In the comments of
Mistakes in the Moral Mathematic of Blogging, I suggested that "[a]s weblogs become less personal expression and more lightweight publishing tool, a number of blogs get a significant amount of traffic from outside the blogosphere (e.g. Gizmodo, which is low on the link chart but very high on traffic.)"
Seb points out an example of that in
William Blaze's analysis of the transmission of the spread of the Linton Freeman’s excellent
Visualizing Social Networks paper. Seeing that Freeman's article was posted here on the 15th, Blaze concludes "Several hours later the link was duplicated on an even more popular (381 inbound blogs) site Many 2 Many, presumably because they saw the link via Blackbelt Jones, or perhaps on one of several other smaller sites that also picked up the link via that site."
This is a telling intuition -- Blaze's assumption is that, once discovered, pointers to Freeman's piece simply passed from weblog to weblog. In our case, though, it came to be posted to Many-to-Many because I saw it on
del.icio.us, a 'social bookmark manager.'
The design pattern of del.icio.us is radically different from that of a weblog. In particular, del.icio.us is designed to be useful to the individual first and foremost, as an easy way to collect and categorize links; its secondary function is to provide an aggregate view of links others find interesting.
Many of us though, possibly a majority of delicious users, find the main page to be as valuable as the bookmark saving service. In this case, I saw
the link appear on del.icio.us on the 15th, and after reading the paper, posted it immediately.
So Blaze is right that context was stripped, but wrong in thinking it was stripped in passing from weblog to weblog. It was stripped when an individual user saved the link for him or herself, in a forum where that behavior is public, and where I happened to be watching the link flow.
So although link credit was stripped, it's not clear that this was because of any negative action. The delicious user who first posted the link can hardly be expected to record contextual data for what is a personal record of interesting material, and webloggers surely shouldn't refrain from posting interesting links simply because they discover them in a low-context environment.
Blaze's core assumption, in other words, is leaky. He starts from the idea that when a link shows up on two weblogs in short succession, it must be that it progressed from one to the other. In this case though, and, I believe in a growing number of cases, the comingling of the weblog world with other forms of both link management and social tools means that even if there were a code of conduct among people who are self-conciously blogging, this would not be enough to produce the effect he wants. As weblogs become increasingly important as general purpose publishing tools, the internal consistency of communal understanding or shared prupose will necessarily shrink.
1. Ken Schafer on January 22, 2004 9:36 AM writes...
Clay,
I love del.icio.us (thanks for pointing me to it).
I agree that the home page has real value in being able to dig up odd links I might not have otherwise found. I also like the fact that the site shows me other users who are linking to the same thing I'm linking to. I like following these links to other user's bookmarks to see what else they might have.
This is roughly equivalent to the old Napster experience of checking the rest of a users collection after finding that they had one track you wanted that no one else seemed to have.
But my concern is that del.icio.us' home page (and the categories) will succumb to spamming as soon as it gets traction. As an open web-based system it is probably doubly attactive because spammers gain the attention of del.icio.us users and Google juice.
Will del.icio.us survive it's own success?
Cheers,
Ken Schafer
Permalink to Comment2. Adam Greenfield on January 22, 2004 11:44 AM writes...
Clay, I'm not sure that WAB would impute any "negative" intention to the manner in which you discovered the link. If I'm understanding him properly, it seems to be that his intention was just to use that particular series of links as an example of memetic amplification and distortion.
That it happens not to have unfolded, in this instance, precisely as he's mapped it does not invalidate the point he's trying to make, although your point about the increasing variety of diffusion channels available is surely apropos as well.
In both cases, architecture has a strong impact on the structure of attributory possibilities. A traditional blog post affords the usual space for full atrribution, linked or otherwise; a "remaindered" link (such as we can see on kottke.org, Anil Dash, or indeed Abstract Dynamics itself) usually does not, an RSS feed may well not, and something like del.icio.us obviously strips everything down to the link itself.
WAB is correct on this: "As blogs begin to emerge as economic entities, both as revenue sources and as means for people to build personal reputations and brands, the danger of stratification increases." The broader point - that the structure of a self-conscious blogging community is bound to change as the architecture of its tools changes - is just as sound, just as relevant.
Even if a bit technodeterminist. ; . )
Permalink to Comment3. William Blaze on January 23, 2004 8:12 PM writes...
Thanks Adam, as always you add clarity to the proceedings.
Going to take a moment to clarify even more. Obviously I was incorrect in guessing that the link on this site came from a blog, and del.ico.us definitely adds an interesting new dimension to the flow of links. But I'm pretty mystified as to how the presumption of that particular link has anything to do with the "core" of the argument.
Quite the contrary the link to Many2Many was the last in a series of links all of which were definitively documented. The final jump to Many2Many can be removed from that chain without effect to the core argument at all. I wrote the piece precisely because I do not make the assumption "when a link shows up on two weblogs in short succession, it must be that it progressed from one to the other". It's the fact that the linkflow in this particular case is in fact documented that makes it interesting. And in the instances where the linkage is not documented I very deliberately revert to the hypothetical.
In addition as Adam rightfully pointed out, I don't look at the stripping of context as being negative at all. Unfortunate perhaps, but as pointed out in this post sometimes necessary.
Permalink to Comment