danah boyd (love child of e.e. cummings and archie), Marko Ahtisaari and Stewart Butterfiled take apart my rationale for claims of fairness in the weblog world, taken from the
powerlaws paper, and find two flaws I have to cop to.
danah
takes on my assertion that the threshold for having a weblog is only slightly higher than the threshold for getting online. She says:
I wish that this was true. This is where issues of social pressure, time, literacy, confidence, etc. come into play. Consumption and production are fundamentally different and there are different forms of pressure when engaging with either. There is no way that one can possibly say that the threshold for consumption is equivalent to the threshold for production.
danah is right, I was wrong.
I am so immersed in the idea of the internet as an enabling technology that I forget that for many people consumption is a mode they are so habituated to that the offer of production does not, by itself, reverse that pattern. Writing takes more time and energy than reading. Writing exposes the writer to critique, where reading does not. Writing is a skill less practised by most people than reading.
In the future, I'll restate this as equality of technological opportunity, but one heavily dependent on other, external factors.
Marko then does a better job than I did of
taking apart my four reasons for claiming, last year, that the inequality in the weblog world was basically fair. (I label this 'last year' because I am in the process, with Tom Coates, or revisiting that work, using a new data set, which may lead to some more re-thinking.)
The fourth part of my 4-part rationale was "There is no real A-list, because there is no discontinuity" in the power law curve. Of this assertion, Marko notes:
Shirky lists a fourth point in favour of his claim that the weblog world is mostly fair but strictly speaking this is not really a feature of the blogging world at all but rather a claim about descriptions of it.
Stewart makes Marko's critique even more explicit in the comments:
"Since there exists no unarbitrary way of drawing the line that separates the A-list from the B-list ..." This is bad reasoning, IMO. There is no non-arbitrary way or demarcating living things from non-living things (even if you think there is a clear distinction between simple single-celled organisms, viruses, various crystals, autocatalytic sets, etc., there are many vague states between, say, a living fish and a dead fish, as some particular fish is dying. But there are clearly moments when *that* is a dead fish and when it is a living fish.
This is also right -- the last item on the list is the one that didn't belong, for the reasons both Marko and Stewart note. What I should have said, and just said on Marko's blog (a year late):
Fair cop on reason #4. I should have separated it from the others, as Marko has done, because it is a critique of descriptions of the weblog world, rather than a description of that world itself.
If I had to re-state that idea, I would say "The idea of an A-list suffers from great vagueness, because power law distributions are self-similar. No matter who you are, _everyone_ to the left of you gets more traffic or links (whichever you are ranking) than you do." (This holds true even in the Glenn Reynolds edge-case, as no one is to his left.)
In other words, in a power law distribution, the sense that a lot of people are getting more X than you is always true, even if you are getting a lot more X than most people (Glenn fails this restated test). Thus locating anything like an A list suffers not from the kind of "living/non-living" edge case Stewart notes, but from an inability to find any locus of the two categories at all other than the 1st and Nth positions. Not only are the edge cases blurry, the "center" doesn't exist at all.
1. Bill Seitz on January 6, 2004 8:25 PM writes...
I think the big problem is that you raised the issue of fairness without defining the damn thing in a useful way.
Could you do so now? Maybe using EPrime? Or shall we have Steven Johnson stop by to deconstruct the hermeneutics of implicit complicity of the penile poopiheads, or however a frenchman would say it...
http://webseitz.fluxent.com/wiki/z2004-01-06-ShirkyItoMarinoffJustBlogs
Permalink to Commenthttp://webseitz.fluxent.com/wiki/EPrime
2. Clay Shirky on January 6, 2004 9:51 PM writes...
I'll say something is fair when the outcome is acheived by people playing by the same set of rules, and those rules don't pre-determine the winners.
Permalink to Comment3. Bill Seitz on January 6, 2004 11:03 PM writes...
So, if the rules "are" "fair", then it's OK if the outcomes are unequal?
Permalink to Comment4. Bill Seitz on January 6, 2004 11:05 PM writes...
It appears that some people (e.g. Mozart) are born with an inherent talent, which pre-determines (barring tragic accidents, laziness, a preference to spend precious spare time watching SpectatorSports) that such people will "win" in the music game.
Is that unfair?
Some people are born into families that honor music and have the money to support such activites. They may never become a Mozart. But they will be more likely to rise to a certain level in the game than someone born with equal determination and raw talent, but less "support".
