Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×
The Internet Math Science

Network Scientists Discover the 'Dark Corners' of the Internet 99

KentuckyFC writes "Network theorists have always simulated the spread of information through the internet using the same models epidemiologists use to study the spread of disease. Now Chinese scientists say this isn't quite right--it's easy to infect everybody you meet with a disease but it's much harder to inform all your contacts of a particular piece of information. So they've redone the conventional network simulations assuming that people only ever transmit messages to a certain fraction of their friends. And their results throw up a surprise. In these models, there are always individuals or clusters of individuals who are unreachable. These people never receive the information and make up a kind of underclass who eke out an information-poor existence in a few dark corners of the network. That has implications for organizations aiming to spread ideas who will have to think more carefully about how to reach people in these dark corners. That includes marketers and advertisers hoping to sell products and services but also agencies hoping to spread different kinds of messages such as safety-related information. It also raises the interesting prospect of individuals seeking out the dark corners of the internet, perhaps to preserve their privacy or perhaps for more nefarious reasons."
This discussion has been archived. No new comments can be posted.

Network Scientists Discover the 'Dark Corners' of the Internet

Comments Filter:
  • by Megane ( 129182 ) on Friday October 25, 2013 @04:57PM (#45240335)
    Network scientists don't get around much.
    • by Anonymous Coward

      Also, scientists apparently don't understand how the internet works.

      A server can be a dark corner if (a) not many clients know about it, (b) not many clients have access to it (* this includes advertising clients).

      A user/client can "live in a dark corner" if he/she (a) doesn't use the internet, or (b) only visits servers that would be classified as dark corners. A user can also choose to visit dark corners to conduct "private" business. Note: Private business does NOT mean nefarious. It just means the scien

      • by Megane ( 129182 )
        A server can also be in a dark corner if someone builds a wall around it. [theregister.co.uk]
  • According to the feds, that's just two ways of saying the same thing.

  • what? (Score:5, Funny)

    by xevioso ( 598654 ) on Friday October 25, 2013 @05:00PM (#45240395)

    I don't really understand what they mean. Are they implying that there are entire pygmy tribes somewhere that spend their entire day on IRC? That somewhere there's a bunch of Tunisian goat-herders that only get their news through Usenet?

    If this is the case, who cares, and why?

    • Re:what? (Score:4, Funny)

      by Nutria ( 679911 ) on Friday October 25, 2013 @05:08PM (#45240481)

      Are they implying that there are entire pygmy tribes somewhere that spend their entire day on IRC?

      Kinda.

      if this is the case, who cares, and why?

      Didn't even read the whole summary?

    • I don't really understand what they mean. Are they implying that there are entire pygmy tribes somewhere that spend their entire day on IRC? That somewhere there's a bunch of Tunisian goat-herders that only get their news through Usenet?

      If this is the case, who cares, and why?

      Well, it *does* seem like there's a whole tribe that have just now discovered every email forward since 1989. And they're all in my friends list.

    • Re: (Score:2, Funny)

      by Anonymous Coward

      They're saying that some people are running ad-blockers and that's totally unacceptable.

    • by Jeng ( 926980 )

      What they are saying is that the way that assumptions were made in regards to older ways of studying this information, your examples could have existed. That with the new way of looking at the problem is a lot more reality based.

      At least that is what I got out of the article.

    • by wbr1 ( 2538558 )
      Join #clickclickwhistle on effnet
    • by nashv ( 1479253 )

      Don't worry too much about what they are saying. They did a formal study of the obvious. The TL:DR is :

      If a transmitting node has finite range and finite time of operation, depending on the size of the network, there will always be nodes that never receive the message because 1. they are out of range 2. They is one other node that can transmit information to them , and it died before it could transmit.

  • by nospam007 ( 722110 ) * on Friday October 25, 2013 @05:06PM (#45240457)

    Because we have our own asocial networks.

  • FTA: "...spreading efficiency is highly correlated with the network heterogeneity..."

    Basically obvious, but is this a negative or positive correlation? For example, disease spread has a positive correlation with decreased heterogeneity. Does their model follow or depart from this? Probably follows, but inquiring mind wants to know!

