told us about his experiences with the Tsunami rescue and rebuild phase and elaborates on how the use of technology may be useful or not in such disasters that happen

Summary

When the Tsunami hit on the 26th of December 2004, Dinesh was taking some relaxed time close to one of the places where the disaster happened. After the first panic in the relief phase, coordination was necessary to start the rebuild. And after a few months many people who were in the relief mode had settled down and long-term rehabilitation needed to be organized. At this moment in time some coastal organizations of fishermen asked Dinesh and his friends to help out. There were lots of local groups, but there were lots and lots of other people from everywhere who had launched themselves in the Tsunami areas as well, because of their expertise in some area. There was a lot of potential for somebody to make an impression. And, very important, a lot of money was dedicated to the Tsunami disaster. A number of international agencies, which specifically had a handle on this money, were quite interested in this precedent to the right cause, the right need.

So in the first few months Dinesh and friends built a portal to show Who's doing What and Where in the open source platform Pantoto, which they had developed before. It wanted to contribute to the re-coordinating of the rehabilitation work. One of the issues they ran into was the fact that very few volunteers on location knew how to handle the platform and for it to be useful reports had to be written, edited and uploaded everyday. Also because the portal was open source, some of the more institutional organizations had difficulty with trusting the thing. Dinesh argues that his team believed that one does not only want to show organizations, but also an individual or group of two that contributes to the rescue. To be acknowledged when being part of a rescue or building up effort is for all involved important, not only for NGO's but for individuals as well.

A few months later, when the rebuild phase was getting into consolidation mode, the large Non Governmental Organizations had obtained a lot of money by that time from all over the world. Faced with the need for coordination, but also faced with the need for establishing their own reputation towards their own funders, the large NGO's refused to collaborate in such a portal anymore. Their identity and their presence in the region were at stake. So they all went to their head offices with a request to build such a coordinating portal and to build it within their own frameworks. As a result of the abundance of money they had received, they all made their own portal in which they only presented their own work and effort with the purpose to satisfy their own sponsors and funders in the first place. The primary thing was to tell the world and your particular funder what you are doing, independent of what other people do. You don't want to potentially relate to the same funders or make them compare. You actually try to capsule it in a way that the portal will say what you would want it to say.

Even in times of crisis reputation building is very important, more important than collaborating in order to solve the crisis, so it seems argues Dinesh. On the long term NGO's have an identity. It's not about how much work the coordinator did with others, but it's what they are going to say about themselves. A portal can now say "we are doing this, this, this and this". It will not be with reference to anybody else. Here Dinesh delineates relief versus rehabilitation. He didn't see any of these kinds of attitudes during the relief process. People are dying, people are hurt, houses are falling down and people don't have anywhere to go. They need to be rescued. But now it was already four months ago that the Tsunami happened and it is merely a thing in memory for most. So now the rehabilitation has a prediction from four months to two and a half years. People are settled in. Then it's about whether we build a house for you or for the hooker. Maybe that we propose that building ten thousand houses is a good idea. This kind of thing, this is where all those politics of presence come in.

Dinesh and his team decided that because of these developments it did not make sense anymore to provide a portal for all since its capacity to show effort and deeds done had seriously been jeopardized by the NGO's will to primarily show their own presence at the rehabilitation places in which they were involved. Having gathered so much information already they decided create a location-based tool that would retrieve all information of all involved other portals, national and international press and other Internet sources to gather and present 'all that mattered to a specific place'. This location-based tool also provides the same place with all the different names that it has in the variety of international and Indian languages. When the rebuild phase was over and the NGO's needed to create solid archives of all that happened, this tool was used regularly because it provided a comprehensive overview of all that was written about specific places.

Dinesh argues that it is very hard to say that a system can be used during the relief; maybe personal calls, cell phones. It's a loose network at such a time. If you can anticipate crisis, like no rain and something else, you can work on these issues. "But something like a Tsunami; you don't even know where it hits, you don't know how wide it will be, how many miles. I can't imagine any anticipated coordination to stimulate redundancy and all that. And it's a very emotional thing you know. Some way that the Tsunami hit, it was slowly flooding, makes a big difference on how rushed you want to be in offering this relief thing. So I wouldn't say I know anything, at least not for this kind of thing."

Transcript

The following is an edited transcription of the conversation. Film fragments of the conversation are included to illustrate parts of the transcribed text.

