Reed's Ruminations: A Blog by Dan Reed

On Licking the Cookies

All organizations and cultures develop their own idiosyncratic acronyms, abbreviations, lingo and metaphors. (Yes, there are probably a few similes and alliterations as well.) One need look no further than the social networks and Twitter to find a veriable cornucopia of such phrases and acronyms (e.g., LOL or OMG). The hacker community (the original one) also gifted us with the gems RTFM and UTSL.

I was once offered my first full time job in the software industry because I knew what JCL was needed to apply a PTF to BTAM. That's IBM speak for applying a software patch to telecommunications software. In this sense, Microsoft is not unique, for its worldwide employees have developed their own lexicon of metaphors and abbreviations, as well as appropriating others were useful. As a relative newcomer to the corporate world after twenty-five years in academia, several of these abbreviations, words and phrases immediately caught my ear.

Licking the Cookies

Not to be confused with the joy of double stuff Oreos or even licking the spoon, licking the cookies refers to claiming action or skill in an area but in a way that prevents anything from happening. It is intended to conjure an image of a cookie plate, where one licks the cookies then returns them to the plate, rendering them inedible by all. As the phrase suggests, you have not eaten the cookie (i.e., taken action) but you have prevented others from doing so either.

Cookie licking is endemic to any large organization or culture where multiple teams compete for credit and accomplishment. We have all seen this behavior in our organizations and occasionally been guilty of it ourselves. This is one of the more pernicious forms of non-collaborative, non-team play.

After understanding the phrase, I realized that subtle and not so subtle forms of cookie licking occur in academia as well. I have seen it in debates over academic directions, discussions of research priorities and in paper and grant proposal reviewing. Of course, academia has its own set of arcane phrases.

Eating Your Own Dog Food (aka Dogfooding)

Dogfooding is a hoary classic, well used and abused across the computing milieu. It refers to using early (alpha or beta) versions of a system you and others have developed before it is released, often as an informal tester. It's not haute cuisine or even nouveau cuisine; it's dog food, edible and nutritious, but neither elegant nor tasty.

Learnings

Finally, there are learnings, as in "Our learnings on this suggest optimization is our first priority." The first time I heard this linguistic oddity, I was nonplussed and puzzled, confused and uncertain, perhaps even mortified, but I was not chagrined. After hearing it several times, I finally asked if the speaker really meant insights, understanding, observation, or even experience. The answer, of course, was yes. Nobody I have yet asked, however, knows the origin of the usage, though I am sure the exegesis would be fascinating.

BFN, I am OOF.

Information Privacy: Changing Norms and Expectations

N.B. I also write for the Communications of the ACM (CACM). The following essay recently appeared on the CACM blog.

Our notions of privacy and security are deeply tied to our social and historical notions of person and place. The aphorism, "A man's home is his castle," captures that notion and its roots in English common law. This Castle Doctrine followed settlers to the colonies and was later codified in the Fourth Amendment to the U.S. Constitution:

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

Your family heirlooms may be secure in your personal castle, but what of the information about you that lives on the Internet? The legacy of physical place and norms around it are far less relevant.

Equally importantly, we often convolve privacy and security without considering their differences. One needs security to protect private information, but one can have security without privacy, as many world events have shown. Security is a topic for another day; let's talk about the evolving notions of electronic information privacy.

Riding the Light

That picture of you at a family reunion, squinting into the sun, can rarely be delimited by a physical location. It might be on disk two, machine nine, rack twenty three in a North Carolina data center, but it probably will not be for long.

Instead, information flows freely, in radio waves among our wireless devices and on photon beams along the fiber optic cables that connect the burgeoning network of worldwide cloud data centers, each themselves larger than the entire Internet was just a few years ago. It's cached, distributed, forwarded, copied, mirrored and indexed.

All of which suggests that we need to rethink our notions of information privacy, moving beyond concepts rooted primarily in person and place, and considering logical privacy. These issues are complex and emotionally charged, for they challenge many of our social, cultural, legal and economic assumptions. I would not presume to offer a definitive answer here. Instead, let me offer three ideas to stimulate our debate the future of information and electronic persona management in this brave new world.

A Simple Family Photo

Let's return to that family reunion photograph, captured on a smartphone and posted to a social network site. What might I, as person in the picture, wish to specify and who else might be involved?

First, I might well like to specify a bounded lifetime for the photograph, after which it would be inaccessible to anyone. Of course, the bound might be infinity, allowing it to remain in the electronic ether forever. That is the current default, as more than one person has learned to their chagrin. I have occasionally noted, tongue in cheek, that one can trace the history of my hair loss on the Internet due to just this default. (For the record, let it be noted that I am at peace with my baldness.)

Second, I might choose to define the transitivity of access. I could share the photograph with my extended family but not allow any of them to share it with their friends. Or, I might limit access to some overlapping circle of personal or professional friends, preventing viral propagation. This is challenging because our overlapping spheres of social, professional and familial influence rarely have hard boundaries, as anyone who has configured their social network privacy settings knows all too well.

The usability of specification interfaces for privacy and security deserves far more attention than it has received. All too often, the only options presented are a broad and vague end user license agreement (EULA) that one must accept to use a service or a byzantine set of confusing service configuration options whose effects are less than obvious. Privacy specifications must be made far simpler and more intuitive.

Third, I might wish to define a claims-based access policy. This is not a binary access specification, but rather a statement that this person or this entity can access this photograph for this and only this purpose. Thus, I might grant my cousin the right to look at this photograph but not to sell it, alter it or combine it with other media.

