digiKam & KIPI coding sprint 2009. Photo by Johannes Wienke.

I just came back from the digiKam / KDE imaging coding sprint in Essen, Germany. My main goal was to start working on an export plugin to Wikimedia Commons from digiKam, which can be a wonderful tool for Wikimedia photographers.

I met a lot of great people there, and it was really nice to benefit from their experience. I learnt a lot during these three days, and being in the same room as other developers made it a lot easier to work together.

By analyzing how the existing Facebook KIPI plugin was made, I was able to create the basis of the communication layer between the plugin and a MediaWiki site, even though I had not done any C++/Qt development before. Admittedly, I had some existing knowledge of Lisp, Java and C, but it was old and limited; I was quite happy to advance so quickly. While my current work focuses on a specific wiki (Wikimedia Commons), it could be later extended to a more general MediaWiki library for Qt.

Special kudos and Vielen Dank to Marcel Wiesweg, who organized the whole event. Everything went great, we had a pretty good Internet connection and we even ate well! (one thing you wouldn’t necessarily expect during a coding event). The trip back was also fun.

From your computer to Commons using digiKam

For several years now, I have been a photographer for Wikimedia Commons. Commons is actually the reason why I became interested in photography, and why I bought a DSLR camera and various lenses. And I have ended up taking a lot of pictures.

The problem when you take a lot of pictures is that the time you need then to process and upload them grows exponentially. Besides, if you want to upload your pictures to several platforms (say, Commons, flickr and your blog), you have to deal with the specifics of each website. It would be much simpler and faster to describe, tag, classify, geotag the pictures only once (using an appropriate batch-editing tool) and then to have each website extract the metadata, or to use an upload tool that extracts them automatically.

Wikimedia Commons

So, on the one hand, you have Wikimedia Commons, a repository of freely-licensed multimedia files (photos, maps, diagrams, sounds, videos, etc.). It is a project of the Wikimedia Foundation, from which uploaded files can be used across all Wikimedia projects in all languages, including the famous encyclopedia Wikipedia.

Wikimedia Commons contains over 5 million files, the overwhelming majority of which have been contributed by volunteers. Such a large amount of files makes it mandatory to properly:

• describe what each work is about
• identify the author and the copyright status of each work
• classify each work in order to be able to find it later and to manage such a large collection.

Besides these requirements, it is also considered a good practice to geotag your pictures (if applicable), i.e. to indicate the geographical location of where the picture was taken.

digiKam

On the other hand, there is digiKam, a FLOSS “digital asset management” application whose special features allows to annotate, classify and catalogue a large media library, although it provides some very handy editing and post-processing tools as well. digiKam uses the KIPI plugins, which provide handy features such as batch editing or export modules to a variety of websites, including Flickr, Picasa and Facebook. The 4th beta release of digiKam 1.0 was recently released and the official 1.0 release is planned for late October this year.

I won’t go through digiKam’s impressive list of features; instead, I will focus on those that make digiKam a very powerful tool for Wikimedia photographers.

The perfect combination?

So let’s take a quick look at the features digiKam offers that would be handy for people who want to contribute their pictures to Wikimedia Commons:

So, what’s missing? Well, you can do a lot of stuff with digiKam to prepare your pictures before they’re uploaded to Commons. Yet, you still have to upload them manually, or using an external tool such as Commonist. And even if you have been spending days to annotate your pictures, describe them in several languages, geotag them, etc., well, all this information remains in the image metadata. Neither Commonist nor MediaWiki knows how to extract these metadata meaningfully (yet); as a consequence, you have to redo all the work when uploading each file to Wikimedia Commons. The Ford Foundation recently awarded a $300,000 grant to the Wikimedia Foundation in order to improve the usability of the upload process. I expect some work will be done regarding the extraction of metadata from files. Yet, when uploading a large number of files, it would be more convenient to provide a standalone desktop upload tool like Commonist. Or…

What would really be neat is an integrated export module to Wikimedia Commons in digiKam; not only would it serve as a batch upload tool, but it would also read all the metadata and create the file description pages accordingly. A few weeks ago, I contacted the main developer of digiKam in order to discuss the possibility of implementing such an export module. He showed interest in this project even if he didn’t have the time to code it himself. As a consequence, I have initiated a roadmap and brainstorming page for the plugin.

Feel free to contact me if you’re interested in helping implement such a plugin.

Notes