Greetings from Mumbai! Hanging out with video hackers from pad.ma
We had made plans to travel back to Pune today, to attend the Gnunify open source conference, have some meetings tomorrow, and work a bit more with IndicTrans and basically chill out from Mumbai. But somewhere on the eastern outskirts of the city, after an hour and a half on the bus already and the guy in front reclining extremely right into my knees, we decided to ditch that plan and stick to Mumbai until Monday when we fly to Trivandrum. Mumbai is actually pretty pleasant when you’re in one place, it’s the moving about which causes problems.
The workshop on Thursday and Friday went really well. Less attended than expected, but the dozen-plus participants were very enthusiastic and patient as we shook out the bugs in the agenda, the software and hardware. We went through the full process of acquiring data, uploading to OSM, and editing the map. Twice. Everyone was very keen to continue, and went away equipped to carry on mapping Mumbai in OSM on their own. The slides from the introduction of the workshop are here, and the Beginners’ Guide on the wiki was effective for the more in depth tutorial section.
Finally had the pleasure to meet Nagarjuna and his assistant Alpesh Gajbe, who have been the primary organizers and herders for our trip. Without their incredible effort, it wouldn’t have happened. Nagarjuna is one of the most unusual thinkers I have met here, or anywhere — just to start he’s an India who loves beef. His main effort is gnowledge, and he’s quite busy developing free and open curriculums. We were interviewed by a Mumbai paper, and put through a slightly ridiculous photo shoot, like taking marketing shots for a band. And the guy who spotted my FooCamp t-shirt on the highway showed up with his brother, and made several business propositions — a total gold rush here.
Also met Terrence D’Souza, an open source hardware hacker, who has built this eyeglass case GPS logger.
He was kind enough to give me one. These are really cool devices, I love how easy they are to open. He’s going to start selling them soon, I’ll give details when that starts.
The AND import was a big topic of discussion. There is a resemblance to reality in that data, but the geographic positioning is inconsistently incorrect in odd ways, with many missing streets as well. We’ve pondered how the data could end up like this, and it looks like it may have been traced from an incorrect paper map? Would be interested to hear from AND. Still there’s something to salvage there, and after some deliberation, we’ve decided to leave it up to individual editors to correct areas as they survey, or by comparing against the excellent Yahoo imagery in Mumbai, or they can just delete the data in a local area if it has no value. In contrast, Pune is just wildly wrong, and we may just ditch that entirely.
There has been a lot of discussion of applications, involving so many people and organizations and connections that my head is swimming, on top of a deep pool of implicit knowledge of the working of India in general. There seems to be great opportunity in education, as part of training IT students. And in mapping urban change, empowering slum dwellers, and redevelopment — the remit of CRIT and PUKAR. (We had a post-workshop Thai dinner celebration with Suhit Kelkar from PUKAR, which was very enjoyable and even more notable since we got home by taxi without getting lost — I made sure to charge the Garmin GPS for this!) Also applications in data collection for rural development, and models for transitioning to open GIS data collection. Thinking about these applications, especially the development and redevelopment of Mumbai, I can not fathom how these problems could be solved without maps as a tool for public discourse. I mean, there are no proper maps you can even buy. The sad fact is that there are maps, but held very tightly by a few, and this is an issue of power. Along the way, I seemed to have been offered a professorship at an Indian university, to provide technical education and help boost the use of open geo data in these fields — an offer I don’t think I’d have time to properly fulfill, but wow.
Btw, did you know that in Indic writing systems, groups of letters might be contracted, switch places, or operated on in all sorts of other fashions. It gets quite complicated in ways I don’t understand, but the upshot is that Indic fonts require not just rendering for individual letters, but also rulesets. So now I fully understand the undertaking of IndicTrans in producing open source Indic localisation.
After the chilled weekend in Mumbai (as chilled as that can be) the next stop is Trivandrum. We’ve also decided to make a visit up to Ludhiana to visit Professor H. S. Rai, who is already actively contributing to OpenStreetMap. Check out the map of his university.