c=US ID Mgt. Research/Development

Oram and Viega have an interesting comment on the Identity convergence in the cloud.

The plethora of Identity folks is really quite small, not sure how lucrative it is (based on an increasing degree of risk). With increasing risk, personal information is increasingly difficult to manage and protect. And there's far more options in play than Open ID and Microsoft Live ID. Since I'm studying fractions I wondered what the common denominator would be if you took all the systems together and came up with a number, and not just the typical ones that are mentioned in press releases and the literature. If so, there would have to be a constant, or unifying element like 137.

In an ecosystem each of the use case actors plays a role, and the public deserves a realistic look at the future. Viega and Oram do this by looking at the criminal underworld of carders who gather up account and personal information, (through techniques such as phishing and database break ins) and others who work with them, and then fast forward those issues into the cloud paradigm.

They pretty much lay out the details of how shoppers information was gathered from retail stores that had been penetrated by poor WIFI security practices, (don't use WEP on your WIFI) stored inappropriately, and siphoned off to the carder marketplace.

And what's really interesting is that the person that LE had working on the case was a former hacker, Gonzales, who the Secret Service said was working with both sides.

However, this was after around 50 to 100 million (or more) accounts had been stolen.

Systemically then, why should something be this vulnerable, or get to that level of loss? Some of it is due to poor security, but also due to the sheer mass of data flows.

These retailers really didn't put sufficient security controls into place. One example, Walmart, apparently reacted quickly with appropriate measures.

I'm reminded of Dirac who saw beauty in equations, predicting particles before they were found through mathematical logic. The quest for beauty is worth exploring here, looking for an elegant solution. So Dirac, Heisenberg, Pauli all shared ideas that opened up our conception of the universe, but how predictable is it really?

There are concerns by a fairly significant amount of evangelicals regarding Revelations, and while religious beliefs don't figure prominently into security systems design, that belief structure has affected society, especially when it comes to apocalyptic thinking. While apocalyptic thinking has had a long history, (numerous examples come to mind), post WWII, we have had to really sacrifice quite a few pre-conceptions to mathematical models of risk writ on a large global stage.

We often hear about proverbial political compromises being compared to "sausage making", a common product we enjoy, but not the specifics of how it gets done.

The expression is actually quite old (the first use I ever came across was in a random browsing of a dusty volume of ancient Greek prophecies), "beware the sausage makers".

Security is essentially an enabler of the new technologies, as a developer, one looks at how difficult it is to enable identity management for the domain that one working on. 

The collaborative nature of working with different technologies (instead of focusing on one approach for the entire domain) leads to many benefits. In order to do this, the organization needs to be a participant in the process. While this comes as second nature to the IDM community, it's also somewhat the nature to also want to "choose" as to winners and losers, regardless of standards. If we look at competition, it's also interesting to look at how "lock-in" of markets occur due to initial conditions in chaotic systems.

Fairness. If a product locks-in a market because it is superior, this is fair, and it would be foolish to penalize such success.
If it locks in merely because user-base was levered over from a neighboring lock-in,
this is unfair [W. Brian Arthur]

That's why some not so very subtle uncompetitive practices occur, which in fact represent either overt or not so overt collusion.

There is a fundamental difference between collusion and collaboration, collaboration being one of the hallmarks of the evolving economy, and collusion being a fundamental element of economies in general, but ends up promoting systemic exploitation as opposed to systemic cooperation.

Short term this benefits the company, like pumping up the stock price, but ultimately affects the overall governance structure, and then the only question is who benefits and is the greater fool. All Ponzi schemes eventually fail, so why do they occur?

Of course when the regulators fail to act effectively, as in the sub-prime mortgage bubble, the promoted benefits of extending housing to a greater number of people ends up in reality, (after greed sets in) a scam where bad debt, no income, no record loans get repackaged and stripped onto Wall Street, based on some  quant's idea of risk. 

The value of what was traded was real, but after a point, broke, and had an unknowable value.  It's that point that Viega recognizes in his risk and security analysis, where good competitive (meaning value creation) mechanisms give way to value destroying bad competitive propositions that reify a position, but still pay well, if temporarily.

It's also worth noting that a cloud service is independent from
any of the participating parties, and therefore can be a neutral and
disinterested facilitator. For a long time, companies have been able to
partition their network to allow limited access to trusted third
parties, or provide a proxy facility accessible to both parties. This
practice was not lost on the plethora of folks trying to compete for the
lucrative and crucial identity management area.

