Economists have long understood the corollary concept of Coase's ceiling, a point above which organizations collapse under their own weight -- where hiring someone, however competent, means more work for everyone else than the new hire contributes. Software projects often bump their heads against Coase's ceiling: recall Frederick P. Brooks Jr.'s seminal study, The Mythical Man-Month (Addison-Wesley, 1975), which showed how adding another person onto a project can slow progress and increase errors.

What's new is something consultant and social technologist Clay Shirky calls "Coase's Floor," below which we find projects and activities that aren't worth their organizational costs -- things so esoteric, so frivolous, so nonsensical, or just so thoroughly unimportant that no organization, large or small, would ever bother with them. Things that you shake your head at when you see them and think, "That's ridiculous."

Sounds a lot like the Internet, doesn't it? And that's precisely Shirky's point. His new book, Here Comes Everybody: The Power of Organizing Without Organizations, explores a world where organizational costs are close to zero and where ad hoc, loosely connected groups of unpaid amateurs can create an encyclopedia larger than the Britannica and a computer operating system to challenge Microsoft's.

Shirky teaches at New York University's Interactive Telecommunications Program, but this is no academic book. Sacrificing rigor for readability, Here Comes Everybody is an entertaining as well as informative romp through some of the Internet's signal moments -- the Howard Dean phenomenon, Belarusian protests organized on LiveJournal, the lost cellphone of a woman named Ivanna, Meetup.com, flash mobs, Twitter, and more -- which Shirky uses to illustrate his points.

The book is filled with bits of insight and common sense, explaining why young people take better advantage of social tools, how the Internet affects social change, and how most Internet discourse falls somewhere between dinnertime conversation and publishing.

Shirky notes that "most user-generated content isn't 'content' at all, in the sense of being created for general consumption, any more than a phone call between you and a sibling is 'family-generated content.' Most of what gets created on any given day is just the ordinary stuff of life -- gossip, little updates, thinking out loud -- but now it's done in the same medium as professionally produced material. Unlike professionally produced material, however, Internet content can be organized after the fact."

No one coordinates Flickr's 6 million to 8 million users. Yet Flickr had the first photos from the 2005 London Transport bombings, beating the traditional news media. Why? People with cellphone cameras uploaded their photos to Flickr. They coordinated themselves using tools that Flickr provides. This is the sort of impromptu organization the Internet is ideally suited for. Shirky explains how these moments are harbingers of a future that can self-organize without formal hierarchies.

These nonorganizations allow for contributions from a wider group of people. A newspaper has to pay someone to take photos; it can't be bothered to hire someone to stand around London underground stations waiting for a major event. Similarly, Microsoft has to pay a programmer full time, and Encyclopedia Britannica has to pay someone to write articles. But Flickr can make use of a person with just one photo to contribute, Linux can harness the work of a programmer with little time, and Wikipedia benefits if someone corrects just a single typo. These aggregations of millions of actions that were previously below the Coasean floor have enormous potential.

But a flash mob is still a mob. In a world where the Coasean floor is at ground level, all sorts of organizations appear, including ones you might not like: violent political organizations, hate groups, Holocaust deniers, and so on. (Shirky's discussion of teen anorexia support groups makes for very disturbing reading.) This has considerable implications for security, both online and off.

We never realized how much our security could be attributed to distance and inconvenience -- how difficult it is to recruit, organize, coordinate, and communicate without formal organizations. That inadvertent measure of security is now gone. Bad guys, from hacker groups to terrorist groups, will use the same ad hoc organizational technologies that the rest of us do. And while there has been some success in closing down individual Web pages, discussion groups, and blogs, these are just stopgap measures.

In the end, a virtual community is still a community, and it needs to be treated as such. And just as the best way to keep a neighborhood safe is for a policeman to walk around it, the best way to keep a virtual community safe is to have a virtual police presence.

Crime isn't the only danger; there is also isolation. If people can segregate themselves in ever-increasingly specialized groups, then they're less likely to be exposed to alternative ideas. We see a mild form of this in the current political trend of rival political parties having their own news sources, their own narratives, and their own facts. Increased radicalization is another danger lurking below the Coasean floor.

There's no going back, though. We've all figured out that the Internet makes freedom of speech a much harder right to take away. As Shirky demonstrates, Web 2.0 is having the same effect on freedom of assembly. The consequences of this won't be fully seen for years.

Here Comes Everybody covers some of the same ground as Yochai Benkler's Wealth of Networks. But when I had to explain to one of my corporate attorneys how the Internet has changed the nature of public discourse, Shirky's book is the one I recommended.

This essay previously appeared in IEEE Spectrum.

Since 9/11, more than three dozen federal air marshals have been charged with crimes, and hundreds more have been accused of misconduct, an investigation by ProPublica, a non-profit journalism organization, has found. Cases range from drunken driving and domestic violence to aiding a human-trafficking ring and trying to smuggle explosives from Afghanistan.

The meta-problem is that the kind of person who wants to be federal air marshal is the exact kind of person we don't want for the job.