[Sequence 1] [Sequence 2] [Sequence 3]

[Top] [Transcript]

Sequence 1

CN: So when I saw you three years ago you were very hard working to make computer support for the Tsunami victims. Can you please tell the story? And I told you before it's about the witnessed presence and what happens then.

D: We just started on the Tsunami because coincidentally we were taking some relief and all that. And we were actually close to where the Tsunami hit. Bangalore is quite far from those areas. And then we met some people who were very active in the rehabilitation. And they were looking for computer support. And what they wanted to do was called "Who's doing What Where?". They wanted to kind of get an idea, a database or an active database, an everyday updated thing to understand who's doing, what's issued, where they are

CN: To try and coordinate the help?

D: Yeah, to try and coordinate. It didn't seem like a hard thing to do something like that, but there are a lot of small issues that need to be taken care of. We did got to ask other people for help, to see if they could do it. And they were all very busy with other things. So we ended up taking up the work. So basically the rush after a disaster is relief. There is no coordination. Actually I personally feel there is no sense in coordination because of the urgency with which things have to go. You can't coordinate so much. You can't say, "Oh, maybe two bags shall go there and we should reduce it." There is no problem in two bags going there instead of one, as long as everybody gets it. I'm particularly saying this because a lot of post relief discussions have been about how they could have optimized it. So it was not about that clinical coordination. This whole project was about re-coordinating the rehabilitation work and not the relief work. You know, when you came it was already march-April. So that's already four months after the Tsunami. So that's when we started. By that time many people who were in the relief mode had settled down and then it was a long-term rehabilitation thing. So there are lots of local groups, but there were lots and lots of other people from everywhere who had launched themselves in the Tsunami areas, because of their expertise in some area. They could be a child psychiatrist group from Bangalore; because there were so many displaced children now and how to make them feel up to the new post disaster thing. Or it could be architects who had experienced the Gujarat earthquake, who had things to do in this new Tsunami thing. From that too lots of international things. One thing was that there was a lot of potential for a lot of money and the potential for somebody to make an impression was also quite large. A lot of money was dedicated to the Tsunami disaster. And a number of international agencies, which specifically had a handle on this money, were quite interested in this precedent to the right cause, the right need. Anyway, when we were there it was just after the rush of relief and now it was time to see what can we coordinate regarding the rehabilitation work. So there were lots of NGO's and community groups. One was the local fisherman's groups, very well established, called... I forget the names, but quite active in the coastal areas. So four of them.. We had a meeting of like thirty people and lots of representation was from these four or five very strong groups who had various activities going on in various places. And a very important thing was this group of volunteers who wanted to do something, but they couldn't come to the location.

CN: Because they witnessed and they wanted to do something....

D: Yeah. So after the Tsunami, even our call, three of individual friends, got like three thousand volunteers who subscribed from all over the world. Most of them couldn't come to the Tsunami area, but they had time and they were willing to do whatever. So like Caroline from Amsterdam says, "What can I do? What can I do?" That kind of thing. So the idea of the coordinate Pantoto was also for those sorts of things. Somebody might be a Tamilian living in the US and they could help with the translation problem, either for the media or for the coordinating people who can't speak Tamil. It would have to be in English. So then there would be better coordination. Or the other way things for news and work were reported in English and it had to be translated to Tamil. You can think of like various of these activities. So that was the work we were doing when you came. So it was basically to map who was doing what and where.

CN: So there is a disaster happening. Many people see it. It's so shocking to witness behind a television screen that people want to do something. And you want to put a system in place so people know where to go to make sense. So you wanted the system to help rebuild in a disaster zone where no system works any more, which was the impression on television at least. And then what happened?

D: This is just a portal, right? So you don't want to give too much credit to how it helps in terms of rebuild. The idea is that there are maybe a thousand people working in various small groups, some of them are working with Dalit, the untouchables, you know.

CN: But the problem is that their work has to be in your database. It has to be written and to be edited. It has to be used.

D: It has to be updated every day.

CN: So what happens with that?

D: What happened is an interesting thing. A lot of people who were interested in this coordination portal, they expressed this idea through their own funders or funding agencies. So they were all fond of this portal, but our approach was to do an open-source quality based something. And it was going reasonably well. But once they got money they had to basically get their own portals. You know, that's their identity. That's their presence, right? The idea of this portal was the presence of individuals and organizations working towards a supporting effort. But the presence they had on the long term, was their own identity. And then they were worried on how they're going to present their identity.