Identity management services such as OpenID and Windows Live ID operate in the cloud, allowing them to
bind users together across domains. [Oram/Viega]

I raise up my mug of hot cider, "Bird in Hand Brand", and propose a toast in cyberspace from the banks of the Schuylkill with libations and honor to Chief Tammany.

While we ride in different canoes, the union that binds us is  strong if the memory of the wampum belt is honored. Before computers, before optical, magnetic drives, and flash memory, we wove our understanding in beads of purple quahog, the path of parallel time lines down the river, without the vescia piscis. without the Riemann manifold, without Penrose tiling.

Those who toil under the buttonwood tree and work on change alley will not understand this; it has no currency for them, they do not yet hear the ringing tones of a nation in harmony.

May the blessings of health be shared equally among it's people, the fish and wildlife be preserved from pollution and multiply, and the trees bear abundant fruit. May we stand by the river and admire the beauty.

The cider mill carved in stone forms a hermeneutic circle of nature and science in harmony.

The apple that can be named is not the eternal apple.

Sit for a while on the banks of the Schuylkill, on the circular pattern of John Bartram's cider mill. Imagine Franklin and Bartram conversing about the useful knowledge obtained by Bartram in his dedicated identification of trees and plants, and the mason's methods used to carve the nether stone from the Wissahickon schist bedrock forged in the Cambrian era.

As they enjoy the cider pressed fresh from the mill, the harmony between the scientific method and the natural process is only eclipsed by the crisp autumn day and the cycle of return.

Billion years of carbon, we return to the garden to enjoy the harvest.

Interesting theory reported in the NYT about the Higgs Boson coming back to break the Hadron Collider. There's an interesting parallel between the loop of the big bang theory, i.e. that because the initial release of energy "broke" the storage containers, thus making Tikkun or repair of the vessels necessary, the LDS view of the Native Americans, and processes involved in the upcoming end and restart of the long count period of the Mayan calendar. Should be interesting to see if the cycle continues or kicks up a notch of understanding.

Here's the gedanken experiment regarding the collider.

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1681680

Dr. Delacroix relates what happened Fri Night, Sat Morning after Katrina.

"Rested for an hour. One of the new doctors went to the airport to see
whether all of these new doctors and supplies could be of use. He
returned with a grim face, saying that the airport was atrocious and
that he had been told by FEMA that we could provide no medical care
because we were not government-licensed physicians. I spoke to EOC/DHH
in Baton Rouge (Jimmy Guidry) who was extremely upset at the FEMA
Bureaucracy. FEMA denied help at the airport from well-staffed local doctors even though people were in need. Bureaucracy at its best. We could not even place an IV on the patients. The same patients who I had been caring for for days. What nonsense."

Tony described at the Tao of Attributes meeting about doctors who were driving boats and could not attend to patients. (Patients who had been dumped by helicopters to an area with little to no supplies). A firetruck that was stopped at the Pentagon during 9-11 for the same reason.

He then explained how he needed to get critical personnel to the operations center to manage a crisis and that they would need to travel sometimes through road blocks set up by 12 different municipalities, and thus used the common credential to get through. Thus a local official would be able to check the personal identity credential (card), and get an answer back from an authority that this person was who they said they were. Better than Dr. Delacroix's experience of having guns pointed at him approaching road blocks during Katrina. So it looks like PKI works efficiently, and at a low cost in these situations where you are dealing with official actors.

What's that going to mean to upcoming generation? PKI or a Facebook login? Quite a few people wear their ID's around their necks in D.C.

Few people know the backstory of FEMA, pre FEMA  was the DCPA, and civil defense. Then they expanded under FEMA as a separate agency (getting a good reputation for efficiency) dealing with natural disasters, and then was folded under DHS, and started having coordination and political overhead problems.

Those drills when you were a kid, "duck and cover", shelters, that stuff? Scared the bejezus out of me when I was a kid. When I learned in college that actually the idea was to keep people off the roads (after all you have seen all the disaster movies right?) there's this whole science around evacuating people and trying to get them through, or around traffic.

This was the Eisenhower defense road network, based on what he saw in Germany as the Autobahn, which became the Interstates. But originally, it was about moving missiles around, and of course the secondary use for transportation.

Meanwhile all these rumors are flying around in any disaster scene, I was watching the replay from 9/11 the other day, and saw this taking place. So there's a huge crowd control problem, and how that is managed is obviously a difficult task. As an adjunct faculty member for communications for FEMA, we found the best approach was to tell the truth, but that's not how many commanders are naturally wired, it has to be a learned habit.

