Hacker News Reader

Comment by pierrefar | Hacker News Reader

This has nothing to do with the fact they released a new version?

And these are just downloads, not necessarily long-term users. I'd rather see the remaining users in, say, 3 months.

Your "new version"-concern is addressed in the third paragraph: Assev said the swell of downloads was above and beyond the increase caused by the final release of Opera 10.5 for Windows yesterday. "We compared the downloads against previous launches, such as Opera 10.0, 10.10 and 9.5, and the tripling is above what we would normally expect with a new version launch," he said.

Yes - "above and beyond" but still affected by the new release which means that the ballot didn't cause downloads to triple - to increase, but not to the extent that the headline makes it seem. They don't even try to explain what kind of increase above what they would expect they are getting. maybe its double what they expected, maybe its 5%. The fact that they don't really go into it makes me a bit suspicious.

I think you parsed that incorectly,

  Normal = X
  Normal after realease = X * Y
  What they saw = (X * Y) * 3

PS: The browser choice is on a per country basis, so they can easly seperate the effect.

the tripling is above what we would normally expect with a new version launch

I read that as, "Our daily download rate tripling is above what we would expect for a new version launch."

However, it is still off base to attribute it to the MS page - I downloaded the new version and have (at least temporarily) switched to Opera from Chrome due to the announcement here on Hacker Hews. I've never seen an Opera release here before, so that must count as a one-off too.

  Normal = X
  Normal after realease = X * Y
  What they saw: (X * Y) < (X * 3)

I understood it the way clusterfu_k wrote it.

Reply on news.ycombinator.com