Previous edition of online WALS
Tuesday, May 10th, 2011The previous edition of the online WALS, the 2008 edition, is available at http://2008.wals.info/.
The previous edition of the online WALS, the 2008 edition, is available at http://2008.wals.info/.
The previous edition of the online WALS, the 2008 edition, is available at http://2008.wals.info/
Over the next couple of days we will push WALS 2011 – the new edition of WALS Online – live! While this should mean more chapters, more data and a better user experience in general, we also expect a transition period with a couple of glitches and hiccups while we reassemble the pieces of the web application. Should you encounter any of these, let us know!
Once this transition period is over, we will detail the changes in a separate post.
Since we introduced the datapoint specific comments functionality, quite a few new posts to this blog have been created – by exactly this functionality. It turns out that most people just click the “comment” buttons out of curiosity, without actually commenting on a datapoint. While this shouldn’t come as a surprise, it is somewhat unintended and renders the feeds for the blog less useful, because they get cluttered with entries for these automatically created posts.
A solution for this problem may be to subscribe to more selective feeds for the blog. The feed URLs for single categories look like this http://blog.wals.info/category/category-name/feed, e.g. the feed for the “News” category is http://blog.wals.info/category/news/feed
If the rate at which new posts are created due to the new functionality increases notably, we may have to reconsider the current implementation, though. But at less than 10 new posts per day, we’ll keep it as it is, and apologize for any inconvenience.
WALS Online is about making the data of the World Atlas of Language Structures accessible – as widely as possible.
To most users the site’s HTML pages will be all the access to the data they care about, but for people who want to work with the data quantitatively a more comprehensive access to the data is necessary.
The Raw Power
For those wanting to unleash the raw power of SQL, we provide the WALS Online SQLite database – the one the web application runs on – for download.
The easiest way we found to work with this database is to use SQLite Manager, a Firefox browser add-on.
The screenshot below shows SQLite Manager with the WALS db loaded, and languages located north of the polar circle – i.e. with a latitude greater than 66.5° – selected.

Power to the Masses
Now SQL isn’t everyone’s first choice for manipulating data; Spreadsheet processors like ms excel or openoffice calc may come closer to that. So since last week, we also provide (most of) the WALS data in a ZIP archive of Delimiter-separated values for download.
Data formatted this way can be easily imported into spreadsheet processors, just make sure to pick the correct character encoding for your platform.
And to the Mapmakers
To include the feature data from WALS in maps, we provide also several exports, explained below in descending order of flexibility.
Single Feature Values as GeoRSS
As announced before, GeoRSS for single values of features are available using the following (somewhat bolted-on) URL syntax. To retrieve the feed for the first value of feature 2:
http://wals.info/feature/2?tg_format=georss
To retrieve other values, you must pass parameters to set the preceding values to “invisible”. So the URL below will give the feed for the third value:
http://wals.info/feature/2?tg_format=georss&v1=a000&v2=a000
GeoRSS is easy to include as layer on maps created with OpenLayers.
Complete Features as KML
With URLs like
http://wals.info/feature/2?tg_format=kml
you can export a feature’s datapoints in KML format, suitable for import in Google Earth.
Features as Mapplets
To add a WALS feature as overlay to a Google map, you can install the mapplet by following the “mapplet” link on the feature’s map page.
To keep track of the usage of the site, we set up a – rather crude – web server log file analysis to be run once a week. You can find the results here. (Note: To keep robots out, the pages require you to authenticate. You may use the account info given in the popup.)