The past fortnight has seen a couple of milestones, a promising start to a regular GIS networking event in Phnom Penh and a planning event for next year.

The Friday before last we all bundled into a rented van and drove over the Tonle Sap via the Japanese Bridge to the peninsula at its confluence with the Tonle Mekong. We crossed the peninsula, past the innate (but solid), imposing architecture of Norton University and waited for the ferry across the Mekong. This was not a typical western ferry with a cavernous enclosed space and aircraft-like seats up above in an airconditioned space with a nice big plasma panel. It had the air of a glorified raft, with an huge gantry straddling the middle where the pilot sat, presumably on a swivel chair since the boat did not turn when going back to the other side.

We went to a lovely little resort, Smamingo, metres from the Mekong which had a pool and a number of bamboo huts much like you see on a holiday program like Getaway. I was given one of these all to myself, but it was a bit of a waste given I spent very little time in it. The point of going there was to spend the next two days planning next year's activities and projects. This also involved a couple of hours reviewing 2011, all very interesting for a new recruit like me. The main highlight of the year, as mentioned in previous updates, was getting suspended for our report on eviction of communities along the railways that are being rehabilitated in Phnom Penh. People highlighted positives and negatives to do with this. It gave us notoriety and credibility (in some circles), leading to more funding. It also meant that our visits to relocation sites had to become more low key and we could no longer do things at sites threatened with eviction. This should change on 1/1/12 with the lifting of our suspension, although we shouldn't assume that this will occur just because the government has said so.

Proceedings were lead by Jill 1, hence it was all in English, handy! I spent most of my time involved in planning activities of the mapping team and advocacy team. It appeared that generally 2012 would involve much the same work as 2011. An interesting thing to do was try to forecast the cost of the advocacy team's community mapping sessions for the year. Yes, the advocacy team does mapping although it is more participatory GIS than typical desktop stuff. It involves consulting with community members, their leaders and local authorities to produce a large hand drawn map of their surroundings. This data is processed further by the advocacy team back in the office. With all of the transport costs for ferrying people to a meeting point multiple times on various discussions, food for each of these occasions, printing costs etc we predicted that we would spend in the region of $US1600 next year. This was within the budget allocated by the donor, but it would be interesting to find out if that's actually how much is spent.

I won't divulge anything else about our plan of attack next year, but we have one particularly cool mapping project coming up, of which many will see the results. On the topic of the retreat generally, I saw a couple of old BMW motorbikes at Smamingo. There was a really old, black one that I couldn't identify (until I remembered this), welcome to comment if you know! It was also nice just to hang out with the guys from work without it being in the office. We swam in the pool, played pool and ate good food. Lastly, I also got to meet the chairman of the board who has 12 years' experience in development work and has a business background. He was a very nice man who was obviously very knowledgable about organisational development. The last order of proceedings was him giving an hour's lecture in Khmer about typically patterns of development of NGOs and how to use this knowledge to improve our's. I couldn't understand a thing he said, but it was funny because the lecture slides were all in English!

Ok, there were a couple of milestones this fortnight. First, phase 1 of the Urban Voice webmap is complete. This is a major achievement for the NGO. It is quite a fantastic deployment of the Ushahidi platform, even thought we haven't enabled reporting via SMS yet. The key ways in which it's better than arguably the best deployment until our's, Libya Crisis Map, are summarised below:

• Other deployments do not enable the user to view multiple layers simultaneously in a simple fashion. The Reports tab can be used to filter on a number of categories, although this does not symbolise points by category such that the user must click each point to know what type of information it is about. We have created a function where users can select multiple categories to view at once. This function has some nice features, for example a clicking on the child of a parent category will deselect the parent, and vice versa. One can also select the parent of one category and one or more children of another parent category.

• The user can select between three basemaps, OpenStreetMap (OSM) or Google's street map or satellite imagery. This is gives them the greatest chance of being able to find the location they are interested in when submitting a report. For example, Google has vacillated over providing street numbers in Phnom Penh during the past year (they're back on again finally as of this week), so OSM may be more useful in finding a location based on street numbers as they remain on the OSM map. Alternatively, the satellite imagery might be more useful for reports where there is no street nearby or in identifying a particular point along a street.

• The map can be embedded in a blog or webpage, including being able to select the size of the map in pixels. This serves a dynamic map to the blog or webpage, displaying the layers selected by the user prior to requesting the embed code. Users that click links in pop ups on the embedded map are taken to the main Urban Voice site, advantageous as they will see the extra functions of the site and perhaps decide to submit a report if they have one to make.

