4 ms·

Facebook works in "over 70" languages: https://developers.facebook.com/docs/internationalization/ https://developers.facebook.com/docs/internationalization/ O

by j_baker 14y ago

Facebook works in "over 70" languages: https://developers.facebook.com/docs/internationalization/ https://developers.facebook.com/docs/internationalization/ One estimate says there are 6,910 languages in the world: http://www.science20.com/science_20/blog/rarity_koro_brand_new_language_discovered http://www.science20.com/science_20/blog/rarity_koro_brand_n... In other words, you're simply wrong that Facebook is running out of people who don't have a Facebook account. I might buy that Facebook is running out of people that don't speak the languages it supports though.

carbocation 14y ago
Facebook has 900 million users: http://www.pcmag.com/article2/0,2817,2403410,00.asp http://www.pcmag.com/article2/0,2817,2403410,00.asp Say we assume that all users have the same economic value to Facebook (an extremely unlikely assumption; it's most likely that the first billion have much more discretionary income than the rest). Assuming Facebook gets the remaining 5.9 billion humans (including all those without Internet), they will only be able to increase their userbase by 7.5X. Assuming the world 'wants' businesses to have a P/E ratio of around ~15, Facebook either believes in the two above totally fallacious assumptions (they don't), or they anticipate rolling out new revenue strategies to create value.
bri3d 14y ago
Your approach to estimating Facebook growth by languages supported is interesting but I suspect fundamentally flawed. There's an incredible long tail when it comes to languages spoken, especially when multilingual individuals who speak both an obscure regional language and a more common one are taken into account. http://en.wikipedia.org/wiki/List_of_languages_by_number_of_native_speakers http://en.wikipedia.org/wiki/List_of_languages_by_number_of_... gives a great overview of this long tail - with 70 languages, Facebook probably cover all but a few hundred million people. Simply by comparing the top 25 languages spoken with the languages Facebook support, you can see that several billion people should be able to access Facebook based on language numbers, but yet they only have 900 million users. I strongly suspect that "working internet-enabled device available frequently" is a much better way to estimate potential Facebook adoption than language support. Having never run the numbers, though, I neither support nor refute that Facebook is running out of potential account-holders.
- j_baker 14y ago
  Two things: 1. I wasn't trying to estimate Facebook's growth. I was merely trying to refute the notion that Facebook is running out of people to use their service. My method is arguably quick and dirty, but it makes for a good heuristic, doesn't it? It's enough to tell me that it's probably not worth my time to try and dig up better data. 2. Fine. The number I gave of 6910 probably was too high, but 70 is still less than half of the languages mentioned in your link. If they have 900 million users, that's less than half of the people in the world who have internet access[1]. Thus, it would seem that although there are certainly a number of problems with my approach, the evidence we have so far would seem to indicate that it did the job I intended it to do, doesn't it? [1] http://www.internetworldstats.com/stats.htm http://www.internetworldstats.com/stats.htm
  - damoncali 14y ago
    Suppose you expect Facebook to grow at 50% per year. How long until you run out of internet users? 50% for the next several years is not an outrageous expectation given their 25x sales multiple. I haven't done the math because I'm both busy and lazy, but I suspect it is actually a good bit lower than what is implied by such massive multiples. They are indeed running out of people to sign up. On the planet. That is amazing. So amazing, I would argue, that one could imagine governments getting nervous. When's the last time you saw that in a prospectus?
    - j_baker 14y ago
      I certainly will grant that such a thing is possible. But I will believe what you're saying when I see it.
      - damoncali 14y ago
        They won't actually run out of course, but I expect their user growth rate to rapidly approach the low single digits. Which leaves an awful lot of innovating money extraction to provide the absurdly high valuations they've asked for and gotten. Time will tell.
freehunter 14y ago
How many people on Earth are there who don't speak one of the 70 languages, though? Imagine if they didn't support German. How many Germans speak English, or French, or even Russian? A lot. So while they might not support Tagalog, how many Tagalog speakers don't also know how to speak Filipino? Going on basis of language support is ridiculous at best without weighting how many exclusive speakers there are in that language. I might buy that Facebook is running out of people who don't have fresh water, paved roads, electricity, or high speed Internet connections.
- j_baker 14y ago
  Of course it's a flawed approach. I was responding to a comment on HN, not writing a paper in an academic journal. Yes, if I really wanted get proper data, I would do the things you mention. But please give me some credit. The comment I responded to offered no data points whatsoever. If I offered one piece of data that showed someone something, isn't that a net positive?
  - freehunter 14y ago
    For that point, yes I will give you credit. At least facts give you something real to argue for or against.
krschultz 14y ago
I don't think that really has anything to with it at all. It comes down to how many people in the world are connected to the internet. I've seen a lot of numbers around 2 billion people. Facebook has 900 million users. Even if they got every person on earth, thats only 7x growth (and that would be impossible). Their valuation is not based on 2x or 6x growth, their valuation requires an order of magnitude of revenue growth. There are not an order of magnitude more users left on this planet. And that completely ignores the fact that most of the remaining people aren't as valuable to advertisers.

