There’s an article at osViews about trying to get together a project to develop a full fledged user desktop package. By this they mean everything that a local computer retailer would need in order to sell full fledged Linux based systems, including artwork and sales and marketing materials. I think this is a great idea. Both in terms of getting together a standard system for use by mere mortals wanting to use Linux, and in pooling together resources by many individual retailers in order to get the project completed.

What concerns me however, is the structure of the proposed project. They’re talking about running this like a contest. The potential sellers of Linux based desktops would provide cash and other resources to be used as prizes for those wishing to submit entries into three different areas. I actually don’t think this is a good idea. The model that I personally think would be more likely to work is to have those interested in seeing a project like this done to directly create work to be distributed under something like a Creative Commons deed.

My reasoning for running this way is because it seems like the closer the contribution is to the user of the system, the more likely the open source model will work. Put another way, the more abstraction there is between user and producer, the worse off the project is. Therefor, have the users of the work directly contribute content they would like to see made part of the project. This doesn’t mean that they have to create the work directly themselves, they can always contract a designer to do the work, pay for it directly, and then put it under Creative Commons and publicly release it. It also ensures that those producing the work without being guaranteed return (the graphic designers, video editors, actors, and whoever else would have to be included) don’t end up wasting their time. If run like a contest, if a graphic designer makes an entry that doesn’t place at all, they end up getting no reward. Whereas a producer of a Linux Desktop does get a reward, the project produces a desktop system either way. Since the vendor gets benefit no matter what the outcome, it makes the most sense that they should directly bear the cost for the time contributed.

The other reason for making all the work available under license and then choosing from it is so that collaboration can happen. If run like a contest, the entrants will most likely want to keep their entries secret until the final judging. If they have a great idea, and their compensation depends on the impact of that great idea, they don’t want to give it away. If the work is made public by the vendor after the producer is already paid, they will want to make it available so that others can pick up and polish the idea. Take an example where one vendor has a great idea for a commercial, but doesn’t have the means to produce the video. They can either write the script themselves, pay a writer, or find a friend to do the script for them. If they license it under CC then someone who does have the means to produce the video, find sets, and locate actors can provide the second part of realizing that vision. Taking the contest out of the picture opens up the possible avenue of collaboration between vendors and volunteers. I think that would be absolutely essential, as it is for any other open source project. I ran this by my girlfriend Eleanor, who does business strategy consulting, and she said a project like this would almost definitely be able to get pro bono work from some established firms just so that they could say they were involved with the project.

Overall, I think it’s a GREAT idea. Kudos to whoever pinned this as a necessary step for the Linux movement. I think the time is finally ripe to start looking at options like this. But it’s one of those areas where the devil is really in the details. One small mistake in the execution of the project at the early stages could really derail the whole thing.