Asa, your link to WebSideStory is blank. Should it link here?

http://www.websidestory.com/services-solutions/datainsights/spotlight.html

Given the latest growth rates, the year end target still appears attainable, but a mid-year achievement is unlikely unless we see increased marketing activity from the Mozilla Foundation.

I wonder how much growing we have yet to accomplish... according to that Websidestory link, it says Firefox growth is slowing... was 34% in about the first month after release, then dropping to 22% for about a month, and then 15% growth for the last 5 weeks or so.

Do you think that's a bad sign that the Firefox userbase won't grow much further? I'm hopin' it's just meaning there was a fast start from all the hype of the release...but still, when will it level off and stop dropping?

I'd like to quote a comment from MozillaZine:

They're clearly assuming compounded growth, while there's no reason to believe that's realistic.

To illustrate: if we added one user per month, the percentage growth in the second month would be 100% (from one to two users), in the third month it would be 50% (from two to three users), in the fourth month it would be 33.3% (from three to four users), etc. So growth would be constant in linear terms, but decreasing if you assumed compounded (ie. exponential) growth.

It should be clear from the outset that there's no way compounded growth figures could be sustained. A more likely model is slow (perhaps linear) growth for a long time, until a certain critical mass is attained. Then a period of exponential growth, leveling off at some equilibrium value that will be very hard to surpass. This model, at least, is consistent with the theory on network effects.

In this case I think the period of exponential growth will begint when IT departments start adopting Firefox en mass. That will expose a lot of new users to the product who will then also install it at home.

I've done some number-crunching based on their figures, and: yes, they're assuming exponential growth. What I see looks much more like a surge around the 1.0 release, which then settled back to a lower (but steady) level of linear growth. *IF* that steady linear growth continues, Firefox will hit 10% sometime in September 2005. That's a very big "if", of course.

Details of my calculations are at http://www.livejournal.com/community/stats_weenie/2005/02/27/

"If" the steady linear growth continues? Personally, I think it would grow linearly if nothing else changed. As we've got a 1.1 release coming up a bit before then, I expect we'll see 10% some time before September due to the 1.1 press we'll get.

"I think it would grow linearly if nothing else changed"

I don't, because the numbers they are looking at are the relative numbers. So if you expect the linear growth to continue, you would expect to see Firefox usage over 100%. That is perhaps a little to optimistic (no offence, Asa :) ).

Therefore, I find the conclusions from WebSideStory a little shaky at best. But hey, they just do this kind of "statistics" for a living.