/Message

Dopplr Raumzeitgeist 2007: Where we went last year

Dopplr Blog » Dopplr Raumzeitgeist 2007: Where we went last year, originally uploaded by Stowe Boyd.

Dopplr has published a great image -- the Raumzeitgeist 2007, or 'space time spirit' 2007 -- based on the travels of all the Dopplr community in 2007. Soon they will be letting all of us make similar graphics, it seems.

Flu Vaccine Shortage Posts From 2004

I am reposting several posts I wrote in 2004 at Get Real during the flu vaccine shortage.

Ethics of the Flu Vaccine Shortage: What Would Network Science Do?

The flu vaccine shortage has brought to high relief the inability of our government to effectively respond to public health threats. This is the result of a laissez faire attitude toward the safety net that the government has an implicit obligation to put and keep in place for the old, young, and needy, but even more chilling, as the direct result of outdated ethics.

Confronted with a shortage of flu vaccine, the health care apparatchiks have responded in a 19th century, "women and children first" approach, which may feel like the ethical response, but is in fact not well-grounded scientifically. It turns out that doling out the scarce flu vaccinations to those most at risk will not counter the threat of epidemic. The government bureaucrats may continue in the old, wrong-headed, and unscientific rhetoric, but the public heath people should know better.

Network science has shown that human interaction is scale free: that is to say, some of us have significantly more contacts that the rest of us. As scale-free networks grow, those with more contacts are more likely to add new contacts. This is the power law of popularity and influence that we have seen at work everywhere in human interactions.

Research into the spread of diseases like AIDS has suggested an alternative approach to breaking the non-linear expansion of the epidemic, which diffuses through the population just like innovations:

[from Linked by Albert-László Barabási]

Despite differences in purpose and detail, all diffusion models predict the same phenomenon: Each innovations has a well-defined spreading rate, representing the likelihood that it will be adopted by a person introduced to it. [...] Yet knowing the spreading rate alone is not sufficient to decide the fate of an innovation. For what we must calculate is the critical threashold, a quantity determined by the properties of the network in which the innovation spreads. If the spreading rate of the innovation is less than the critical threshold, it will die out quickly. If it is over the threshold, however, than the number of people adopting it will increase exponentially until everybody who could use it does.

Recognizing that passing a critical threshold is the prerequisite for the spread of fads and viruses was probably the most important conceptual advance in understanding in spreading and diffusion. Currently the critical threshold is part of every diffusion theory. Epidemiologists work with it when they model the probability that a new infection will turn into an epidemic, as the AIDS virus did. [...]

For decades, a simple but powerful paradigm dominated our treatment of diffusion problems. If we wanted to estimate the probability that an innovation would spread, we needed only to know its spreading rate and the critical threshold it faced. Recently, however, we learned that some viruses and innovations are oblivious to it.

Research into the spread of computer viruses has led to a new, network science-based approach to modeling -- and countering -- human epidemics.

[from Linked by Albert-László Barabási]

The deadly virus [AIDS] must have followed the route already spotted in the spread of innovation and computer viruses: Hubs are among the first infected thanks to their numerous sexual contacts. Once infected, they quickly infect hundreds of others. If our sex web formed a homogeneous, random network, AIDS might have died out long ago. The scale-free topology of AIDS dispersal allowed the virus to spread and persist.

So the problem before us, ethically, should not be "find those who are the most at risk, and vaccinate them," but rather "who among us are most likely to be the hubs in the spread of the flu? Find them, and vaccinate them."

Especially when supplies are limited, the best hope to stem the rise of the epidemic is to find the most connected individuals in the population -- which in this case means physically connected, not virtually -- and immediately vaccinate them.

I am no expert in the determination of who are the most connected people, but it would likely include some obvious -- and non-obvious -- walks of life. Various kinds of public and health service workers come to mind: doctors, nurses, and other folks in hospitals, clinics, and medical offices who come in contact with the old, young, and infirm are obvious candidates. Starbucks barristas, taxi cab drivers, and bartenders -- while not at risk, necessarily, to succumbing to the flu -- are likely disease vectors who might spread the disease to hundreds of others if they were to contract it. I am even willing to concede that our Senators and Representatives to Congress are likely to fall into this group, and therefore administering vaccine to this group is in the public interest, even while it may appear to be self-serving.

While some have argued that administering AIDS counteragents (we still have no vaccine) to those who are most promiscuous only rewards unsavory and immoral behavior, we have no such quandary in this case. The flu is not a sexually transmitted disease, so there is no moral dimension to vaccinating the bus driver: he could infect hundreds of old, young, and infirm every day. And he would do so simply through doing his job, not through some arguably anti-social act. And worst of all, he could infect two dozen other bus drivers, who would infect thousands, again.

The scale-free network is there, we all know it, and you can't wish it away. Our best choice is to apply what we know.