Is that unfair? More unfair than the first case? Why?
Some people are born into families with so much money that they can buy a new building for Julliard. They will never be a Mozart, but ...
Is that unfair? More... Why?
All these rules seem to "predetermine" winning to some extent.
Permalink to Comment5. Bill Seitz on January 6, 2004 11:18 PM writes...
What is the "outcome" to each participant, anyway? Attention?
To what end?
(Not sure what my point is here... maybe that none of this matters, so the market is even more likely to be skewed... kinda like "university faculty infighting is so vicious because the stakes are so *low*")
Permalink to Comment6. Bill Seitz on January 6, 2004 11:25 PM writes...
More meta (echoing Stewart's point on Markho's page)...
What stakeholder perspective "is" (most) appropriate? Is it the individual, or "the public" as a whole? Markho himself says "What kinds of features should tools and social networks support in order for them to form stable sustainable social systems?" I don't know what a "stable sustainable social system" is, but it sounds like a goal which has nothing to do with the fairness of allocation of chits to individual players. Though one could maybe argue that a system based on "unfair" rules will demonstrate pathological behavior on the part of individuals leading to "instability" of the overall system. But I think that might require a different standard of "fairness" than you define.
Permalink to Comment7. Bill Seitz on January 6, 2004 11:31 PM writes...
And what's the appropriate scale to look at here? Does it really make sense to look at this as 1 big ocean that all the fish are in?
There are many small ponds... each with its own A-list.
Permalink to Comment8. Bill Seitz on January 6, 2004 11:34 PM writes...
I return to the point that we should concentrate more on the aggregate outcome of the system, and discuss the impact that various "rules" have in furthering that outcome.
I'll pull lines from the Seligman page I first linked to:
"A process that continually selects for more complexity is ultimately aimed at nothing less than omniscience, omnipotence, and goodness.
The best we can do as individuals is to choose to be a small part of furthering this progress (EvolutIon). This is the door through which meaning (MeaningOfLife) that transcends us can enter our lives.
The good life consists in deriving happiness by using your signature strengths every day in the main realms of living.
The meaningful life adds one more component: using these same strengths to forward knowledge, power, or goodness."
Permalink to Comment9. Ian Grove-Stephensen on January 7, 2004 2:39 AM writes...
Danah draws a hard and fast distinction between consumption and production that no more valid than the defined A-list. The two blend; for example is this comment an act of consumption or production? Right on the cusp is Dave Winer's Share your Opml
Permalink to Commenthttp://feeds.scripting.com/
(via John Batelle
http://battellemedia.com/archives/000192.php )
10. zephoria on January 7, 2004 6:23 AM writes...
Ian - consumption is intake while production is offering. Sharing one's feeds is the offering of one's consumption and thus the production of one's consumption. There are different consequences to consumption and production. I take one risk by choosing to read the 92 RSS feeds listed; i take another by offering them for you to analyze me through.
Permalink to Comment11. Daniel Varga on January 7, 2004 8:22 AM writes...
"And what’s the appropriate scale to look at here? Does it really make sense to look at this as 1 big ocean that all the fish are in? There are many small ponds… each with its own A-list."
I think this is a very important point. There is a nice paper formalizing this idea for web links. It's titled "Winners Don't Take All". You might find it under
http://modelingtheweb.com/modelingtheweb.pdf
It says that inequalities are not so large when you concentrate on only one pond.
Permalink to Comment12. Seth Finkelstein on January 8, 2004 6:41 AM writes...
Clay, that fact that there's no central cluster shouldn't translate into "there is no A-list". After all, why should it matter?
In a Gaussian distribution, there's a center, but not a right-hand side.
In an exponential distribution, there's no center, but there *is* a right-hand side.
So A-list is naturally defined as units from the right-hand side, rather than center.
It's not like "living" vs "dead". Rather, it's more like "rich" versus "poor".
We'd never write "There's no such thing as "the super-rich", because income is a continuous distribution. Where do we make the cut-off between "the rich" and "the poor"? No matter how little money you have, there's always someone poorer than you.
Permalink to Comment13. Mark on January 8, 2004 1:49 PM writes...
This entire discussion has now landed squarely in the Philosophy of Law arena. Go read "Justice as Fairness" by John Rawls, then read the mountain of critique for it and against it, then come back and bicker about justice and fairness.
Permalink to Comment