Photos:

Track:

You can see a live demo running on my Moodle installation.

Any feedback or comments welcome

Online Users Map with OpenStreetMap

I’ve finally had a bit of time to work on the online users map block I wrote a while ago – the recent release of Moodle 2 being a bit of a spur. Two main changes I’ve made:

Firstly, I added the option to use either Google Maps or OpenStreetMap, with the default being OpenStreetMap, but you can easily change it in the block settings.

Secondly, I managed to get the block up and running in Moodle 2 (see it in action). It needed a few changes over the previous version to get it to work. I’m still having a few issues with getting the block to update the cron field in the mdl_blocks table – so currently you need to update the blocks table manually to enable the cron function, which automatically updates the user locations.

Any feedback welcome

I recently became the owner of an unlocked HTC Dream smartphone (running Android 1.6). Smart phones are still quite a rarity in Mekelle (and I’d guess in much of the rest of Ethiopia), so despite this not being the most recent model, everyone who sees me using it asks me to have a look & play around. I have seen a few people with Nokia E71 phones, but when you look closer they’re actually Nokla E71′s (yes, that’s Nokia with an L instead of an i).

In a couple of days I will be heading out to some rural areas with a colleague doing his doctorate in public health. He’s testing different smartphones and applications for data collection whilst he’s interviewing Health Extension Workers (HEWs). I’m joining him to see what some of the issues are with using these types of phones and applications in this context, with a view to spending some time over the coming months seeing how these devices may be used to deliver training.

I’ve only really been using the phone for the past week or so and there are a couple of areas where I can already see we may run into problems.

Firstly, the battery life. With my usage, not particularly heavy, the battery usually only lasts just over a day. Given that we’ll be using these devices for data collection, then they’re likely to be having heavy use in areas with little or no mains power. We are testing out some small solar power chargers.

Secondly, the GPRS coverage. GPRS is not used widely here and coverage in extremely patchy (even in large city like Mekelle) and it’s not yet been rolled out to other more rural areas (or even large towns). Sim cards need to be specifically enabled to use GPRS – it’s not turned on by default. The applications we’re testing out (EpiSurveyor and Sana) will both allow data to stored until an area with coverage is reached, but unless the user visits Mekelle on a regular basis then the data will never get uploaded.

I’m sure that improvements in the phones and the phone network infrastructure will eventually make both of my concerns invalid – it’s just a question of when they will be addressed.

The other questions and areas I’d like to look at include:

1) How easy is inputting the data on such a small screen? Might a tablet or netbook PC be more appropriate? Perhaps they’ll work well for short, relatively simple surveys, but not for others?
2) Do any of the HEW’s already have java enabled phones? If so, this would enable them to use the EpiSurveyor application without any new phones.
3) Do any of the phones support input using ge’ez (the alphabet used for Amharic and Tigrinian)? I can’t see how to input these characters on my phone (if anyone knows how I’d be pleased to hear from you), but I can display the characters.
4) How long do the phones take to get a GPS signal? For each record input we can automatically attach the location coordinates – but I’ve noticed that sometimes the phones can take a long time getting a GPS fix. With the power issues it’s unlikely they’d want to leave the GPS on all the time.
5) Would they really be used? Getting reliable data in these areas (even just for the number of births/deaths) is extremely difficult – reporting processes are often unreliable or just not used. Using these phones could help with gathering this info – but obviously only if they are used.
5) What are the other uses for the phones? E.g. providing remote diagnostic support, clinical support, training content/activities or reference, or perhaps for fun/social activities.

Plus I’m sure many other questions and possibilities will arise over the coming days.

Heatmap with weighting applied

Following up on the playing around I was having with heatmaps the other day, I made a few updates to the heatmap.py script to allow for each point to have an associated ‘intensity’. When the points are plotted, the intensities are normalised and each point plotted with a relative intensity (as shown in the image on the right).
Few other little changes I made:

• allow the area covered by the size of the final image to be specified, rather than using the max & min x/y coordinates from the input points
• change how the the dots are built – in theory should be a little quicker – but I’ve not tested with large enough dataset to know if it makes a big difference
• in my demo script I’ve shown how you can convert the latatitude to its Mercartor projection coordinate

For those interested in the changes I made to the heatmap.py script you can download the diff and an example script. Any feedback, comments welcome

My 'hello world' heatmap

Over that past few days I’ve been playing around with creating heatmaps to overlay onto OpenStreetMap – the image on the left is where I’ve got to so far. My interest in this is for creating heatmaps from data collected during my colleagues health sciences research in Ethiopia.

There are already several programs and services available to create heatmaps, for example gheat and OpenHeatMap. But none of these quite suited me, gheat because I didn’t want to create a full tile server – just an image to overlay on a particular area of the map and OpenHeatMap because I wanted to have access to the code to tweak how I wanted.

I then found heatmap.py which (using a similar algorithm to gheat) did almost exactly what I wanted. There were only a couple of changes that I needed to make:

1) Update to account for the Mercator projection – I wanted my overlay to be on a view of the whole world (zoom level 2 in OpenStreetMap). When I first ran the program the areas over northern Europe (and others) were almost but not quite inline with the marker overlays, but this was due to the projection. I just edited the input script to convert my lat/lng coordinates into Mercator coordinates. I also hooked up the heatmap.py script to read the lat/lng coordinated from a MySQL database.