Before 9/11, the Air Marshal Service was a nearly forgotten force of 33 agents with a $4.4 million annual budget. Now housed in the Transportation Security Administration, the agency has a $786 million budget and an estimated 3,000 to 4,000 air marshals, although the official number is classified.

And 3,000 to 4,000 is a lot of people to hire quickly; it's hard to weed out the bad eggs.

After Interop Las Vegas you may have seen our analysis of the data that we collected and delivered in our NOC view.  I thought I’d recreate the same data for NY and do a short comparison.

1) Like in Vegas, uptime for the network 100%.  This is no small feat considering that we introduced a new wrinkle in NY, taking down the primary NOC while the education portion of the show was still going on.  This was a forced failover to the backup systems, and it went flawlessly.  I’d like to give a little credit to EM7 on the 100% uptime as it caught a failover to battery power that allowed AC to be restored before a series of critical equipment would have gone down.

2) Again like Vegas, the average monitored device in the show network didn’t even hit 10% CPU utilization.  Still lots of computing overhead availabe in the show network.

3) The NY show network wasn’t nearly as busy as in Las Vegas, sustaining an average of only 27Mbps of usage (versus 56 Mbps) in Vegas.

4) Power consumption for the network and NOC in NY clocked in at 445kwh per day, about 25% less than the Las Vegas show.  This wasn’t because the equipment was any more power efficient, but instead because the show was smaller and therefore there was less network gear.

5) Finally, a stat we didn’t track too carefully in Las Vegas, but that I find interesting.  During show hours the wireless network average 1,100 users attached.  That’s a lot of people and a lot of wireless devices.

The good news is there was nothing too unexpected in the data, overall the smaller show led to a smaller number of tickets and smaller consumption of resources across the board.  We hope to have the opportunity to work with the InteropNet team again next year and take a look at this data year-over-year for each show.

I’ve often read people who say that we (security, risk management) need to “think like the attacker”.  And when you read this sort of article, that usually alludes to trying to anticipate the tactics an attacker might use to mess with your C, I, or A.  Smart stuff, that, and very useful when architecting security solutions.  But as I was training some folks Monday, I was thinking in the back of my head about Threat Capability (TCap) in FAIR.  As you might know, we like to estimate the capability of a threat to apply some level of “force” against our assets.  This ability to apply force is a byproduct of the attacker’s skills and resources.  And thinking of how an attacker applies skills and resources, I came across another way we might “think” like an attacker.

Traditionally, I’ve thought of “skills” as being a byproduct of the toolset an attacker has.  This mindset probably stems from my time with Penetration Testing teams, where in the process of scoping the  PenTest I would ask our clients to select the level of effort that they wanted us to throw at them.  If a client chose “high” we’d throw every ‘spoit we had at them.  If they chose “low” we’d limit ourselves to a more commonly available toolset.

But while the resources part of TCap is time & materials (money) - the skills are really more than just the toolset.  Skills would include the ability of the attacker to be creative and innovative.    As an example of that innovation from those PenTesting days - when we got a “high” effort request, we would always try to couple that with some “social engineering”-type of attack, or some unique means of delivering an existing exploit.  Our creativity was not necessarily a byproduct of a unique exploit or tool we had, but the process by which we might deliver pre-existing or commonly available exploits.  I remember when we first got ahold of a handful of 32mb thumb drives (hey, 32mb was huge back then) and “dropped” a few in the lobby of a client’s retail space.  The keystroke loggers and phone-home script weren’t new, but using the thumb drive as delivery vehicle certainly was.

So I’ve started to really think about this concept of innovation, and how if “thinking like an attacker” means to be innovative, we ought to do the same.  I’ve been thinking of two main categories of innovation this morning.

INNOVATION

The first I’ll call Technology Innovation.  And by Technology Innovation, I mean some new, unique, “ahead of the curve” technology that an attacker can use against us.  The obvious example of which is a zero-day.  It’s that “high” tool set our PenTesters would use against the clients.  For security departments, this might be the latest security product designed to enhance our ability to P, D, and/or R.

Alternately, we can be creative in the way we deliver (manage) existing technology.  I think of this as Process Innovation.  It’s doing more with what we already have, just like the PenTest team would be creative in the delivery of an existing exploit.

Unfortunately for us - attackers have traditionally had quite a leg up on us in terms of Process Innovation.  It is much easier fro them to be creative, as they are free of political constraints and bureaucracy.  In contrast, when the security industry tries Process Innovation, the results are checklists and “standards”.  It’s committees and consensus.  An extreme example of which might be something like SABSA - a great work if you want to understand some very smart people’s comprehensive understanding of organizational security  - but the “adoption”of which will do very little to help you be innovative in P/D/R.

It’s worth noting that ultimately, this is one reason I don’t like regulatory compliance efforts - they simply serve to prove how mundane your security department is,  wasting valuable resources that could be spent on creating ways to be more effective.

