Comments on: State of the Add-ons Ecosystem http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/ Add-ons, metrics, distribution and other Mozilla matters Tue, 14 Oct 2008 11:14:55 -0700 http://wordpress.org/?v=2.8.6 hourly 1 By: bhashem http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/comment-page-1/#comment-1045 bhashem Mon, 11 Aug 2008 05:44:46 +0000 http://blog.mozilla.com/basil/?p=27#comment-1045 @Ethan: Unfortunately, we are not able to offer up this statistic because of how we currently implement the Firefox Automated Update System (AUS) and add-on updates. There is currently no unique ID, cookie or other correlation done to identify users. Add-on updates are checked on a per add-on basis so the update service gets 10 requests for example if a user has 10 add-ons. We've tried looking at IP addresses but that also is not unique enough given our volume. Firefox is architected this way in order to take advantage of the caching infrastructure on the server side. If you batch them up as a single request you'll lose the cache advantage. @Ethan: Unfortunately, we are not able to offer up this statistic because of how we currently implement the Firefox Automated Update System (AUS) and add-on updates. There is currently no unique ID, cookie or other correlation done to identify users. Add-on updates are checked on a per add-on basis so the update service gets 10 requests for example if a user has 10 add-ons. We’ve tried looking at IP addresses but that also is not unique enough given our volume. Firefox is architected this way in order to take advantage of the caching infrastructure on the server side. If you batch them up as a single request you’ll lose the cache advantage.

]]> By: bhashem http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/comment-page-1/#comment-1035 bhashem Wed, 06 Aug 2008 18:41:05 +0000 http://blog.mozilla.com/basil/?p=27#comment-1035 @Ethan: And that 2.5 add-ons/per user calculation is not exactly right either since it's not necessarily the same users who happen to visit and the ones that are ping'ing since they are not time or cookie correlated. It also tells us nothing about the distribution of add-ons per user. (Some may have 50 add-ons while others may have 2). You can't have 1/2 an add-on :-) @Ethan: And that 2.5 add-ons/per user calculation is not exactly right either since it’s not necessarily the same users who happen to visit and the ones that are ping’ing since they are not time or cookie correlated. It also tells us nothing about the distribution of add-ons per user. (Some may have 50 add-ons while others may have 2). You can’t have 1/2 an add-on :-)

]]> By: Ethan Bauman http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/comment-page-1/#comment-1032 Ethan Bauman Wed, 06 Aug 2008 17:06:55 +0000 http://blog.mozilla.com/basil/?p=27#comment-1032 Ooops. Just realized I compared daily add-on pings to monthly active users of Firefox. Should be 150 million daily add-on pings divided by 60 million active Firefox users or 2.5 add-ons per user. Ooops. Just realized I compared daily add-on pings to monthly active users of Firefox. Should be 150 million daily add-on pings divided by 60 million active Firefox users or 2.5 add-ons per user.

]]> By: Ethan Bauman http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/comment-page-1/#comment-1029 Ethan Bauman Wed, 06 Aug 2008 03:33:21 +0000 http://blog.mozilla.com/basil/?p=27#comment-1029 Hi Basil, I'd be really interested to know what percentage of your active users have multiple add-ons installed. For example, how many active Firefox users currently have 3 or more active add-ons? While I can't prove it yet, I suspect that these are the folks who really "get" the customization aspect of Firefox and are core to evangelizing Firefox. Best, Ethan Hi Basil,

I’d be really interested to know what percentage of your active users have multiple add-ons installed.

For example, how many active Firefox users currently have 3 or more active add-ons?

While I can’t prove it yet, I suspect that these are the folks who really “get” the customization aspect of Firefox and are core to evangelizing Firefox.

Best,

Ethan

]]> By: Morbus http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/comment-page-1/#comment-897 Morbus Mon, 30 Jun 2008 18:57:30 +0000 http://blog.mozilla.com/basil/?p=27#comment-897 Thanks for the heads up basil. Thanks for the heads up basil.

]]> By: bhashem http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/comment-page-1/#comment-896 bhashem Mon, 30 Jun 2008 15:44:30 +0000 http://blog.mozilla.com/basil/?p=27#comment-896 @Morbus: Excellent points. Have no fear, discussions for AMO are coming back in the next revision. We initially removed them because there seemed to be other alternatives (developer's web site, MozillaZine forums, etc...) Discussions will be turned on for add-ons and authors can opt to redirect discussions to a private forum. Hopefully that will offload some of the feedback from the review system back into discussions. @Morbus: Excellent points. Have no fear, discussions for AMO are coming back in the next revision. We initially removed them because there seemed to be other alternatives (developer’s web site, MozillaZine forums, etc…) Discussions will be turned on for add-ons and authors can opt to redirect discussions to a private forum. Hopefully that will offload some of the feedback from the review system back into discussions.

]]> By: Morbus http://blog.mozilla.com/basil/2008/06/29/state-of-the-add-ons-ecosystem/comment-page-1/#comment-895 Morbus Mon, 30 Jun 2008 13:18:29 +0000 http://blog.mozilla.com/basil/?p=27#comment-895 About users with add-ons being more loyal, it's a tricky issue. It's not that they are more loyal (i.e., that they won't dump firefox as easily for the competition) by default. The thing is, the more extension we have in our firefox, the more our firefox becomes OURS, and that's a very important thing to take into account when about to go and change our browser. Look at it this way: firefox without extensions isn't as good a browser as it is with extensions, and it's arguably worse than Opera or Safari. But with extensions, it's better, or it can be better, and why would someone dump his browser for a worse one? Also, reading "given the above, the AMO site’s function is to present end users with add-ons, keep them up to date and support the developers who publish these add-ons" reminds me of a thing that's been bugging me (and probably many others) since the last big remodelling of AMO: no discussions for extensions. It's less functional, and encourages users to post complains/problems/suggestions in their reviews. Why the hell did they cut discussions off anyway? I know, it's better for the developers, but IT'S WORSE FOR THE END USERS!!! Jush sain. About users with add-ons being more loyal, it’s a tricky issue. It’s not that they are more loyal (i.e., that they won’t dump firefox as easily for the competition) by default. The thing is, the more extension we have in our firefox, the more our firefox becomes OURS, and that’s a very important thing to take into account when about to go and change our browser. Look at it this way: firefox without extensions isn’t as good a browser as it is with extensions, and it’s arguably worse than Opera or Safari. But with extensions, it’s better, or it can be better, and why would someone dump his browser for a worse one?

Also, reading “given the above, the AMO site’s function is to present end users with add-ons, keep them up to date and support the developers who publish these add-ons” reminds me of a thing that’s been bugging me (and probably many others) since the last big remodelling of AMO: no discussions for extensions. It’s less functional, and encourages users to post complains/problems/suggestions in their reviews. Why the hell did they cut discussions off anyway? I know, it’s better for the developers, but IT’S WORSE FOR THE END USERS!!!

Jush sain.

]]>