The stupidity du jour is the bone-headed notion of vaccine lotteries. This is totally dumb, and intelligent people everywhere should rise up against its apparent "fairness." It flies in the face of reason, and squanders perhaps our only chance to stop the spread of the flu in a population confronted with an inadequate supply of vaccine.

The outcome of better science should be the betterment of society, on the whole, and an improvement in every individual's life. However, this is only true if those that govern our collective resources wisely take into account the best scientific thought. If they, on the other hand, disregard science and devolve into outdated ethics and pseudo-scientific mumbo gumbo, they should be hounded from office.

More on Flu Vaccination: Kids are the Supernodes

I posted a plea for the application of network science to the distribution of flu vaccine, last week. Various Corante contributors pointed out some fallacies in my arguments, or at least the fact that my arguments are not well supported by network research to date.

Both danah boyd and Clay Shirky pointed out via email that the studies I cited re: AIDS dispersal were much simpler to model because of the relative difficulty involved in spreading AIDS. The flu, on the other hand, is spread by very casual interaction -- breathing other person's exhalations, or using a cup touched by a flu sufferer -- so that the dispersal is much more general and open.

I concur, as far as the analysis goes. But I maintain that there is still a network gradient involved, and that people should be sorted out to those least likely to spread the disease -- older shut-ins, for example -- and those who are more likely to spread the bug.

Renee Hopkins Callahan came across an interesting support for this position:

[from NPR : Health Experts: Kids Should Get Flu Shots First]

Health Experts: Kids Should Get Flu Shots First

Morning Edition, November 1, 2004 · In a typical flu season, more than 40 percent of school-age kids get the flu. But health officials are trying to get the vaccine to Americans over age 50. New findings suggest children should be vaccinated first to reduce the spread of flu to older adults. Hear NPR's Richard Knox.

Turns out that kids get the flu at over 7 times the rate of adults, and then infect at risk adults. As Renee notes, "This test is based on a case in Japan where flu in older people almost disappeared after a period of years of vaccinating all school-age children, then returned after the vaccination program was discontinued."

So the empirical results suggest that vaccinating school age children may break the epidemic explosion, because schools turn out to be a hot zone for the disease, even though the children themselves are not at risk. So, when we are short of vaccine, we should target the kids to quell the epidemic. Of course, as is noted in the report, if you really are confronted with a pandemic, you should inoculate nearly everyone, but if you inoculate even 25% of the kids, you will see a drastic downturn in the overall infection in the population as a whole.

Action Streams And The Next Cycle In The Social Revolution

Got email from David Recordon of Six Apart yesterday, alerting me to some news:

[from Six Apart - News and Events: Time for Action: What We’re Opening Next]

Today, we're shipping the next step in our vision of openness -- the Action Streams plugin -- an amazing new plugin for Movable Type 4.1 that lets you aggregate, control, and share your actions around the web. Now of course, there are some social networking services that have similar features, but if you're using one of today's hosted services to share your actions it's quite possible that you're giving up either control over your privacy, management of your identity or profile, or support for open standards. With the Action Streams plugin you keep control over the record of your actions on the web.

David Sippey and Byrne Reese of Six Apart also add some insights:

[from Blogging Evolved by Byrne Reese]

But this plugin is not just about activity aggregation, it about control.

But this plugin is not just about activity aggregation, it about control. Because if there is one thing to learn from the one service that even remotely capable of performing this service for you, is that control and privacy is not just important, it is paramount. That is why this plugin:

• is 100% free and open source


• is available for a 100% free and open source blogging platform


• allows users to select which events are public and which are private


• allows users to select which services to aggregate and show activity from


• utilizes open standards to collect and publish data


• and allows users to distribute and do with this data what they please

You can see this plugin in action in a number of different places:

• on the front door of my blog, and on an archive I created
• on Sippey's, Mark's and David's blogs
• on Movable Type dot org's web site to collect and aggregate activity for the entire MT dev team
• and more places to come I am sure.

But no matter how "cool" I think this is, the single most important thing to me is that Action Streams has helped majordojo return to its original purpose: to act as a central aggregation for all of my activity online, and to do so in a way that just works that doesn't require me to do any extra work. Just use the tools I like to use, and let it do the rest.

Never, never put cool in quotes, man!

Michael Sippey Action Streams, originally uploaded by Stowe Boyd.

So, a neat plugin. Yes, people would like to coalesce their myriad streams into something, but I don't think directing that into a old-fashioned blog -- into a publishing metaphor -- is really what people are dreaming of. Yes, I would like to have an archive to search against, although a tree of published pages is not really the best for that: that's what databases are for.

The real issue is that real control will have to be incredibly fine-grained, and the blog publishing approach has been too coarsely-grained in general for that purpose. Blogging is grounded in the 'everything is public' and 'everything is a page' model, and the technologies all reflect that to a great extent.

So, it's not blogging evolution that we need, but a new fusion of the concepts of social media reflected in real social context and the nature of the current web. Like many others, I believe we have to draw on what we know about social networks -- not just their current implementation on the web in Facebook or socially networked services like Flickr or Dopplr -- but real-world social networks: the ways that people share and don't share things with each other.