But all these people (the incident commanders) are trying to deal with chaos. Stuff breaks down. Alternate communications needs to be used, and people (not just other responders) need to be looked on as an asset rather than a liability. Tony was talking about people having guns and pretending to be police in Katrina, and there was some effort in getting guns out of the hands of people in New Orleans, who were presumably trying to protect their property. The stories of looters, true or not, caused FEMA to withdraw and put the police on edge.

So I have my own use case of a man who was in Katrina, and had climbed up an exit ramp to a bridge, and was met by police with guns drawn, and told he could not cross that bridge to evacuate, he was going towards one of the neighboring towns. They would not let him pass. He was forced to go back to the river, and float among the corpses until he could reach a point where he could get out of the river.

It reminded me of older times in the civil rights struggle where police would barricade a bridge and not allow the marchers to pass. And that is the question I posed to Tony, which he could not answer, what does U.S. identity mean when it comes to Force Majure and Posse Comitatus?

These are major events (Force Majure) which break systems. Posse Comitatus which limits the degree to which Federal troops can be used for law enforcement, had very racist roots in pulling back the troops that were protecting freed slaves in the re-constructionist South. Americans want some guarantees that the Constitution is going to be followed in regards to their identity, and he was making no statements what would take place under martial law. Naturally without some kind of clear guidelines about what happens when troops are deployed in terms of identity (not their identity so much) there's going to questions, and now is the time to ask them.

The point I made is that if you want to extend Identity in the U.S., it has to be in context of everything that has happened historically. If you really want to understand the requirements, you don't stop with 9/11, you go back to the Founders, and farther back than that. And accurately mind you, there's valid reasons and situations, that can get confused, especially when it comes to Posse Comitatus, precisely because arguments were linked back to the Founders, when the reality was more advanced.

But let's start with the Founders, because most people get that. "We the people" in order to form a more perfect union... There's something here. It's the start of identity. Of course, that does not get us across the bridge. But it's a start. Let's skip to the Civil War and see if it gets clearer, Union, what's that mean?

John Payton, President and Director-Counsel of the NAACP Legal Defense and Educational Fund spoke to a group of people at Oberlin on Constitution Day about who was included in "We the People". At the Founder's time, this did not include everyone, and it was Lincoln who firmly cemented U.S. identity.

"Equally significant, in the middle of the Civil War, at Gettysburg,
President Lincoln reframed the entire Constitution by forever linking
it to the Declaration of Independence by his pronouncement that “All
men are created equal” and “that government of the people, by the
people, for the people” is the central mission of the Constitution.  We
the People became a unifying identity."

That unifying identity is a powerful thing, it should not be divided into identity have and have nots.

Al-Qaeda was not able to reach Washington because people on Flight 93 had found out with their cell phones what had happened. With all our technical capability, it came down to this, we the people. The first responders at the trade towers, we the people. It was not the coordination of the Boston airport screeners, the DMV people who got the terrorists fake or real IDs, the officials that stalled the FBI investigation before 9/11 based on a misunderstanding of the separation of Intelligence and LE in the special wiretapping court, and so on, the folks who taught the terrorists to fly.  No, what was effective in the end was simply some folks with cell phones. The system failed that day, but ordinary citizens gave their lives, so that other people could live. And in Katrina, people who could help, medical professionals who tried to help were denied access.

Can we be critical of those people who are trying to wrest control from chaos? They are in fact in extremely complex, dangerous and difficult situations, where dedicated people are putting themselves in harms way.

So we can sit home and watch television?

Sorry, I don't buy it. Yes those people deserve the highest respect and admiration for doing their job, absolutely. But we are all aboard on this one, we the people, the union, and if we say stand down and let this man cross the bridge, it's for a reason that goes beyond your fear of looting and lack of control, so recognize his ID and let him pass, he's probably a friendly. It's time to think out of the box on this one. That's going to be the win on collaboration with the Web 2.0 technologies with Tony's PKI architecture, but there have to be some lines drawn on civil liberties, that are baked in.

I attended this meeting as an observer, making a few comments. It was certainly interesting to see many of the "identerati"in attendance and hear what they had to say. The meeting (which lasted two days) is available now online.

The focus of the meeting was on the convergence of Web 2.0 ID technologies like OpenID and O-auth and traditional X.509v3 certificate systems, such as the Federal Bridge PKI, and related systems such as the SAFE PKI for Bio-Pharma, DOD, and DHS. Certain technologies have been vetted by NIST with levels of information assurance, or LOA.

