Hacker News Reader

Comment by rw | Hacker News Reader

The TC hype machine's two-step pass is operating at full efficiency:

1) Make a dramatic generalization from obviously-inadequate data. 2) Write a followup when better data proves it wrong.

what if the down-trend is as short lived as that initial spike was? nothing has been proven wrong yet (based on your own argument, in fact.) the data isn't "better" yet.

Uhm... where better data is the exact same source as they used before?

Better data is over a longer span of time.

I'm not sure what you're saying. StatsCounter (the source of the data) reported that Bing was briefly a more popular referral than Yahoo, but that the trend has now reversed. If you don't trust the June 4th data, I can't see how you can then accept the June 6th/7th data. The time span over each day is exactly the same.

The point is that the one day they were above yahoo is also the one day they launched, and got everyone trying it for the first time and tons of launch-day media coverage. The number could be 100% accurate, it still doesn't mean much.

TC is drawing conclusions based on one data point. I dispute how they are using the data, not whether that data is accurate.

I wonder if their CPM will go down as a result of all this hype-cycling.

My apologies, when you said "write a followup when better data proves it wrong," I thought you were suggesting that it was now good data.

Would you prefer that they not report on available data? Everyone knows that it's pop journalism, not a government commission report.

Arrington is closer to the tech Zeitgeist than he gets credit for on this site.

Reply on news.ycombinator.com