6 ms·

So the argument is: - Facebook's growth curve looks like that of an infectious disease and - "Ideas, like diseases, have been shown to spread infectiously be

• bwood 13y ago
  The largest indicator of bullshit (as I've commented below) is that based on their model, ALL infections of this type go to zero. They could have modeled Apple, Google, Christianity, or Bitcoin and reach the same conclusion (albeit with different timeframes, I'm sure).
  • civilian 13y ago
    The rise in consumption of hamburgers follows an infectious curve-- thus we predict that humanity will be hamburger-free within 2090. /s Agreed. Not everything is a fad.
  • Fomite 13y ago
    I am rather surprised by this constraint on the model, as an extremely straightforward modification to the standard SIR model allows for diseases to reach an endemic equilibrium state.
• jgalt212 13y ago
  > Bullshit. Shoddy analogy at best, wilfully stupid attention whoring at worst. how did you get so many karma points using such coarse language? It brings down the level of discourse.
  • mcphilip 13y ago
    Even pg noted that top comments are largely awful. I find just skipping the first one makes reading HN a much more interesting experience.
    • megrimlock 13y ago
      Your observation reminded me of the Million Short search engine that skips past the first N ranked hits. Is there a name for the phenomenon that the first entries in any contested ranking tend to be less interesting? There's something about the ranking systems used on the web in the way they tally the nearly-effort-free actions of hordes (clicks, likes, upvotes, star ratings) that means they represent one kind of magnitude (statistical average impulse behavior) but very little of another (deeper reflection & significance). Deeper stuff tends to see less impulse attention and therefore sinks down the ranking, but is there some other way to identify it?
      • boomzilla 13y ago
        It's intrinsic vs perceived values in ranking. You could have some rough measure of intrinsic value of a comment by analyzing its content: length, grammars, wording sophistication, etc. It's definitely not trivial and maybe open to spams.
      • freehunter 13y ago
        It's the same phenomenon that creates pop music, pulp literature, prime time TV, summer blockbusters, two-party voting systems, and AAA games. The mainstream tends to be safe, non-threatening, and appeal to a wide variety of audiences. There may be a genre/topic/idea that better suits people, but it takes work to find it and not everyone wants to put in that work. Not everyone will like Thelonious Monk, and the same people who like him might not like Americana or electro house or black metal. But no one complains about smooth or cool jazz. No one complains about mainstream country. No one complains about techno, no one complains about hard rock on the radio. They're watered down, appeal to wide audiences, and are non-offensive in their broad market. It takes no effort to like mainstream culture, and it doesn't demand anything from your intellectually. You're going to like the Billboard Hot 100, or at the very least you're not going to hate it, because it's been manufactured to be liked by everyone. Same thing with high rated comments in a lot of cases. While the poster might not actually be saying anything, might not have a point, might not have any intellectual depth to their comment, their comment offended no one, was easy to understand, and can be appreciated at face value without needing to think about it. If you can get people to like your ideas without making them think, you're going to be popular. If you make them think, they have the chance to decide if they like what you're saying or not, and if you make them think, you give them the chance to decide that they don't like you.
  • VMG 13y ago
    the topic doesn't warrant a high level of discourse
  • swombat 13y ago
    The level of discourse was brought down to below zero by the article. I'm actually raising it, if anything. This is an appallingly bad "forecast" and I'm slightly ashamed on behalf of Princeton for allowing their name to be associated with it. Also, if you look at my comment history, I don't habitually use "such coarse language". I only use it in appropriate circumstances - like in response to this article. The coarse language has little to do with the karma. Edit: worth noting that the title of the article on Guardian is "Facebook will lose 80% of its users by 2017", not may. That has been editorialised by the submitter. The article is even worse than you may think if all you read was the title.
    • smrtinsert 13y ago
      Well I think ur bullshit. What a great way to communicate.
      • swombat 13y ago
        I wasn't communicating with the authors of the article, just with the commenters here. I was not even insulting the authors (only the "forecast"). You, on the other hand, are throwing insults at people (me in particular) - and doing so in a largely inarticulate manner. Another pretty poor analogy, I suppose. I can see why you liked the article. </sheldon-joke-mode>
    • jrs99 13y ago
      it's only a bad forecast if it doesn't happen. if it does happen, then it's a freaking awesome INTUITIVE forecast. Okay, so there's no science behind it? It doesn't matter. There's no science behind a lot of things, yet they work. When you recognize insane patterns and connections, much like an Artist as opposed to a Scientist, things can work. All that needs to matter is that the intuition has to be right. The science can come after.
  • TelmoMenezes 13y ago
    Oh give me a break. He made a very good point -- the article takes the epidemiological model analogy too far and in a non-rigorous fashion. You are the one "lowering the level of discourse" by focusing on form instead of content and by writing in clichés.