The focus of the problem space is attributes. I will attempt to explain it as best as I understand it, and leave the rest of the context to be filled in.

Simply put, every identity system maintains some attributes, some are edited or updated by administrators, and some can be directly changed by end users. Attributes that have been determined to be valuable are persisted, certainly depending on culture, so what attributes are important is culturally dependent. We commonly use prefixes such as Mr. Mrs. Ms. Dr., before names. (However in the U.S. we do not bestow royal titles).  This largely means that there is no purely appropriate schema for every instance, and it is difficult for a relying party to query an authoritative party, or identity provider for a definite answer. This has led to a great deal of probabilistic matching.  There are definite issues here because this means a great deal of personal information is being sent, often without adequate controls. In the larger sense the great deal of exposure of identity data gathered in this attributes presents a very large risk.

 The classic example of an attribute that an end user is expected to modify would be "Favorite Drink" documented  by Kurt Zeilinga in RFC-4254.  If you wanted to tell a consumer (meaning a consumer of information) what your favorite drink was, you could update this field.

2.11.  drink

   The 'drink' (favoriteDrink) attribute specifies the favorite drinks
   of an object (or person), for instance, "cola" and "beer".

      ( 0.9.2342.19200300.100.1.5 NAME 'drink'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15{256} )

   The DirectoryString (1.3.6.1.4.1.1466.115.121.1.15) syntax and the
   'caseIgnoreMatch' and 'caseIgnoreSubstringsMatch' rules are described
   in [RFC4517].

There a couple (actually several) very interesting ideas here. Let's take a look a just a few to avoid information overload.

First of all, the attribute is located within the directory server at a very specific place, a unique place. If that attribute is well formed it then can hold a value 

Kurt has "requested" through a RFC, which is how the IETF works, on the basis of rough consenus and running code, this Standard.

This standard is in turn based on other Standards, notably that of the ITU, which does global telecommunication standards, and is part of the United Nations.

For LDAP or X.500 directory servers that use this attribute, in this form, they can communicate (in terms of millions of people) what you think your favorite drink is, and for those systems which exchange information with other directory systems, that could, (and does) extend world wide.

Since the attribute was based on objects, (and people represented as objects) an object could have this attribute if this was allowed in the schema. For example, while I managed the schema for the Cosine Pilot server in the U.S., I could have added favorite drink, or any other attribute to the schema loaded in the root server. While this was useful in the pilot project, I only had one request to add an attribute to the schema, in order to accommodate the addition of X.509v3 certificates for email encryption for scientific researchers as part of the global energy community.

Now as part of the Tao, and the oft quoted "Tao of the IETF" which addressed issues such as what to wear to IETF meetings, (dress for the weather) and other logical approaches on how to get to rough consensus. Good practical advice.

Certainly when we have to deal with "Identity Management" (a term of diplomatic ambiguity which we invented to broaden the discussion beyond the RFC above), and add something as impossible to define as the Tao, we are broadening the discussion to include the entire universe, (or not). At least the entire universe as it applies to First Responders and FEMA folks in the use case.

Since it was so long since I had read the Tao Te Ching, I looked at Wikipedia and came across the following snippet on names.

The Way that can be told of is not an unvarying way;

The names that can be named are not unvarying names.

It was from the Nameless that Heaven and Earth sprang;

The named is but the mother that rears the ten thousand creatures, each after its kind. (chap. 1, tr. Waley)

Since Identity Management is somewhat concerned with names, names of objects, or how they are identified, but not entirely, since names are hardly unique, the concept of distinguished and common names was developed.

A distinguished name has enough elements that it is unique to that specific object. A common name is a shortened form.

This is an important feature because it means if you are dealing with an object that you can't hold in your hand and look at, that you can determine that the object is what it claims to be. We do that by "signing" objects with a digital signature, a "ceremony". Then we can compare a "hash" to see if the object that we got is the same thing. Without getting into complicated crypto, some things on the Network can be unambiguously identified (with room for error), and a great deal can not.

This is important, because the ability to network information about objects, and by inference attributes about people is that for many social reasons we don't trust this information to be in any one place. While this may not be inherently any more secure, since a purely "centralized" network system is unlikely (as opposed to a distributed system). If we took the Tao approach, the invariant or permanent name could not be named! Interesting problem. But the Tao gives us a way or path to resolve the problem when we consider the pattern of "return".