  • by paj1234 ( 234750 ) on Friday October 25, 2013 @05:14PM (#45240547)

    Perhaps Malcolm Gladwell had better update his book, "The Tipping Point". It's about how fads, crazes and fashions take off. In the book, he doesn't mention people who remain impervious to such things. They are indeed an interesting group.

    • by Anonymous Coward

      We're called "trendsetters", and yes, our ability to form our own unique opinion make us able to convince the whole planet of our ideas.
      Often because we're fundamentally "right", in the ways evolution is "right".
      It just takes time to penetrate.

      Nothing will ever be perfect in the world.
      The hidden assumptions and premises in this study has another world: authoritarian hell

      Captcha: demise

      • We're called "trendsetters", and yes, our ability to form our own unique opinion make us able to convince the whole planet of our ideas.

        No, you are called basement dwellers.
        People that sign up for FB, but never get contacted by anyone.

  • Why people never read the article and instead comment title/description? This article is not about warez and pron but about network theory.

  • by Archangel Michael ( 180766 ) on Friday October 25, 2013 @05:28PM (#45240709) Journal

    I thought they were talking about 4Chan. Imagine my surprise they weren't.

    • by memnock ( 466995 )

      No, they meant /. That's all I ever read and considering the drek that ends up here, it means I might as well live in a cave.

      • No, they meant /. That's all I ever read and considering the drek that ends up here, it means I might as well live in a cave.

        I assumed /. was supposed to be /./ and was at the end, you know /b/ etc ends with a full stop, or, /./

        So slashdot is secretly part of 4chan.

  • Not internet (Score:5, Informative)

    by Dan East ( 318230 ) on Friday October 25, 2013 @05:29PM (#45240721) Journal

    First of all, by "internet" they mean social networks like Facebook and Twitter and the interpersonal communication between people. Second, they have created a simulation, but it's not clear how it actually correlates to the real world. The key thing is they have the concept of "exhausting" sources, so once a person has communicated something, they won't receive or communicate that information again. Obviously that's not the case in the real world, because some people are more interested in certain pieces of information and will continue propagating them much longer than others, potentially seeding enough to compensate for the "exhaustion" of other average users.

    Further, social networks all have a backlog where previous posts can be viewed (particularly true with FB), thus a person still "transmits" a given piece of information indefinitely as other people view their wall going back far in time. Thus it is always possible for a "dark corner" of the "internet" to always catch up by seeing a piece of information in that way instead of only real-time.

    • Congratulations, you've discovered the problem with reading non-physics headlines from Arxiv: an unbelievably vast mishmash of nonsensical assumptions that prohibit publication in any peer-reviewed journal. Solution: avoid doing it at all costs.
    • First of all, by "internet" they mean social networks like Facebook and Twitter and the interpersonal communication between people. Second, they have created a simulation, but it's not clear how it actually correlates to the real world. The key thing is they have the concept of "exhausting" sources, so once a person has communicated something, they won't receive or communicate that information again. Obviously that's not the case in the real world, because some people are more interested in certain pieces of information and will continue propagating them much longer than others, potentially seeding enough to compensate for the "exhaustion" of other average users.

      Further, social networks all have a backlog where previous posts can be viewed (particularly true with FB), thus a person still "transmits" a given piece of information indefinitely as other people view their wall going back far in time. Thus it is always possible for a "dark corner" of the "internet" to always catch up by seeing a piece of information in that way instead of only real-time.

      I like your theory better than theirs... use Bittorrent as a model for social networks. This will reflect reality much better, including "information decay" and "not getting the whole story" as well as information poisoning and all the rest.

      I like it!

    • by http ( 589131 )

      Further, social networks all have a backlog where previous posts can be viewed (particularly true with FB), thus a person still "transmits" a given piece of information indefinitely as other people view their wall going back far in time. Thus it is always possible for a "dark corner" of the "internet" to always catch up by seeing a piece of information in that way instead of only real-time.

      You're so funny. Just try going back a week on facebook. I'll wait while you restart your browser a few times, waiti

    • by HtR ( 240250 )

      So, because I haven't logged into my Facebook account for 4 years, TFA says I'm "uninformed" and part of the "information-poor underclass"?

      Funny - I was actually avoiding Facebook and Twitter because I prefer information, as opposed to, you know, gossip, cat pictures, and what my acquaintances are having for lunch.

