Best point of the article is that you need to keep in mind your 'leverage' as owner of a product when implementing large scale changes. In Digg's case, the community had all the leverage: they posted links to content, curated the content, provided the pageviews for advertisers, and spent the time making the site great. In return, Digg had little leverage at all to cause users to stay if they pissed them off sufficiently.

Facebook and Twitter have leverage: their content is user-generated, and cannot be found elsewhere, this combined with their social graphs make it so if you switched you will not be seeing content from a majority of the people you know or care about any longer. Marketplaces like eBay and Craigslist have enormous leverage: eBay can push out a feature that pisses off 100% of their sellers, but sellers go where the buyers are, and the buyers are on eBay. Google has leverage over its advertisers for the same reason, though it's hard to think of how Google has leverage over its main audience, which is a testament to their cautiousness when messing with the search engine (at least until recently!)

I think this aspect is lost when you have sites like TechCrunch telling product people to just forge ahead and break stuff. Break stuff, sure, but remember that when Facebook break things and users eventually come crawling back, its certainly not due to loyalty or warm-fuzzy feelings but due to leverage. Do you have that kind of leverage?

The question of leverage can be an extremely difficult question to answer honestly to oneself. that gets even harder when you have to show the press/investors that you're staying relevant and continuing to evolve. Innovation for the web tends to mean redesigns, which users hate. if you admit to yourself that users have all the leverage, then you're taking away the go-to tool to prove you're innovating. Most CEOs and product teams would rather justify they're like Facebook or Apple in their own special way and do a shiny new redesign.

My impression of Digg power users at the time was that many of them were participating in voting rings to promote each other's sites and a few were even making money pushing links to the home page. Not all of them of course, but the few that did were ruining it for everyone. So I question the conclusions that the article makes based on asking these power users.

I think that power users (the bad ones) were the root cause of Digg's demise as they pushed bad content to the home page and forced the company to impose more restrictions and limit communication (collaboration) channels.

On the other hand, Reddit reduced the potential reward for gaming the system by 1. dividing the community into smaller subreddits, 2. Allowing down votes, and 3. varying the time a link stays on the home page based on user feedback. They did many other things better as well, but these stand out as important differences that allowed reddit to control the spam problem better and keeping the communities interesting.

I probably would have stayed with Digg if they had implemented one feature: "ignore all by X". I wish I could have, for example, hidden all posts by MrBabyMan as easily as I hide all Farmville updates on Facebook.

Instead, they kept tweaking their design and algorithms until I got annoyed and left.

All the top users on Digg aggressively dugg each others submissions, that is undeniable. What's more, because of the "I'll vote your shit up if you vote mine up" mentality, and the ability to easily follow users to see what they submit and vote on, they also acquired hoards of normal users who would do the same without reciprocity in hopes of joining the top ranks.

It would've been a fairly worthless currency if not for the blogspam networks like Gizmodo, Engadget, Tech Crunch, Mashable, Cracked, Life Hacker, etc. that sprung up to funnel Digg's firehose of users on to ad-laden pages.

While no system would be perfect, Digg's made it excessively easy to game (maliciously or otherwise) and the content of submissions soon became almost irrelevant compared to who was submitting it.

That sounds very eerily similar to J.K. Galbraith's description of trader collusion on the New York Stock Exchange in his classic The Great Crash 1929. In it he describes how traders would fix prices. The general public, though aware of this, still participated in the market -- there was no where else to invest, and there were some gains that could accrue to them as well.

In a sense, a stock exchange is also a sort of social network. Curious thought, that.

Some say you should have such a strong vision that listening to users stuck in the past is a mistake. Others say shame on you for not listening to users, how stupid. Apparently it's only OK to ignore users when it works out.