by j_baker 14y ago

carbocation 14y ago

Facebook has 900 million users: http://www.pcmag.com/article2/0,2817,2403410,00.asp http://www.pcmag.com/article2/0,2817,2403410,00.asp Say we assume that all users have the same economic value to Facebook (an extremely unlikely assumption; it's most likely that the first billion have much more discretionary income than the rest). Assuming Facebook gets the remaining 5.9 billion humans (including all those without Internet), they will only be able to increase their userbase by 7.5X. Assuming the world 'wants' businesses to have a P/E ratio of around ~15, Facebook either believes in the two above totally fallacious assumptions (they don't), or they anticipate rolling out new revenue strategies to create value.

bri3d 14y ago

Your approach to estimating Facebook growth by languages supported is interesting but I suspect fundamentally flawed. There's an incredible long tail when it comes to languages spoken, especially when multilingual individuals who speak both an obscure regional language and a more common one are taken into account. http://en.wikipedia.org/wiki/List_of_languages_by_number_of_native_speakers http://en.wikipedia.org/wiki/List_of_languages_by_number_of_... gives a great overview of this long tail - with 70 languages, Facebook probably cover all but a few hundred million people. Simply by comparing the top 25 languages spoken with the languages Facebook support, you can see that several billion people should be able to access Facebook based on language numbers, but yet they only have 900 million users. I strongly suspect that "working internet-enabled device available frequently" is a much better way to estimate potential Facebook adoption than language support. Having never run the numbers, though, I neither support nor refute that Facebook is running out of potential account-holders.

j_baker 14y ago
Two things: 1. I wasn't trying to estimate Facebook's growth. I was merely trying to refute the notion that Facebook is running out of people to use their service. My method is arguably quick and dirty, but it makes for a good heuristic, doesn't it? It's enough to tell me that it's probably not worth my time to try and dig up better data. 2. Fine. The number I gave of 6910 probably was too high, but 70 is still less than half of the languages mentioned in your link. If they have 900 million users, that's less than half of the people in the world who have internet access[1]. Thus, it would seem that although there are certainly a number of problems with my approach, the evidence we have so far would seem to indicate that it did the job I intended it to do, doesn't it? [1] http://www.internetworldstats.com/stats.htm http://www.internetworldstats.com/stats.htm
- damoncali 14y ago
  Suppose you expect Facebook to grow at 50% per year. How long until you run out of internet users? 50% for the next several years is not an outrageous expectation given their 25x sales multiple. I haven't done the math because I'm both busy and lazy, but I suspect it is actually a good bit lower than what is implied by such massive multiples. They are indeed running out of people to sign up. On the planet. That is amazing. So amazing, I would argue, that one could imagine governments getting nervous. When's the last time you saw that in a prospectus?
  - j_baker 14y ago
    I certainly will grant that such a thing is possible. But I will believe what you're saying when I see it.
    - damoncali 14y ago
      They won't actually run out of course, but I expect their user growth rate to rapidly approach the low single digits. Which leaves an awful lot of innovating money extraction to provide the absurdly high valuations they've asked for and gotten. Time will tell.

freehunter 14y ago

How many people on Earth are there who don't speak one of the 70 languages, though? Imagine if they didn't support German. How many Germans speak English, or French, or even Russian? A lot. So while they might not support Tagalog, how many Tagalog speakers don't also know how to speak Filipino? Going on basis of language support is ridiculous at best without weighting how many exclusive speakers there are in that language. I might buy that Facebook is running out of people who don't have fresh water, paved roads, electricity, or high speed Internet connections.

j_baker 14y ago
Of course it's a flawed approach. I was responding to a comment on HN, not writing a paper in an academic journal. Yes, if I really wanted get proper data, I would do the things you mention. But please give me some credit. The comment I responded to offered no data points whatsoever. If I offered one piece of data that showed someone something, isn't that a net positive?
- freehunter 14y ago
  For that point, yes I will give you credit. At least facts give you something real to argue for or against.

krschultz 14y ago

I don't think that really has anything to with it at all. It comes down to how many people in the world are connected to the internet. I've seen a lot of numbers around 2 billion people. Facebook has 900 million users. Even if they got every person on earth, thats only 7x growth (and that would be impossible). Their valuation is not based on 2x or 6x growth, their valuation requires an order of magnitude of revenue growth. There are not an order of magnitude more users left on this planet. And that completely ignores the fact that most of the remaining people aren't as valuable to advertisers.