Here it is:

I really enjoyed Ignite Dublin 8 last Thursday. There was a huge variety of talks, and I was lucky not to be following hilarious presentations by Shirley Temple-Bar and David McKeown.

This was Conor Houghton‘s last time at the helm of Ignite Dublin and I wish him well in his future endeavours in Bristol, he’ll be missed here in Dublin for sure.

I mentioned previously that the talks will be posted on the Ignite Dublin YouTube channel, I’ll update this post when mine goes online, which will be in a few weeks. But in the meantime my slidedeck is below, you may want to click through to see the speaker notes.

Now to plan my next Ignite talk…

I’m pretty exited by my next speaking appointment, Ignite Dublin 8, which takes place in The Science Gallery on Pearse St in Dublin next Thursday 8th December. Each Ignite presentation is recorded and posted on YouTube afterwards, I’ll post mine when it’s online, I’m busy writing and re-writing it now.

If you’re not familiar with the Ignite concept, the idea is that each speaker has only 5 minutes and has 20 slides which auto-advance every 15 seconds. This keeps things short and sweet and ensures a wide range of topics can be fitted into each Ignite session – there will be 18 speakers at Ignite Dublin 8, talking about things ranging from Archaeology to Robotics. And Maps, of course!

I gave a presentation on ‘The Do’s and Don’ts of Irish Addressing’ and my slide deck is now online on SlideShare here:

There will shortly be even more open data to work with, in Dublin at least, when the DubLinked initiative launches in September.

The project I chose to work on was led by Jane Ruffino, and focussed on opening, tagging and imbuing with context the vast amounts of Archaeological data that are currently locked away in reports and dusty museum stores. The interactive app is aimed at people who are interested in the heritage of their city, and Jane is hoping to develop the concept further via the NDRC LaunchPad programme, and the very best of luck to her.

One of the tasks I set myself during the challenge was to grab some of the publically available data from archaeology.ie and make it available in a more app friendly manner.  To do this I used Google Fusion Tables, and this is how I did it.

Before you start you’ll need a Google account.  If you use gmail or other google services you probably have one already, but if not you’ll need to set one up to gain access to Fusion Tables.

That noted, the first step is to go to archaeology.ie, click through to the Map Viewer and then click on the Tools icon and choose Download data.  You can filter your download by Class or by County. You should download the data in Shapefile format:

Once you have downloaded the data you need to upload it again.  A guy by the name of Josh Livni has built an online service which takes .zipped up shapefiles, reformats them to Google KML and uploads them to your Fusion Tables.  Simple and rather brilliant.  The service is called ‘Shape to FT’ and is available at http://www.shpescape.com/

You will need to grant the service access to your Google Account for it to work it’s magic. Simply browse to the .zip file you downloaded earlier and hit upload.  the import will take a minute or two.  Once it’s finished just click on the Fusion Table link and you’ll be able to see your data in tabular form in Fusion Tables. To view it in Map form, simply click on ‘Visualise > Map’ From there you can choose to share the table, or even embed it in another site, like this:

There are plenty of other options, including customising the info boxes, using color-coded symbols and classifications etc. but I’ll leave those as an exercise for the reader.

This lecture covers a lot more ground than the previous two, as I’ll be talking a lot about Open Source Geo Tools as well as covering Open GeoData.

Here’s the slide deck:

My employer specialises in this kind of work and has built a number of systems to partly automate the task and verify the accuracy of the results.  It’s pretty impressive stuff and I hope to write about it in more detail in the near future.

Today I ran a public dataset of some interest through the system – the locations of the 84 properties auctioned at last week’s high profile sale in The Shelbourne.

An Excel spreadsheet of the data is available from here and a Google Map/Earth compatible .kml file can be downloaded from here.

Here’s the map, you can click on the pins to reveal underlying data, such as the reserve price and price achieved:

View Larger Map

As is always the case in dealing with Irish addresses, there may be some error in the precise location of these points.  If you want to know more about the process, or if you’re interested in visualising your address data I’d be delighted to hear from you.

Next Sunday, April 10th, is Census day here in Ireland.  There has been much debate in the media and online over some of the questions asked on the census, particularly those relating to religion and partnership status.  The raising of these two issues in particular could be considered to be a reflection of social change in Ireland, something which the census itself sets out to measure, even if it does so in a way which many consider to be flawed.

Trying to accurately reflect these changes while maintaining links back to previous censuses dating back to 1901 and beyond is a very difficult problem and one which I feel that the CSO has done as well as it can to solve.

I am a GIS specialist who makes use of census data on an almost daily basis. I have written before about some of the kinds of analysis that can be done with the anonymised data which the CSO makes available to researchers.  I think it is fair to say that the census is the most important dataset which people working with Geographic Information in Ireland have available to them.  The fact that the results are made available at no charge to the end user is admirable and long may it continue.

There has been some debate, particularly in the UK, as to whether a census is needed at all.  States such as Denmark are frequently mentioned, as they did away with the manual census process used here in favour of a register based system back in the 1970′s.  A register based system is where different Government departments share their databases and citizens are obliged to inform the government of any change of residence, employment etc.  Some of the required registers are in place in Ireland, notably the General Register Office and GeoDirectory.  But I have my doubts that a Register based census could be established here given our current data protection legislation (among other issues)

What excites me most about Census 2011 is the new ‘Small Areas’ (PDF link) which are the smallest spatial unit for which data will be made available.  This will allow for a much more detailed view of all aspects of our society.  How so?  Well consider the case of the current smallest spatial unit used by the CSO since the early 1980′s – the District Electoral Division (DED).

There are approximately 3,400 DED’s covering the country, and they range in population size from 16 people (Arigna in Leitrim and Lackagh in Tipperary) up to 32,305 in the case of Blanchardstown-Blakestown.  That variation alone is a major problem and makes for skewed results if extreme care is not taken when doing any analysis of census data.  The new Small Areas, of which there are over 19,000, are nested within the DED’s for backwards compatibility, and have been designed to follow urban geometry and natural features.

Here’s an image of the Blanchardstown-Blakestown DED (in blue) with the new Small Areas shown in red:

Blanchardstown-Blakestown DED and Small Areas - Data Copyright OSi and Microsoft Bing

There are 119 Small Areas in the Blanchardstown-Blakestown DED alone, which will allow for much more insightful analysis of the population there – meaning more efficient and effective delivery of services, among a myriad of other benefits.

There has been much speculation that these small areas will form the basis of our new Postcode system (I’ve written about this before too, and I’m not expecting any system to be launched soon), but that remains to be seen.

It will be 18 months to 2 years before we see the census data released at the small area level, but I’m looking forward to being able to map our population in a never before seen level of detail.

The presentation is a refresh of the one I gave back in December, and it is embedded below.

I hope you find this as interesting to browse as I did putting it together.  I’d be delighted to give more lectures on this topic in future, please get in touch if you know of a suitable event.

It was a very interesting couple of years. I worked with some great clients and on some very interesting projects, but I have a roof to keep over my family’s head and it just wasn’t paying the bills.

I fully intend to keep up with the blogging on here, with less focus on professional promotion and more on my thoughts of what’s going on in the wider geo-sphere. Of course these views will be my own and shouldn’t be interpreted as being those of my employer.

The first of the changes to this blog might be apparent to frequent readers – I’ve imported some posts from my (slightly less professionally focussed) Posterous site to here, you can see them in the From Posterous category tab.

I’m facing a steep but enjoyable learning curve over the next few weeks as I get to grips with new software platforms, databases, client requirements and workflows. But I hope to be back with an end-of-year round-up before the month is out.