PROCESS INNOVATION AS A SUBSTITUTE FOR TECHNOLOGY INNOVATION

As we come to the close of 2009, some surveys suggest that security spending isn’t horribly impacted yet by the economy (the latest from E&Y points to only 5% of their respondents getting budget cuts).  But if this is a protracted downturn, and because InfoSec is an operational expense, I would expect cash to become more and more difficult to keep.  And regardless if technology spends do slow, I believe it makes sense to think about Process Innovation because I see Process Innovation as a means to increase effectiveness without significant capital expenditures (effectiveness increases because our ability to manage risk has a direct correlation to the amount of risk we have).

The bad news is, of course, that great innovation is hard.  It is R & D.  Failure is usually a pre-requisite to success.

The good news is, our current state is so bad that many of us don’t need to come up with a whizbang new way of reducing software defects in the SDLC as innovation.  Simply inserting a risk analyst into the PMO’s processes might count as a big enough victory. Be cautioned, though,  that if we’re substituting the risk reductions provided by technology acquisition - Process Innovation might actually be even more “expensive” as it requires us to expend political capital.   But there are (forgive the term) innovative ways to spend this political capital.

For example, by taking a second now and figuring out the 3 things that the rest of the organization can do to make your life easier, when that “I need to reduce your budget” talk comes, you can be prepared to negotiate.  Get a political capital “loan” or “investment” from the C-Suite reducing your budget.  Something to the effect of: “I expected this, and am happy to give up my budget.  But if our tolerance for risk hasn’t changed, what I’d like to do is get you to personally back my office on three projects I’ve identified that can reduce our risk without requiring significant capital expenditure.”

After 26 days, and almost 350 million e-mail messages, only 28 sales resulted -- a conversion rate of well under 0.00001%. Of these, all but one were for male-enhancement products and the average purchase price was close to $100. Taken together, these conversions would have resulted in revenues of $2,731.88 -- a bit over $100 a day for the measurement period or $140 per day for periods when the campaign was active. However, our study interposed on only a small fraction of the overall Storm network -- we estimate roughly 1.5 percent based on the fraction of worker bots we proxy. Thus, the total daily revenue attributable to Storm's pharmacy campaign is likely closer to $7000 (or $9500 during periods of campaign activity). By the same logic, we estimate that Storm self-propagation campaigns can produce between 3500 and 8500 new bots per day.

Under the assumption that our measurements are representative over time (an admittedly dangerous assumption when dealing with such small samples), we can extrapolate that, were it sent continuously at the same rate, Storm-generated pharmaceutical spam would produce roughly 3.5 million dollars of revenue in a year. This number could be even higher if spam-advertised pharmacies experience repeat business. A bit less than "millions of dollars every day," but certainly a healthy enterprise.

Of course, the authors point out that it's dangerous to make these sorts of generalizations:

We would be the first to admit that these results represent a single data point and are not necessarily representative of spam as a whole. Different campaigns, using different tactics and marketing different products will undoubtedly produce different outcomes. Indeed, we caution strongly against researchers using the conversion rates we have measured for these Storm-based campaigns to justify assumptions in any other context.

Spam is all about economics. When sending junk mail costs a dollar in paper, list rental, and postage, a marketer needs a reasonable conversion rate to make the campaign worthwhile. When sending junk mail is almost free, a one in ten million conversion rate is acceptable.

News articles.

But that credit card weirdness is nothing compared to the one I’m about to describe.

Before we do that, let’s take a moment to discuss the design principle of failing securely. The general idea is that if a security mechanism fails, it should fail closed. If your firewall crashes, it should block all traffic, not allow all the packets through. If the power source to your card key system is interrupted, it shouldn’t unlock all the doors. If the connection between your application server and your LDAP directory is severed, subsequent authentication requests should be rejected, not approved. This is not rocket science.

So back to credit cards. I had a conversation last night with an old friend who related a bizarre situation they had encountered during the QA process for one of their web applications. One of their tests involved repeatedly attempting a credit card transaction using a canceled/expired American Express card. Here’s what they saw in their logs, paraphrased by me:

Attempt 1: Denied
Attempt 2: Denied
Attempt 3: Denied
 .
 .
 .
Attempt 49: Denied
Attempt 50: Denied
Attempt 51: Approved

What the…? Approved? That can’t be right. So they ran the test again. Every time, after multiple consecutive rejected attempts, the transaction would inexplicably go through. The threshold wasn’t always 50, but the general pattern was consistent — keep trying and eventually it’ll work. Clearly, this had to be a bug in the code, but a deep-dive into the guts of the application turned up nothing. The application security group got American Express on the phone to see if they had any insight on this odd behavior. The answer? They didn’t concede the failure was on their end, despite log data showing the successful authorization codes.

My gut instinct would be that the application requesting the transactions wasn’t failing securely (e.g. network connection to AmEx timed out, so just approve the transaction). But that explanation wouldn’t account for authorization codes coming back.

So what in the world is going on here? Why would the system behave this way? Is it by design? I can’t think of a single legitimate use case for failing open like this. If this is actually a design decision by the credit card companies, I have no doubt that someone in our audience knows the rest of the story.