growth around the world

We have a few things we keep an eye on here at Mozilla — one of them is where in the world our users are coming from — or, more specifically, what language version of Firefox they’re using. In particular, I’ve been watching this lately because the % of our users using the en-US (English, United States version) has been right at 50% for a while — and yesterday, for the first time, it went under half, to 49%, which means that we no longer have an English/US majority of users — and, in fact, have no particular language that’s in the minority. (In truth, our en-GB — English, Great Britain — version accounts for about 4% still, so we’re still English-language in the majority.)

This represents a ton of hard work by communities around the world — obviously our German & French communities are exceptionally strong as well, with the rest of Europe also making gains. Brazil is our most significant user population outside of North America & Europe, followed by Japan. But the fastest growth is coming from “other” — our long tail of languages. I’ll grab some more information over the next few weeks to talk about growth rates & what composes other, but thought I should highlight this milestone now.

I’m very proud of Mozilla to get to this point, and think it shows something very special about the global nature of what we’re up to.

Picture 2

  • en-US (English, US) 49%
  • de (German) 13%
  • fr (French) 7%
  • es-ES (Spanish, Spain) 4%
  • pl (Polish) 4%
  • en-GB (English, Great Britain) 4%
  • pt-BR (Portugese, Brazil) 3%
  • it (Italian) 2%
  • ja (Japanese) 2%
  • other (everything else) 12%

21 comments

  1. Are these data download stats, update stats, some kind of web usage stats or sth else?

  2. Are these data download stats, update stats, some kind of web usage stats or sth else?

  3. These figures probably over-represent en-US at the expense of other English versions. Often it is easier to download the US version than it is to download a specific localisation, so there will be lots of Brit using the US version. And are there Australian, Canadian and other English localisations?

  4. marcoos: they’re update stats.

    ian: they’re just representative of what people are using. there are 2 main English-language versions of Firefox: US, which is what we build here, and GB, which is localized by our Great Britain community. we can, of course, do other localizations, but it’s not clear that there’s a huge need.

  5. These figures probably over-represent en-US at the expense of other English versions. Often it is easier to download the US version than it is to download a specific localisation, so there will be lots of Brit using the US version. And are there Australian, Canadian and other English localisations?

  6. marcoos: they’re update stats.

    ian: they’re just representative of what people are using. there are 2 main English-language versions of Firefox: US, which is what we build here, and GB, which is localized by our Great Britain community. we can, of course, do other localizations, but it’s not clear that there’s a huge need.

  7. Yup and the localisers still get the disc/disk thing wrong. Disc is used for round things like CDs, Disk for things like hard drives (which may contain a round thing but themselves are not round).

    More info here:

    http://en.wikipedia.org/wiki/Wikipedia:Manual_o

  8. Link got broken up in previous posts, I’ve made my name the link this time.

  9. Yup and the localisers still get the disc/disk thing wrong. Disc is used for round things like CDs, Disk for things like hard drives (which may contain a round thing but themselves are not round).

    More info here:

    http://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style_(spelling)#Different_spellings_.E2.80.93_different_meanings

  10. Link got broken up in previous posts, I’ve made my name the link this time.

  11. John, also think about that a lot of users outside of US or GB are using en-US builds. I think it would be better not to analyze the UA string but the accepted language. Users with en-US builds want to have the language of sites mostly in their own language. So this way should give a better result but also need more work to do and a bit of overhead while it needs extra processing of the HTTP header. But it’s just an idea…

  12. John, also think about that a lot of users outside of US or GB are using en-US builds. I think it would be better not to analyze the UA string but the accepted language. Users with en-US builds want to have the language of sites mostly in their own language. So this way should give a better result but also need more work to do and a bit of overhead while it needs extra processing of the HTTP header. But it’s just an idea…

  13. henrik, absolutely — that measures something different, and you’re right that it’s another important one to understand.

  14. henrik, absolutely — that measures something different, and you’re right that it’s another important one to understand.

  15. Are these data download stats, update stats, some kind of web usage stats

  16. Are these data download stats, update stats, some kind of web usage stats

  17. Kirill Koroteev

    I am a native Russian speaker living in France and using the English (UK or US – I have no idea) version. And I think there are lots of people in more or less comparative situation. How would you assess this type of situation? For example, are you able to trace in which countries which versions are used? (If it at all necessary)

  18. forex trade: they’re update stats, so reflect usage, not downloads.

    kirill: i suspect there are a lot of people in your situation — we don’t actively try to figure out country of origin right now, although we may in the future. for now, i’m just trying to share usage by language setting in the browser. (you’re almost certainly using en-US, since that’s the easiest version to get.)

  19. Kirill Koroteev

    I am a native Russian speaker living in France and using the English (UK or US – I have no idea) version. And I think there are lots of people in more or less comparative situation. How would you assess this type of situation? For example, are you able to trace in which countries which versions are used? (If it at all necessary)

  20. forex trade: they’re update stats, so reflect usage, not downloads.

    kirill: i suspect there are a lot of people in your situation — we don’t actively try to figure out country of origin right now, although we may in the future. for now, i’m just trying to share usage by language setting in the browser. (you’re almost certainly using en-US, since that’s the easiest version to get.)