    • jrs99 13y ago
      we don't even know yet if it takes it too far. What happens when 2017 comes along and facebook has lost exactly 80.00% of users since 2014? Will it still have been an analogy taken too far?
      • mbillie1 13y ago
        Yes - 'a broken clock is still right twice a day', and scientific conclusions cannot be accurately or scientifically applied via analogy.
        • jrs99 13y ago
          i guess that could be true. The other day i was trying to convince someone that you cannot use a mathematical conjecture if a proof has not been found yet. I think I succeeded.
• ronaldx 13y ago
  Why do you believe this analogy will not apply to Facebook? Social media is called 'viral' for a reason: because its spread is apparently analogous to infectious disease. This analogy certainly does seem to apply to other memes and social networks of the past - why would Facebook be an exception?
  • jarrett 13y ago
    The argument isn't that it won't apply to Facebook. Rather, the assertion that the analogy will apply is unscientific. It is an assertion of the form: A and B have behaved the same so far. Thus, A and B are of the same nature, and will continue to behave the same indefinitely--an improperly drawn conclusion "Viral" is just a word. The fact that we have repurposed it to refer to electronic media does not magically impart to electronic media the properties of biological viruses.
    • ronaldx 13y ago
      The conclusive headline (use of the word will) is the Guardian's editorial decision, not used by the scientists, if that's the main objection. The article has a more measured statement: "Extrapolating the best fit model into the future suggests that Facebook will undergo a rapid decline in the coming years, losing 80% of its peak user base between 2015 and 2017." Observing common network effects and using it to make predictions seems like valid scienfitic research to me. The hypothesis may or may not be correct, but it's a (very) testable and falsifiable hypothesis. The science is good. http://arxiv.org/pdf/1401.4208v1.pdf http://arxiv.org/pdf/1401.4208v1.pdf Edited to add - jarrett: I accept your point that the research could be more thorough; it's a shame only Myspace data appears to have been used. However, I think this is a valid thread of research and the word 'bullshit' was unwarranted.
      • jarrett 13y ago
        Yes, it is a testable and falsifiable hypothesis. So is "George Washington will be elected President in 2016." Nonetheless, we can fairly argue that both hypotheses lack a firm grounding in the evidence. > Observing common network effects and using it to make predictions seems like valid scienfitic research to me. It's valid I suppose, but purely speculative at this point. It would be much more meaningful if a study demonstrated that social networks almost always behave like viruses in both their growth and decline, than to simply observe a similarity and growth and extrapolate the decline.
• gaius 13y ago
  Right but we can do science here. Myspace, Friendster, Google+... We know a lot about how social networking plays out. There's enough of a sample size there to start doing some observations and making some hypothesis.
  • morgante 13y ago
    No, there's not. In what universe is (optimistically) 5 data points a sufficiently large sample?
    • gaius 13y ago
      If you have 6 things, then 5 is as good as you can get to extrapolate about the 6th.
      • swombat 13y ago
        Yes, when you have only 6 things to look at you might take an educated guess based on what might have happened to the first 5. It's the sort of contextual, gut-driven judgement people call a hunch. That's fine for a conversation about facebook after a few beers... it's pretty silly to take that and write a flippin' paper about it and present it as "research".
    • Fomite 13y ago
      It's not five data points. It's five trials with many, many data points - what you're fitting is a time series. Hell, for some epidemics (epidemic modeling being what I do for a living) I'd kill for five example cases to fit a model against.
  • mbillie1 13y ago
    So there's 'we know the way social media plays out' and there's actual science and the two phenomena have nothing in common beyond 'we drew conclusions based on what we saw'. The difference is that the latter is rigorously controlled, tested and peer-reviewed for according to a vetted methodology, and the former is no different from saying 'basically all asian people I have met are good at math, thus this new asian person I have just now met must also be good at math'
• gdulli 13y ago
  "So the argument is:" It's an interesting way of looking at it, but there doesn't need to be a more complicated explanation for the eventual fall of Facebook than the fact that it's always been the pattern for social networks. There's fundamentally more novelty than real value to it so people will lose interest and it will be supplanted by whatever else has the property of being novel. (And lucky, and in the right place at the right time.) You just don't need something like Facebook to stay connected to your close friends, and the interactions you have with your larger "network" are inherently shallow and capable of being disrupted in the move from one social network to another or disposed of completely.
  • scott_s 13y ago
    it's always been the pattern for social networks. For individual social networks, yes. But what if we lumped all social networks together, and looked at aggregate usage over time? It would constantly increase. I think that instead of dying out, it will plateau until we reach saturation. Other communication technologies follow this model: radio, television, telephones, cell phones, etc. Rather than being a novelty, I think that the past decade has demonstrated that online social networks are something most people want and will continue to use. To be clear, you and I are making contradicting factual predictions about the future. Neither of us knows which prediction will turn out to be true. At the moment, the available data fits both of our predictions. We'll know in about 20 years.