2) Allow the script to have weights against each point. The current script looks at the number of points in a particular area (or on top of each other) to generate the ‘heat’. I also wanted to allow for points to have a weight – as mentioned here. I’ve not yet implemented the weighting, but I’m not anticipating this to be too tricky. GHeat (as far as I can tell) doesn’t allow for weights on points, but OpenHeatMap does (please let me know if I’m wrong about this).

All has been much easier than I’d expected, I’ve learnt a little more about python and once I’ve got the weighting working how I’d like, then I’ll share the code back.

Visiting a school near Wukro

Yesterday, I had a visit out to Wukro, to meet some the heads from the local health bureaus and visit some schools in the area. Two of the PhD students from Mekelle have projects in this area, looking at prevalence of parasites in children and how this may be reduced. We’re investigating using phones and GPS to collect the survey data, and have it automatically sent to a server (via GPRS) when the data collector comes into a mobile reception area. We’re also looking at how learning materials and/or diagnostic tools could be made available electronically (again using phones) to the Health Extension Workers.

We visited two schools, both a fairly short distance (10-15km) from the main Mekelle to Adigrat road, but both feeling quite remote. Both were primary schools with around 1000 pupils each – but neither had access to clean water and one had no electricity – although both did have mobile phone reception.

Getting connected to GPRS ought to be relatively straightforward, however the SIM card needs to be GPRS enabled first. It’s free to enable, but isn’t automatic. We went to the ETC office in Mekelle to ask them to do this, but for some reason currently it can’t be enabled for Tigray mobile numbers (0914xxxxxx), I then asked them to enable my Addis mobile number (0910xxxxxx), but was then told it can only be enabled in Addis. So we’re a little stuck, being unable to test the GPRS availability on any of the phones that we have between us. The good news is that SIM cards are now much cheaper – 85 birr each (with 15 birr credit included), compared to the 360+ birr I paid less than a couple of years ago.

The past month has flown by, feels as if I’ve only be back for 5 minutes. I’m generally very happy with the progress we’ve made on the project over the past year. Looking forward to coming back again in September again to see how everything has moved on again.

UK map generated on my OSM server

After using OpenStreetMap in a fairly limited way for the past year or so, essentially just to upload data/tracks from wandering around Ethiopia, I thought I’d have a closer look, especially as a possible replacement for Google Maps which I was using a lot whilst at the OU.

Firstly I looked at how I could use OSM and the OpenLayers API as an alternative for the map in my Online Users Map – which certainly looks feasible. So I’m now thinking about making this an option in the block settings, as shouldn’t be too much work to provide the option.

I also set up my own open street map server (instructions for installing on Ubuntu Lucid Lynx). Although quite a long process everything worked well first time with no errors. I didn’t download the entire 8Gb+ OSM dataset to generate my tiles, rather I just selected a much smaller area of detailed data (using extracts from GeoFabrik and CloudMade). This avoided the potential ’30 hours or longer’ process of importing the world into my database – the whole dataset for Ethiopia took less than 30 seconds to import.

The only part where I ran into a problem was at the very end when I wanted to generate all the tiles. When I ran ./generate_tiles.py I kept getting an error telling me that osm-local.xml didn’t exist. Fortunately this was quite an easy fix, I just needed to edit ~/bin/mapnik/generate_tiles.py to point to ~/bin/mapnik/osm.xml instead, then all ran fine (generating 55Mb+ of png images).

I could then create an alias in my Apache to point to my generated tiles and create a slippy map using OpenLayer and the tiles generated on my desktop.

I still have a lot to learn about all this, especially with what can be done with Mapnik for generating tiles (and the python script necessary to achieve it) and using OpenLayers as a replacement for Google Maps API – but all going well so far

It turned out to be a relatively simple issue to resolve, basically I wasn’t correctly constructing the call to the geonames website, so it gave the ‘bad gateway’ response.

If you are using (or trying to use) this block then please update to the latest version as soon as possible (go to: Moodle Modules and Plugins).

I’ve not yet had chance to test the changes when the Moodle server is behind a proxy server, I’ll try to get this tested soon. If you are using a proxy server I’d be very interested to know if it’s now working (or not).

[Update 20 Apr 12:30]: have just tested by going through a proxy server and appears to be working fine, but please let me know if it doesn’t work for you.

In an attempt to avoid getting stopped quite so often I thought it’d be better to go out on my bike. This mainly worked, though there we a few times when I was almost knocked off by groups of kids trying to stop me to give them money! Fortunately I was going quite slowly and didn’t actually get knocked off. Out on the bike I was able to do some more mapping for OpenStreetMap, adding some roads that I don’t usually walk along. After uploading these updates for OpenStreetMap I took a look at Mekelle on Google Maps. Last time I looked, several months ago now, there was virtually nothing marked, but now there’s almost a complete map. I’m interested to know where they got this data from, given that most of the printed maps I’ve seen of this region aren’t great and certainly don’t include many of the new residential areas on the outskirts of town.

Yesterday I also went to the cinema for the first time I’ve been back, to watch Teza, an Ethiopian film about a doctor returning to his home village near Lake Tana during the 70′s and 80′s. Although depressing in parts, it was well worth watching – the film won best film at an African film festival earlier this year. I had expected it to mainly be in Amharic, hopefully with English subtitles, but it was actually almost all in English. For info, the word ‘teza’ roughly translates as drop or drip.

If anyone feels like posting something over to me then anything, postcards, packet cheese sauce, etc, it would be much appreciated… the address is:

PO Box 3060
Mekelle University
Mekelle
Tigray
Ethiopia

Cheers!