Delicious and the Preservation of "Platforms"
Just as plenty of others have, I recoiled in horror when I heard that Delicious (née del.icio.us) was being "sunsetted". Regardless of the red flags that have been raised about its potentially imminent demise, I've still been using it on a daily basis. I've been an active user for over 6.5 years, which is longer than I can say for just about any other web platform or service. I deleted my Friendster and Myspace accounts quite a while ago; I've been on Flickr almost as long as Delicious, but the bookmarking wins out by a good four months or so.
I started using Delicious in my final semester of library school, and it shows. I used it for procrastinating as well as a way to organize research materials before I had Zotero. The bulk of the bookmarks from that first day of use (February 24, 2004) were likely imports from my browser, but I quickly showed a facility for adding stuff that I saw as interesting, useful, etc. I became frantic about tagging, and quickly developed my own conventions for adding tags. One of the things I remember being impressed by was that people quickly found ways to build on top of Delicious, and going through my old bookmarks with the tag "del.icio.us" shows a whole number of these, and an October 2004 blog post from the RSS Weblog lists even more. I was particularly surprised and impressed with the number of alternate posting interfaces that sprung up, and by sid.vicio.us, which built ontologies based on Delicious tags.
As my friend and colleague Amelia Abreu writes, part of the value of Delicious has been its ability to allow for individuals interact with others they didn't know well. Arguably, it was one of the first social networking platforms that really valued the importance of weak ties instead of strong ones. I think it's also the platform in which I first encountered Jo Guldi, who also wrote a fantastic blog post in 2007 about how Delicious changes academic research. Specifically, it allows you to sort things, and arguably more importantly, it makes things public:
What's rapidly happening with these shared tags is academics finding each other in rapid numbers. I have some twenty people in my network, at least half of whom I've never met in real life. ... Each of these is another intellectual putting together rarified connections about strange pieces of thought somehow related to my world. ... This is the nearest thing to running into someone else at the card catalog yet. I don't check in with them. I don't have, nor do I really need, the capacity to send email to them. Some of them I may actually encounter at academic conferences later, and we'll share more of a bond, through our years of doing collaborative research, than many scholars who have labored through the years in adjoining offices.
Could Joshua Schacter, the creator of Delicious, or even Yahoo!, anticipate all of the potential uses of Delicious as a platform? Definitely not. There were countless unsupported or under-the-radar ways to use it. I remember a Delicious hack (for which I can't find a link) wherein you could send someone a private message by tagging the bookmark as for:username and marking it as private.
But what is to be done given that it's not dead yet pining for the fjords? It's obviously important stuff, but I feel that it's a bit hyperbolic to say, as ReadWriteWeb has, that Delicious's data policy is like setting a museum on fire. I don't think a large cultural heritage institution such as the Smithsonian could really ever support Delicious as it is, despite what Gabe Rivera thinks should happen.
In his post about preserving Delicious bookmarks through migration, Ed Summers advocates for releasing the Delicious data to the Web somehow. As he writes, this could be mediated by an institution like the Smithsonian or the Library of Congress. Nonetheless, we have to consider what this mediation means, or even what data would be made available. As Stephen Hood notes, as a platform, Delicious has a whole host important features that need saving: the networks of users, and users' inboxes, which contain bookmarks shared by others.
Although the Library of Congress reached an agreement with Twitter to acquire its data regularly, they are not promising all that much. To wit, user interaction with the data won't follow the paradigm of using Twitter (emphasis added):
Tools and processes for researcher access will be developed from interaction with researchers as well as from the Library’s ongoing experience with serving collections and protecting privacy and rights. The Library is not Twitter and will not try to reproduce its functionality. We are interested in offering collections of tweets that are complementary to some of the Library’s digital collections...
Additionally, Martha Anderson of the National Digital Information Infrastructure and Preservation Program was quoted in a New York Times article from May 2010 stating that access would only be granted to "qualified researchers." I have seen no other coverage that qualifies who or what constitutes a "qualified researcher" in this context. In short, I don't think the Library of Congress is a suitable candidate for transferring something that arguably needs to have open, unfettered access from the beginning.
In the meantime, we can migrate our bookmarks to other services, but how do we migrate or emulate the experience of Delicious as a platform? Is the a platform a significant property, or is it an essential part of the context of the records created on that platform? Or, perhaps more curiously, is the platform a record itself? In my opinion, these questions require in-depth critical work if we want to have any hope for the long-term preservation and access to Delicious and other platforms with social networks and user-generated content.