So I have been looking at the large and small scale implications of data sharing, the NHIN Connect software which acts as an open source gateway between different systems. Use cases presented at the NHIN Connect conference indicates a large number of stakeholders (including individual users) who are confronted with systems that have varying amounts of connectivity and vastly different data sets. This makes sense if you realize that a great deal of this software initially came from billing systems to deal with insurance companies to keep track of the vast amounts of data required for payment. For the systems where payment was not the major goal, the focus was on providing care. Of course, that's very simplistic, but this is painting with the broadest brush. It's an awful amount of work to keep up with medical paperwork.

Myself, I try to stay within a shared system so that it's simple for my doctor to look up my data and a pharmacy that's computerized to the point that they can fill what ever is required with a minimum amount of fuss since they draw from a common database.

My Health provider has had a limited amount of information directly available to me in electronic form, but that looks to be improving for my PHR so that it might be useful.

What I have noticed is that the whole process is getting considerably smoother. But, I'm also more sensitive as to what costs I am generating for my own health care. Asking more questions and interacting with my primary care physician to take more direct care of my health, thinking that some expense now, will save money later. Is that not a bit of the microcosm of the larger picture?

Except with the economy continuing to look grim, there are a lot of people who simply can't afford healthcare and most countries realize that it is cheaper to simply take care of this first. That's a fact. A recent book about what healthcare system we have in the U.S. turns out to be correct, we have a little bit of every country, it just depends on your economic status, your age, your military status.

For some reason some people want to put health on the same level as everything else, another product to consume, and someone whom you trust who knows how to provide that product. In reality that's true here, but probably should be limited to just that, access to healthcare should be a public good, no matter how it is delivered. And not the television attempting to scare you about everything, an over abundance of information, more like we make it closer to the goals of the profession, where someone can point with pride, and say, that procedure that you just got done, you know that was developed here. But overall, this inequity in availability of health care exacts its own costs.

This brings up an interesting thought as to what kind of feedback we have in regards to our health.

Of course the obvious indicators come to mind, weight, blood pressure, Hdl/Ldl, amount of exercise, stress, etc.

But something I had not considered was the degree that people are beginning to manage their own data streams.

There are some obvious examples for bikers, and runners, who map their runs, or rides to get feedback on a training schedule.

There are the way more sophisticated wireless monitoring devices for heart patients, but although the data is self generated, the processing is taking place remotely at a doctor's office.


But tweeting to a database to keep track of self assigned attributes?, that's really interesting and simple to do. The CENS urban sensing project, http://urban.cens.ucla.edu/ lets you keep track of attributes that are important to you, by providing a way to tweet a simple message that you can then track later to see what happened.

For example, what sleep patterns do I have, how much water did I drink, would be useful to know over a few month period. How much chocolate did I eat? These are really things that one could do with paper and pencil by keeping a journal, but with smart phones, it's a simple thing to tweet it and get some long term data.


In terms of sensors (MEMS) stuff, of course Nike has had a sensor for a few years that works with the Ipod and Iphone for runners, but the Iphone has spawned remarkable amount of apps that can combine GPS location data, mapping, into these personal tools.

The ads for Google Android equipped cell phones really hammer this home, the data is all about you, it's an extension of your personal identity.



This leads eventually to a huge personalization of data, (not talking here about the data which is gathered by cookies and data mining on the web) but user controlled for whatever Walden Pond use that one wants to dream up.

For the mass amount of personal data there's currently a tradeoff for "free services". In quotes because actually this is a quid pro quo. In terms of equity in the tradeoff of personal information being processed, people have totally different opinions, (and as many options) about how they want to share their data or keep it private.

But the mechanisms, privacy policies, etc. vary in quality. and don't scale well to the multiplicity of data points of entry.


So what is scaring people about Health Care Reform? More security and privacy for data? That's a good thing.

Getting better care at a lower cost?

Making sure that the total cost of health care does not go up to 20 or 22 percent of GDP?

Seeing that veterans get better care?

The AMA? No I think they support it.

Having to take some responsibility for ones actions that in fact lead to bad health outcomes? Getting closer.

Health care reform is remarkable in that has taken so long, and been opposed in so many different ways. It's actually a long American tradition to oppose health care reform even while there has been at the same time some significant social progress. It turns out that there have been smear campaigns that go back to the beginnings of the twentieth century, usually involving some foreign takeover. David Blumenthal has just written a book on how Presidents have  faced absurd amounts of vitriol in trying to get Health Care reform passed, I can't wait to read it.