CN: But this is also very interesting for disaster management. One of the big problems is that different services in society don't collaborate. But even in rehabilitation when you offer an open-source, but there is no chief, no authority, it's a portal, so the egos or the organizations cannot shine and everybody builds their own and there is no more coordination.

D: Yes, see what happens especially in an Indian context is that even to use a portal you need to have some kind of a confidence in your capacity to interact with it, to feel comfortable with it and all that. That already defines a certain kind of a person. Most of these people are part of these organizations but the volunteers, who can do this, are mostly elsewhere in the world. On the location there are not these people who have this capacity. So one of the main things that happen is that all the people, with this capacity who are of these organizations, immediately see that if they lose their capacity for their own portal. On the long term one will be able to use tele-funding agencies and so much money has gone into satisfying these dissipation needs. The second one is that on the long term they have an identity. It's not about how much work the coordinator did with others but it's what they are going to say about themselves. A portal can now say "we are doing this, this, this and this". It will not be with reference to anybody else. That freedom and the style of representation and all that can all be very autonomous.

CN: But here the argument is always that open-source functions on reputation that all social networks function on reputation. Why was not your portal the reputation place? Or is it because of the disaster moment?

D: No, no, I think the disaster moment brings a greater need for this kind of a reputation. It increases the need for this kind of reputation because everyday something is happening and you don't know; you might come and help or you might come and you will input data. It's an open community thing, so anybody can input data. So you can input data that really highlights yourself or your organization. So this reputation thing and rating thing is very important. So we have worked on that over time. It doesn't matter how you say that you are great or somebody else is, because the system kind of nurtures the long-term effect of this. But again I already told you why this was not the need, because of the local capacities. Imagine there were the local volunteers, a hundred people who were very comfortable with computers and they wanted to use a portal and they didn't belong to all these other organizations, then what you're saying makes perfect sense. But if I am part of this organization, which has gotten funding and has their own portal, your primary thing is to tell the world and your funder what you are doing, independent of what other people do. You don't want to potentially relate to the same funders or make them compare. You actually try to capsule it in a way that the portal will say what you would want it to say.

[Top] [Transcript]

Sequence 2

CN: Yes, but I'm just really trying to understand, because it resonates with research that's being done by IIDS, where it is analyzed that in crisis people still go for their own interest. So even in super crisis moments, people will make choices for their own interest and not to solve the crisis. So this is a point that is about witnessing. If I'm witnessed to do the great job and get more funding, the witnesses judging will generate future for me and my organization. So how could you get around this?

D: I try to delineate relief versus rehabilitation. I don't see any of these kinds of attitudes during relief process. People are dying, people are hurt, houses are falling down and they don't have anywhere to go. They need to be rescued. I don't see any of these things coming in during that time. But this was already four months. So now we have prediction from four months to two and a half years. People are settled in. Then it's about whether we build a house for you or for the hooker. Maybe that we propose that building ten thousand houses is a good idea. This kind of thing, this is where all those politics of presence come in.

CN: So presence as power game?

D: Yes. So the crisis is just in your mind, in the sense that it's four, five months after the Tsunami. All these issues have settled.

CN: And was there any roll for systems during the crisis?

D: Personally I don't think there was any role for the system we were trying to do, because you want to give the benefit of doubt to a person that's suffering rather than not. Meaning that if Caroline is also taking some supplies and if I'm also taking some supplies to that place, and suppose you barely make it, it's better that two people going to the same place, than saying only you going and me not going. So it's very hard to say that a system can be used during the relief, maybe personal calls, and cell phones. It's a loose network you know. I'd rather go to the side of redundancy than coordinated control of the sources.

CN: How can you stimulate redundancy and self-organization?

D: If you anticipate crisis, you probably can work on all this. So if you anticipate no rain and something else, you can work on these issues. But something like a Tsunami; you don't even know where it hits, you don't know how wide it will be, how many miles. I can't imagine any anticipated coordination to stimulate redundancy and all that. And it's a very emotional thing you know. Some way that the Tsunami hit, it was slowly flooding, makes a big difference on how rushed you want to be in offering this relief thing. So I wouldn't say I know anything, at least not for this kind of thing.

CN: Can you think of something now that you would have wished to have done something different with the portal, should you have the position to do so?

D: I mean it can be used for many other things.

CN: But I mean at that time, in hindsight.

