The KDE Social Desktop's first appearance

The KDE Social Desktop was first proposed in Frank Karlitschek's keynote speech at the 2008 Akademy in Belgium. However, it has only recently received widespread attention with the announcement of its first step: a desktop plasmoid scheduled for inclusion in KDE 4.3. The resulting publicity has left as many online commenters praising the concept as criticizing it, with both sides having the potential to improve subsequent development on the project.

The KDE Social Desktop should not be confused with the countless other efforts on every platform to integrate general social network tools more tightly into the desktop. Nor should it be confused with the semantic desktop, the ongoing efforts to add a data layer based on Nepomuk to the KDE desktop for annotating and tracking information. Instead, the Social Desktop is specifically an effort to bring the advantages of the KDE community to the desktop.

As described in the PDF of Karlitschek's slideshow from the 2008 Akademy, the Social Desktop began with his observation that the KDE project has a long history of community-oriented sites such as the now-discontinued KDE-Look.org and the ongoing KDE-Apps.org. He particularly emphasized his own highly-successful meta-portal openDesktop.org, which, now receives some 60 million page impressions per month, and boasts over 100,000 users.

Yet, despite this ability to create active communities, Karlitschek noted, KDE has only a fraction of the general desktop market. Arguing that this organized community is something that Windows lacks, he proposed increasing KDE's market share by making these communities accessible directly from the desktop with what he calls "Open Collaboration Services". When finished, these services will include such elements as links to developers in the About window of applications, and others to allow users to become fans of an application and follow its development. Similarly, online help could include links to people with specific problems, and hardware dialogs could include links to those with the same hardware. In the same way, to "welcome new users into the family," as Karlitschek put it, an applet could show nearby KDE users and events in order to help newbies join the community more easily.

The recently announced openDesktop plasmoid is a simple tool in itself, more a proof of concept than anything astonishingly different. It is a client for the Open Collaboration Service API being developed by Kügler. It also draws as necessary upon a new geo-location engine that can use either a GPS device connected to the computer, or estimate location based on IP Address.

You can download and compile the source code for the plasmoid and the engines it depends upon, but, if you are simply curious, you will learn almost as much with far less effort by watching the Flash video by Sebastian Kügler that was part of the announcement.

To use the plasmoid, you must register with openDesktop.org. It opens on your openDesktop.org user page, and other tabs show friends and nearby users who are currently logged into openDesktop.org, any of whom you can contact. In the setup, you can also decided whether to publish your location, or preserve your privacy by not revealing it. All in all, the plasmoid resembles nothing so much as a dedicated IRC application. However, as limited as it is, the plasmoid marks the accomplishment of the first stage of Karliltschek's plans.

In a comment that accompanies the announcement, Kügler wrote, "The goal is not to write a desktop Facebook client, and I'm not aware of anyone working on this right now. That doesn't mean that it wouldn't be welcome (it certainly is if that's what you want to work on). I'm aware that DigiKam does integrate with Facebook to upload photos."

However, Karliltschek emphasizes that the effort is intended to create a specifically KDE community. "Please see this only as a first step," he wrote, suggesting that the social desktop could eventually be used for "Free Software Events, Knowledge Base and user Support, document sharing, location-based features and more. We try to create something really new and innovative here."

The next step, if things go according to Karliltschek's original presentation will be to encourage KDE sub-projects like Akonadi, the KDE Personal Information Manager, and Decibel, the desktop communications framework, to add support for the Social Desktop to their back ends. That will be followed by integration of the Social Desktop into specific KDE applications. However, if the slowness of integration for Nepomuk is any indication, these steps may take several releases to accomplish, assuming that they succeed at all. Probably, they will be as much a matter of diplomacy as the Social Desktop developers persuade others of the usefulness of their project.

Reactions to the Social Desktop

At least half the reactions to the announcement of the Social Desktop are positive, but offer little except encouragement for the concept. For observers of the free desktop's evolution, the questions poised by the negative comments seem more thoughtful, regardless of their validity.

For example, when the announcement was linked on Slashdot, a poster called speedtux questioned the need for the project. "The 'social desktop' is already here," speedtux wrote. "It consists of web sites, site specific browsers, instant messenger apps, feed readers, desktop notification, and widgets. Some people also still use local mail, calendar, and address book apps. What is KDE trying to contribute to that? Even more heavy-weight local apps and new protocols? How are they going to keep up with the rapidly evolving set of protocols and features available through web apps? And why bother?"

In another comment on the Slashdot discussion, the same poster compared the project to the long-retired CDE graphical interface, arguing that "KDE is repeating the CDE mistake: instead of focusing on what people need right now and doing a really good job at it, KDE is trying to realize some long term pie-in-the-sky technical visions of its developers that no user asked for."

However, perhaps the most fruitful reaction appeared on the KDE Plasma-devel mailing list, where Richard Dale pointed out that at least some of the concepts being developed for the Social Desktop overlapped with ones that already existed in Nepomuk's semantic desktop. In particular, Dale suggested that a number of ontologies (that is, higher level organizational concepts for information) such as FOAF (Friend of a Friend) might be as applicable to the social desktop as to the semantic one.

To his credit, Karlitschek promptly welcomed the idea of working with the semantic desktop, a move that might ultimately reduce the amount of new code that the Social Desktop adds to KDE, as well as saving development time. Perhaps, too, it will prove easier to to have both the Social Desktop and the semantic one accepted together, rather than separately.

Is the Social Desktop needed? Could it be what KDE needs to increase its market share? These questions cannot be answered yet, largely because the details of where the KDE Social Desktop will go from here is still unannounced.

However, it may be that the next stages of the project will be stronger for both the enthusiasm and the probing criticism. The enthusiasm may expand the concept into general social networking, while the criticism, if nothing else, will remind the project of the need to allow users to opt out and suggest further collaboration possibilities. In the end, the Social Desktop may be more successful because of its implementation in stages.