Jul 142010
 

Hello – and welcome to my first post!

I’d like to let everyone know how things are progressing with the software side of AddressingHistory.  I’ve been working on all aspects of AddressingHistory, from the database (at the back-end), storing information from the Post Office Directories, to the public-facing webpages at the front-end.

A large part of the challenge so far has been to take raw text from the Post Office Directories and turn it in to useful, structured data. This is necessary before you, our future users, can search through it – and add your own data!  I’ve created software that parses the Post Office Directory text, extracts the useful information and loads it in to a spatial database (a database with special features to manage geography). For those who are interested, all the software I’ve written is made with Java, using Spring MCV, runs on Apache Tomcat and the database is PostgreSQL with PostGIS extensions.

I’ve written software which allows easy access to the organised, structured information from the Post Office Directories. It’s known as middleware, or an API.

There is a development version of the API available here, where you can change the parameters to search for your own surname, or address:

http://devel.edina.ac.uk:8082/ah/ws/search?surname=Alexander

http://devel.edina.ac.uk:8082/ah/ws/search?profession=baker

You can also search for addressess (using an ‘address=’ parameter) and perform spatial searches on specific areas.  Results are returned in plain-text (comma separated format) or, by default, in JavaScript Object Notation (JSON).

I’ve also been experimenting with Google’s Geocoding API, with some success!  After extracting the address text from each entry in the directories, I send a query to Google’s mapping service in much the same way as you’re probably used to using Google Maps.  It looks as though we can get accurate locations (a process known as ‘geocoding’) for the majority of entries in the Post Office directories.  That will mean we will be asking you to help us locate the small percentage of addresses we cannot automatically geocode, and to help us make sure what we have coded automatically is in the right place. Once we have the coordinates of each entry, they can be shown on a map – and be used to search for results.

So combining the data loaded in the database, the web service to request entries using specific search terms – and the newly geocoded data, we’re able to make some quite interesting maps. For example, this map (unfortunately shown on a modern map of the city for now) is a quick look at the location of some of the bakers, yes, bakers, in Edinburgh, in 1905.

2

And here, are all the people with a surname of ‘Alexander’…

1

Thanks for reading – there’ll be more soon!

– Joe.

Welcome to the AddressingHistory Blog

 Project Updates, Site Admin  Comments Off on Welcome to the AddressingHistory Blog
Apr 012010
 

Welcome to the AddressingHistory blog!

We will be using this space to let you know about ongoing developments around the AddressingHistory project as we work, over the next few months, to build on online tool for letting you explore and georeference historical post office directory information from Edinburgh. This is information like the names associated with a particular address, directories of professions, and adverts for businesses.

You can find out more about the project on our About page and we would welcome your comments, suggestions and feedback on the project and updates to the blog. You can also become our fan on Facebook or follow us on Twitter.

If you would like to know more about the project or have suggestions for the project team please drop us an email to: addressing.history@ed.ac.uk