D: It's a very small subtle thing. It could have gone well, it's not like the capacity is not there. But I just think we got de-motivated, when in continuing with this in a meeting with the main four or five organizations and when we realized three or four of the five all have their own thing. We thought our work was done. Otherwise it's a very good idea.

CN: But maybe the conclusion should be that even in a moment of crisis that taking care of identities is very important? Maybe that was underestimated?

D: No, we actually tried to take this into account, but maybe they didn't know of course. See there are two things about the identity. In how the system is built, we are very careful in providing these long-term identities for all these people who are involved with all these different works, whether you're small or large. Because the nice thing about this portal is that even a two-person group, who work with some community, will have a presence. Where as now it's only these large groups. So we had worked out this long-term identity thing. It was a very complex; it had the potential of doing friend of a friend, rating, to all kinds of issues that come up. In the long term it had that capacity. But the fact that this cannot be expressed and convinced, that this is a better identity, is one of the main reasons they couldn't go for our system. That's a system that they do not know. Print a book, Ok? You already know how it goes. Say put it on the web and that becomes an acceptable thing. I think people understand that. "Ok, putting it on the web works like this. It brings me this kind of authentication". But concerning identity there were issues. Because anybody can put anything right? How does it become yours? People can doubt so many things. So I think that's where the problem was, in terms of identity establishment. They were established reputation systems. The same day you reported to your funding agency also your identity was established.

[Top] [Transcript]

Sequence 3

CN: Can you maybe the last ten-minute elaborate because you have made several tools for communities, successful tools. Can you elaborate on how the witnessing, of people seeing each other, how that dynamic actually functions? The dynamic of a community; the difference between a group of web pages and a community is that people see each other, there is witnessing happening.

D: Ok, so one thing that we did, even after we stopped using the coordination tool, was to still provide location specific data, regardless of in who's portal it is. This location, we can get data from all of them and maybe the news from maybe the Washington Times to whatever, as long as this location is mentioned somewhere, so that's relevant to them. So then each location can catch up with what is happening in their thing, in terms of how people are reporting, what they're doing to what the news contains to all that. So we worked next on that kind of a thing. So that tool basically defined a whole location tree, different names like Namipratnar can be spelled ten different ways because it's an Indian word, now it is translated in English, even in Indian, with so many languages, it can have more names. So we kind of need a system that kind of builds a concept of this location. And say like "anything that matters to this location". It's like taking the feeds of all these activities and channeling it through a filter to get only the relevant feeds for this location. So that's the kind of thing that we did.

CN: So if you translate this in terms of semantic web. Because you say you have to make a concept to represent a place. So how would you rephrase it into--?

D: Yeah. A concept is not actually a sense of a semantic word, but it comes really close to getting the best out of text. So you just have text, you mine it to see if this concept matches. With semantic work you already have different tags, using semantic tags.

CN: But here the local connection, the GPS connection is the discriminating criterium.

D: Yeah. It was not a GPS kind of thing, but it's close. It's a location. It's a location system.

CN: It's a real place, so people can also check it to trust it.

CN: Because that's the thing, because if there's no trust, concepts don't --it's like what happened with you thing. The larger organizations didn't trust you, because they don't recognize you. So now when you start doing "place" people can check it. They live there.

D: It's true. And some of the organizations later used this thing, because they were building long-term digital archives. Related not just to Tsunami's, it's the general documentation centre. So they then got into adding also the Tsunami stuff, so they used some of this stuff to enlarge their archives.

CN: Is there anything you want to add?

D: Yes. I don't think it is effort loss because it's like this ideal. It gives a sort of wide understanding of the systems and how complicated or intricate the thinking of this process is, to appreciate different levels of identity. Like in the end we will still be asking, "What is the identity of this guy who doesn't have access to any of this?" So none of these kinds of systems should leave out that kind of person. So if our system worked very well and in that process we left out these people, then it would be that everybody accepted everything in this system as 'The identity of the people who volunteered for the Tsunami'. Then we would have a very bad system. Like now it's very clear. Volunteers work for organizations, so they represent their own thing. But there are all these other people who needed an opportunity to represent them selves as Tsunami volunteer. But if this had worked very well, we probably would have had respect, you don't know.

CN: It's like a choice?

D: Yeah.

CN: Thank you.

This website is work in progress. You can use the content of this website under the Creative Commons license Attribution-NonCommercial-NoDerivs.