There are some disadvantages of the site, though, as well. For example, Libya Crisis Map contains a 'Download Reports' tab. I think we should have this. Then we have another avenue for sharing our data. Second, the website is in English only. Phnom Penh citizenry are more likely to use it, and enjoy it, were there Khmer script. Third, as mentioned, we have not set up a number for submitting reports via SMS due to privacy concerns. These disadvantages may be resolved in the next phase of development, with a priority on making a Khmer version of the site.

Now we've got the tool working, we need to put data on the site for people to look at. This has been on the backburner, though, due to my focus on achieving another milestone. This was done last night with the completion of the five eviction analysis maps. These have been a major source of frustration over the past seven weeks. I have learned a lot about cartography in ArcMap though, so that's a plus. I think, also, that the maps may have been pushing the boundaries of what the program is meant for, and possibly what my computer is meant for. One of the maps had 8 data frames. At least twice, one of the data frames just stopped working (trying to bring up its properties would crash ArcMap), so I just had to copy all of its layers to a new data frame. One particularly frustrating thing was the leaders going from the extent indicators to the inset data frames. These were obscuring labels and points in the path between the extent indicator and the inset. Even replacing them with lines from the drawing toolbar caused the same problem, no matter how many levels I sent them back. So what I found I had to do was put everything that I wanted the leaders to go underneath in another data frame, which had exactly the same scale, extent and size as the original. This new one sat in front, with the extent indicators and leaders in the original data frame beneath. Voi la, the points and their labels were no longer obscured.

Anyway, so it's good to have the maps looking coherent. Perhaps I am a little to keen to say I've achieved this milestone, though, since there is one final thing to do. There are four more to make, one for each of the inner districts of PP. These I thought I'd finished off last night, using data driven pages to shift automatically between the districts and update the subtitle of the map. Only thing is, I did not know that converting labels to annotations in a geodatabase removes the ability to dynamically scale the labels, so they looked far too large when it got to Pram Bpi Makara, which is a small district. This is because I converted them to annotation at a much smaller scale (which means a map covering a larger extent - to understand scale think of the size of the objects in the map, large scale must mean the map covers a small area as the objects, like houses, are large in the map). So I can say that I have finished all 9 maps, only I need to go back and convert the labels to annotation in the map itself, set the annotation group to zero reference scale, place the labels where I want them and then I'll be done. Fingers crossed. Then it's off to get them published!! Will be interesting to see the response. I will just have made 9 maps about one of the most hot button topics in Phnom Penh.

Last of all I should mention the Inaugural Phnom Penh GIS Meetup (map)! This was a huge success! I was so happy at the end of the night. I knew it would all work out when the first two people that showed up I'd never met before. They were both from Aruna Technology, Phnom Penh's biggest spatial science firm. One of them was even Khmer, which was a big win. In all, 10 people showed up and we discussed everything from our Urban Voice webmap, Mango Maps (a simple webmapping tool just developed by Aruna), the application of GIS in research on flooding in Phnom Penh, crowdsourcing GIS, the Land, Management and Administration Project (for systematic land registration and titling of most plots in Cambodia), a project surveying the boundaries of Cambodia currently underway, another project collecting the locations of all hotels across Cambodia... and many more discussions not at my side of the table. We even had someone looking for a GIS job and got him some contacts. It was nice also to receive congratulations on organising it from the technical director at Aruna, who said he'd tried to host a similar thing but it was not so successful... We'll be doing it again.

Making Trouble in Phnom Penh

Discussions have, inevitably, begun about Doing A Big Motorcycle Trip. Hong Kong has even been raised. We'll see. The inaugural ride of the Phnom Penh Crash Test Motorcycle Club was ridden. There were two members, Jim 2 and I. We went to the Bassac River in Kandal Province then back up Norodom Bvd to circle around the remnants of Boeung Kak. First time I've been there at night. Caught glimpses of the expanse of sand it now is, although we didn't stop as it was a pretty unsafe area. Obviously there was a risk we could have been targeted by a rival motorcycle gang. I presume such other gangs will be more wary of us now though, given that my bike scored 106 points on Rideit though this fortnight. We also did recruitment for the club at the appropriately named Garage, on St 108 near the Night Market on the riverside. Motorcycles were tested. Cigars were smoked. Girls were impressed. A story was begun.