Power laws as a thinking tool

It's an open secret in the news industry that while all of those longform articles, parallaxin' interactives and news apps do get great traffic, that traffic is nothing in comparison to that totally random and not even very good quick hit that went viral yesterday.

Personally, I also agree that you shouldn't worry too much about polishing content and instead just get it out there. That's what's so lovely about the blog of statistician Andrew Gelman for example [1], he just writes what comes to mind and his readers love it because they care more about his insights than about whether or not he's presenting an eloquently argued, watertight argument. It reminds me of the old days of blogging, and I think many of us miss those days.

That said, Kent Beck displays some really sloppy thinking here. "I have no clue about what the actual numbers are, and even though I could get them and run some analysis on them to determine the effects of quality vs. quantity, instead I'm just going to philosophize about how I think the world works, and justify my ramblings by saying that it's a POWER LAW, and power laws, that's like nature, dude, you can't fight it."

[1] http://andrewgelman.com/

Here the TL;DR is the following: because the popularity of your blogposts will follow a power law distribution (in which you believe), if you want to be popular you just need to put out more content instead of trying to improve the quality.

An analogy: the popularity of painters follow a power law (or say we believe it). So to be popular, you don't need to be good, but to put out more paintings. Then if you look at the most popular artists, they shouldn't be the most talented, but the most prolific...

And then there are the arguments against the "power law everywhere" meme, that can be found in Mitzenmacher for example.

I can see how this fits with Facebook's "move fast and break things" philosophy. Quantity over Quality, and I'm okay with that if you're in a nice sandboxed environment where you're not accountable for your mistakes. In the case of Facebook, or indeed any vertically integrated environment this works fine but in a great many other environments you need to be more careful about mistakes. I agree that the approach of applying a strict quality control pipeline to creative output is over-applied in the industry as a whole - which Facebook have had some success with challenging - but in a great many other areas it is still absolutely necessary (advances in cloud platforms and devops perhaps notwithstanding). In many cases we have a lesser degree of control over software once it leaves the office, and it is a great pain for us to fix (more costly, more embarassing) once it is in the wild.

Similarly as a more general principle, or as more specifically applied to blog posts - I like other more "ordinary" people am held accountable for the the words that come out of my mouth, and for the material I put up on my blog. Since I'm not a lionized "agent of influence" the words I write will be more critically appraised; could easily be misunderstood, or taken out of context.

That said, and I have to admit to only reading a handful of his posts, the quality of Kent's output could just be naturally "that good" that he doesn't need to worry about these kinds of things. Which could well be one of the characteristics that has got him where he is. But I have to question this approach as applied more generally, to be emulated by mere mortals.

I'd like to add a perspective to the posts questioning the "power law" and it's consequences in this area.

"Gesinnungs Ethik" VS "Verantwortungs-Ethik" (very roughly 'unquestioned attitude' vs 'accountability' Ethics)

for me this blog-post is a fine example of technocratic "Gesinnungs Ethik": the power law is a law of nature, the popular get (easier) more popular, the rich get (easier) richer... that's by law of nature, you can, actually you _have_ to submit to it otherwise your efforts are wasted. That's the way it is. (Other examples are: the market, the rational agent, management principles etc)

Sure, there are processes in nature following the power laws where it is really futile for us humans to intervene or ignore it.

AND

there are other areas where this presented futility most probably is NOT true.

That's where "Verantwortungs-Ethik" comes into play, yes an "antagonist" to the other one: do we want accept the consequences of following/applying this law? Is this driving the development of an area into the right direction? If you are accountable for the large scale consequences f.e., would you follow this law? It's a less passive, "well, that's the way things are", attitude and a more conscious and active attitude, where putting out a lot of posts just in case and for the sake of the one lucky viral attention-grab in a "big data experiment" manner is not ignoring the overall effect on the whole, where there is accountability for collateral damage.

It does not make sense to apply "Verantwortungs-Ethik" to gravity laws. ;) In most social contexts, and in contexts we ourselves construct, we are free to do so. And I think, we should more intensively.

There are some good comments here that get at the false dilemma. The observation that your blog posts will fall on a power law curve is only valuable a posteriori. After writing a blog post, if it doesn't do well you can chalk it up to power law.

A priori (meaning at the moment you sit down to write your next blog post), it's entirely useless, and maybe detrimental. At this moment it's more like the 10,000 hours of practice theory. It's not 10,000 hours of random application that will get you success. It's 10,000 hours of deliberate practice, directed skill-building behavior directed towards marginal improvements.

Will your portfolio of blog posts live on a power law curve? Yes. Does that mean you don't have to put any effort in? No. What it means is do your best every time, but don't expect success every time.

In a basement laboratory somewhere there are 1,000 monkeys typing (and generating a power law curve) and after a few centuries they still haven't written any Shakespeare.

Locally your own oeuvre will follow a power law. Globally your work, when compared to the work of others, will follow a power law. It's fractal in that way. However, if your work is shit it's still going to be hard to break through.

Side note, what OP doesn't realize about blue pill-red pill is that red pill is based entirely on a priori principles while blue pill is the opposite.

tl;dr You can make content so bad that it won't spread, but you can't make it so good that it is certain to spread. Virality grows at the confluence of skill, timing, and luck. Be aware when you've hit the point of diminishing returns on quality and publish.