      Now I know better. Thanks, Network Scientists!

  • by Dunbal ( 464142 ) * on Friday October 25, 2013 @05:31PM (#45240745)
    WWW != internet. It is merely a sub-set.
  • I suppose this would be similar to rural backwaters. Those middle of nowhere places where people labeled hicks, red necks, libertarians, tea baggers, and the like live cut off from the realities of the other 95% of the world surrounding them. Places where education is more about athletics than academics and knowledge is substituted with opinion.
  • by RedHackTea ( 2779623 ) on Friday October 25, 2013 @05:36PM (#45240797)
    Some people don't want to be marketed or advertised at, so they avoid these gateways. Maybe you should be a kind asshole and not find a way to these "dark corners."
    • by Anonymous Coward
      It does kinda sound like a frustrated telemarketer, annoyed as he places call after call only to discover people are ignoring calls from numbers they do not recognize.
  • It appears that my aggressive spam filter is now a "dark corner of the Internet."

  • The places where you're being bullshitted and force fed shit until you don't even WANT to look for information anymore?

    Did someone get grant money to come up with this "DUH, you don't say..." conclusion?

  • In these models, there are always individuals or clusters of individuals who are unreachable. These people never receive the information and make up a kind of underclass who eke out an information-poor existence in a few dark corners of the network.

    Like the cluster of sites on each extreme of the political spectrum that become completely detatched from reality because they only ever get information from each other.

    • by Megane ( 129182 )
      So places like DailyKos and Infowars? Except of course those two don't get stuff from each other.
  • Is a national passtime in the US.

  • "These people never receive the information and make up a kind of underclass who eke out an information-poor existence in a few dark corners of the network. "

    Assuming they are not uncomfortable or in need, they are not an underclass. In fact, they may be an uberclass.

    I know some super-rich people (a few, not many) and all of them don't really bother with the internet. One of them told me, "Look, I pay people to deal with that bullshit." That struck me as interesting. Many people today *must* be informed

    • "Utility is when you have one telephone, luxury is when you have two, opulence is when you have three—and paradise is when you have none."

      Doug Larson

  • You know you were thinking of it while you read the title.

  • The paper is all about social networks, and has way too many wrong or uninteresting assumptions to be of any real use as is.

    However, there are really unreachable areas of the Internet, because of too long routing paths. These areas change depending on where you are (network-wise), for obvious reasons...

    Not to mention the dark internet, areas that are walled-off or partitioned (don't confuse them with the darknet overlay networks).

  • by briancox2 ( 2417470 ) on Friday October 25, 2013 @06:31PM (#45241267) Homepage Journal

    "It also raises the interesting prospect of individuals seeking out the dark corners of the internet, perhaps to preserve their privacy or perhaps for more nefarious reasons."

    Nefarious reasons is always a subcategory of 'to preserve their privacy'. But more and more it is a small subsection as interest in privacy grows.

  • ...make up a kind of underclass who eke out an information-poor existence in a few dark corners of the network.

    ...preserve their privacy or perhaps for more nefarious reasons.

    ...have scripts to kick out adbots.

  • ...there's always ten percent who don't get the word.

  • Good, as i like my little dark corner of the world where those leeches don't exist.

  • I'm always the last to hear about anything. :-/

    But seriously, I often find that people assume I'm better connected than I actually am. I'm the tech guy, so people figure I already heard about stuff before they did anyways. While that is true it's only in a specific subset of information. .. Things posted about on Slashdot.

    So by all means please do repeat stories of things that broke last week because if it's not here I have no idea.

  • Dark Matter, Dark Energy, Dark Corners, this science fad has got to end!

  • The dark corner is well known. It's called 4chan.

  • Simple fix: clue 'em to Slashdot.
  • The scenario created by these researchers does not take into account information that is received yet purposefully ignored, information lost and rediscovered, or how information can be received more than once, in multiple formats and/or revisions, as well as their flawed case where a node can only transmit once. There is nothing real world about this study and the few correct points the article makes is merely common sense that any average person could have concluded by a few moments of simple thought on t

"If the code and the comments disagree, then both are probably wrong." -- Norm Schryer

Working...