by Steve Mallett
Tim O'Reilly has written and spoken often on what he coins The Open Source Paradigm Shift. I've heard Tim give this speech a few times, and read it a few to boot. The one major point that sticks with me is that the software we use is no longer just on your desktop/laptop, but the software of the internet that we use everyday a la Google, eBay, Salesforce.com & Amazon to use his prime examples.
Tim goes on to point out that this software that exists only through our browsers or APIs, doesn't play by the same rules as does software that we download and use on our own machines. If I download the source code to the Apache HTTP server I can then compile it and use it in accordance to its open source license. This does not apply to a Google or an eBay. Even if you could download the code that runs Google you couldn't just stick it you home directory and start it up.... there's no value there. It's not the same thing at all. It's Infoware.
This is the point in Tim's speech that the brakes go on for me. For me open source is two things. One practical, the other touchy-feely. The first thing is that open source creates a practical benefit to me in that is works better. The other is the value of trust it gives me. The code is open, it can be forked at will when someone does something evil. Those two characteristics in combination make my wheels turn for open source software. So, what happens when the software I depend on slowly shifts to Infoware that I can never really touch and that while still immediately practical gives me no assurance that it can't be taken away or misused at will without any recourse available to me?
I think we can apply the same principles to the data as we have to the source code. Google, eBay, Amazon, et al. are really only as useful as we allow them to be through the information we give them. We still hold the cards here which means we have options.
My proposed solution is based on backlash at social network sites and some XML based projects I follow. Social networking sites, like friendster, orkut, etc, are really the ultimate in Infoware. There is no value whatsoever in the sites without the data we supply. In this case it is our network of acquaintances... our friends in XML.
When the first social networking site came out we all saw some value in it. It genuinely would be helpful to be able to reach out through people we know to find the perfect match for some need. Then came the copycats. Are you my friend? quickly became a joke and people tired
of giving up their info. At about the same time those who continued to like the idea of social networks devised a project named FOAF (Friend of a Friend). The concept here is that the owner of the data (that's you) creates one XML file containing your acquaintances (the info in 'Infoware') and distribute that as you like.
Another XML file based project I've been following is DOAP. Edd Dumbill wanted to apply the same idea as FOAF to Description of a Project. This is an XML file that contains all the info you'd ever want to know about a software project in one place that doesn't require being duplicated by hand in the handful of open source project sites.
Both of these projects are based on reducing the bother of an activity centered around the Infoware concept. But, there is a further use of following this model. We own and control the data. The info in Infoware is ours to dictate the terms of its use.
Let's apply this to an sample case. A good one is Google. You can and sometimes do tell Google to bug someone else. You do it with a robots.txt file on your webserver. For those unfamiliar Google looks for these in websites and if it says Google, bugger off! it does.
I could extend this model to an Amazon, or whomever challenges it (Amazone), with the data I provide it in terms of book reviews. Here I register as an Amazone user, tell it where it can find my bookreview.xml file and go my merry way knowing that if Amazone decides that if it wants to pull a fast one in the future I can change access to that information and give it to their competitor thus 'forking' them.
[This would have been an extemely useful feature this week with one of Friendster's employees being fired for blogging. We could have collectively pulled our network of friends in the blink of an eye, but as is, they 0wn J00!]
What led me to thinking about this are GPX files. These are GPS data files that describe GPS location co-ordinates. They are written in XML to insure interoperability of the data among GPS handhelds and software. There is another website that is basic Infoware: Geocaching.com It specializes in collecting and distributing information that it collects from users. They haven't done anything evil that I'm aware of, but they don't make GPX files freely available. People upload GPS information in a webform, the website turns it into a GPX file, and hides it behind a specialty 'service'. It does make the information available in a normal web page form, but this still seemed a bit weird to me and a first step towards begging to be forked. Plus, I'd like to make GPX forms from scratch. How do I make them distributable? Like DOAP and FOAF.
So let's apply my homemade GPX files to the Open Source Paradigm Shift. I create the valuable data, I tell those who are interested in it where it is under the condition that it is theirs as long as I choose to grant it to them. That's to say, conduct yourself as to make me want to continue to help make your Infoware useful.
In this model I believe that the freedom to innovate and improve data, as opposed to software code, is best served by being largely distributed and in the hands of the many.
There are some legal considerations here along the line of granting copyright of the information to one large organization to fight on one's behalf as the FSF encourages, and some attribution rights that people would want preserved. I think this could be best addressed with a very simple combination of the FSF copyright assignment and a creative common's attribution license. We'll leave that racket for another day.
Steve Mallett is the founder and managing editor of OSDir.com, the care taker for opensource.org, general software wonk, and a dreadful programmer. You can read about his life in one light meaty snake at http://steve.osdir.com.