We are moving to a fragmented world of cascading streams: del.icio.us bookmarks, twitter updates, blog posts, Google reader recommendations, Basecamp milestone updates. Just like the old days of American wireless telephony, the various services don't interoperate well, and a variety of gasketry is emerging to bridge the various services in what turn out to be frustrating and partial solutions.

I get an RSS stream from Dopplr, for example, and plug it into my blog as a widget. However, the blog technology knows nothing about it, and the various trips that Dopplr is streaming don't flow into my blog's RSS feed. And should they? After all, Dopplr has a well-defined sharing model within the application, which allows me to share trip information in a controlled fashion. Exporting it outside of the context of Dopplr means that those controls are lost. I have opted to allow that info to be public, but others wouldn't.

In a sense, I am moving from a higher level of control to a lower one. Like many lowest common denominator approaches, this leads to something getting lost.

If we are going to bring down all these socially rich opportunities for sharing down to the level of open publishing we are going to lose a lot. We will bleed out all the social subtleties. We need to devise tools and technologies -- and the social expectations -- that will allow us to bring much more socially rich interaction into the context of social media.

One of the things I am focused on these days is a startup where my partners and I are are actively exploring these topics. Without revealing too much of our plans, I can characterize the world we see changing and that we hope to midwife. As we move from the 'web of pages' metaphor -- based on hypertext and publishing metaphors -- and enter a 'web of flow' -- based on information streaming through interpersonal relationships -- we will have some major disruptions.

The proliferation of hundreds -- perhaps thousands -- of highly specific social applications that will produce streams derived from people's activities, writings, and media is leading to a fragmentation of web identity and social connection.

There is no hope of 'one ring to bind them' at least in the near term. There is no standard, no app, no movement that is going to head off the next few years of chaos as various contending models are dreamed up and tested in the marketplace of ideas as applications, platforms, or widgety glue.

I am betting on a step-wise process, and the announcement from Six Apart is an example of that. We will see increasingly sophisticated widgetry being created, leading us one step at a time up and away from the page/ publishing model toward a new world of socially constrained streams. It will take years, but new standards will emerge, either de facto or de jure, that will allow us to control access to information flowing through a sprawling web of flow based on social controls. Whatever form this takes will certainly require the controls -- to some extent -- being pushed into the information that is flowing. The updates, photos, notes, recommendations, and videos that we will share through these streams will have to be locked (encrypted) so that only those to whom we have given the keys can open them.

We will need a federated identity system as the core of this, but just as important is the basic notion that a vast conglomeration of interoperating applications would have to share the metaphor of streams as the basis for everything, and move away from pages. We will need to federate the identity of the applications in exactly the same way that we will validate our human identities, so that applications can stream information on our behalf. A project management tool might serve a user the status update from a friend that was created in a Twitter-like app of the not-too distant future, and vice versa, the status update of that first user could travel out to the Twitter-like app, and be propagated to anyone allowed to see it. For this to work, the lock has to be in the stream droplet, and every application will have to allow us to slide in the key to open it.

Ultimately the ability to create a published page at a specific URL on the web will be something like the ability to dump the bits stored at a specific location on my hard drive: occasionally useful, but not the way that we generally interact with the information there.

I believe this is as profound a change as the movement from text to windows-based UI, or the movement from disconnected computing to the Web.

The final completion of the social revolution will be, not surprisingly, wiring the sociality that connects us together into the architecture of the platform we will be standing on.

links for 2008-01-31

• CIO.com - Business Technology Leadership - Hot Tech Toys from the Past: Geek Chic from the '90s Doesn't Look So Cool Today - Slide 06
  Man, seeing a Sony Mavica camera zoomed me back so hard I almost whiplashed.
  (tags: cio.com diann+daniel sony+mavica palm gear tech+toys)

links for 2008-01-29

• Pistachio Consulting Inc. » DEMO: last-minute tips & audience “scorecard”
  Tios for DEMOites and other pitchers: 'You might think presentations answer “What is it?” Nope. The good ones answer “What does it do?” and “What does that matter?”'
  (tags: demo pitches pistachio presentations)
• Eye-Fi » Home
  Wireless SD card. Smart!
  (tags: eye.fi wireless photography sd-card)
• Is the Tipping Point Toast? -- Duncan Watts -- Trendsetting
  Duncan Watts doesn't believe that influencers are critical for viral trends to spread. Hmmm. I bet he's wrong. Like saying school kids don't spread the flu.
  (tags: flu trends sociality social+networks virality tippingpoint malcolm+gladwell big+seed yahoo hushpuppies)
• Pete Cashmore, Bambi Francisco Discuss Mashable 2.0
  Drowning in "ish" "kinda" and "like", Pete fails to illuminate his plans for M 2.0.
  (tags: pete+cashmore mashable vator.tv bambi+francisco)