Ownership, privacy, reputation and decision making are intertwined in subtle ways. What if I posed for a reunion photograph but one of my crazy cousins was dancing on the table behind me? Who controls that family reunion photograph, me, the drunken dancer in the background, the photographer with the smartphone, all of us? The shifting nature of social relationships further exacerbates these challenges.

Let Us Reason Together

Let me end with another aphorism, "Possession is nine tenths of the law." In a digital world where images, video and text can proliferate globally in seconds, we need to rethink what "possession" means.

I don't have all the answers, but I do have lots of questions.

Battling Evil: Dark Silicon

As a boy, I remember reading all the superhero comic books, from the standard Superman and Batman to the Green Lantern, Thor, The Flash and Spider-Man. Some superhero stories and backstories were better than others, but all were escapism from small town life. For me, Thor was always a favorite, though I am not quite sure why. Maybe I just liked the head gear! The summer "action movie" season is upon us, and several of these superheros will grace the screen in new Hollywood fantasies, battling the forces of evil.

All of this made me think about the resident evil now facing us in computing – the rise of dark silicon. The very phrase sounds ominious and it is, for I believe it will profoundly reshape how we think about computing in the next decade. But before we talk about the grand denouement where our Justice League of silicon process experts, computer architects and software gurus do battle with dark silicon, perhaps we should begin with a bit of backstory.

Moore's Law and Silicon Scaling

We are all familiar with Moore's Law, namely that the number of transistors in a given area doubles roughly every two years. It is worth noting that Gordon Moore never really claimed this was a law, merely that it was a statistical truth based on observational data and technology trends. It is also interesting how a rather informal 1965 Electronics paper, entitled "Cramming More Components onto Integrated Circuits," had such impact. Over the past thirty years, the "law" has held true due to the hard work and creativity of a great many people, as well as many billions of dollars of investment in process technology and silicon foundries.

For a detailed understanding of why reducing transistor feature sizes really has been possible, I highly recommend reading the original, seminal paper by Robert Dennard, Fritz Gaensslen, Hwa-Nien Yu, V. Leo Rideout, Ernest Bassous and Andre LeBlanc, entitled "Design of ion-implanted MOSFET's with Very Small Physical Dimensions." It appeared in the October 1974 issue of the IEEE Journal of Solid-State Circuits, and it defined a scaling process that has driven the semiconductor industry ever since.

Simply put, Dennard scaling says that reducing the length, width and gate oxide thickness of transistor features by a constant k results in transistors that are k² times smaller, k times faster, and dissipate k² less power. This was Moore's Law made manifest and has let us move from feature sizes measured in micrometers to nanometers. Since the elucidation of Dennard scaling, all has been well in Metropolis, at least until very recently.

Power and Performance: Evil Lurks

As transistor sizes continued to shrink and clock rates continued to increase, we have seen the rise of high-k dielectrics and silicon on insulator (SOI) to address static current leakage. In that spirit, one of my colleagues recently remarked that transistors are now less like solid doors that keep out the cold (current flow) unless open and more like screen doors that always have some air flow (current leakage) even when closed.

Transistor density, clock rates and chip sizes have also made heat dissipation an ever bigger problem. While at Intel, Pat Gelsinger once remarked that unless things changed, chip temperatures would approach that of the sun's surface, an obvious absurdity only possible in the comic books. All of this has led to multicore designs, bounds on clock rates for most microprocessors and decreases in operating voltages.

We have also seen new semiconductor processes and transistor design options to reduce leakage current. (Intel's recent FinFET announcement is one such example.) Although I expect to see transistor feature sizes continue to shrink, with sub-11 nanometer feature sizes achievable, it will not be easy, nor will it be cheap. Lest there be any doubt, the cheap part really matters, for it is the economics that will determine if scaling continues.

These semiconductor challenges have been mirrored by an increasingly empty bag of general purpose architectural tricks. Almost all of the techniques once found in supercomputer processor designs – superpipelining, scoreboarding, and vectorization – are now in mainstream microprocessors. It is not obvious that we can extract more parallelism while maintaining the illusion of sequential execution. In fact, the evidence is substantial that we cannot.

Battling Evil: Dark Silicon

That's the backstory. Now let's turn to the true evil lurking among us, dark silicon. If one considers process scaling trends and limits on heat dissipation, the conclusion is rather obvious. We soon will have (and in many cases already do have) chips with more transistors than can be concurrently activated. The practical implication is that most of the chip silicon will be "dark" (unpowered) most of the time.

If it's dark most of the time, is it of any value? Yes, but only if we embrace our fears. This is the true end of the hegemony of von Neumann computing and the birth of a new era, one based on chips with a diverse set of specialized functional units, accelerators and tailored architectures that each target specific tasks. The mobile world has already learned this lesson, for functional specialization increases energy efficiency (more operations/joule) and allows designers to mix and match the set of functional units to target domains. Arguably, heterogeneous multicore is one, very tentative step in this direction for client and server computing.

This transition to diverse functional units, accelerators and tailored architectures will also challenge many of our assumptions about programming models and system software, as well as the economics of chip design. In this new world, hardware/software co-design becomes de rigeur, with devices and software systems deeply interdependent.

It is a different ecosystem, with different cultures and different processes. It means far less general-purpose performance increases, much more diversity in the hardware, elevation of multivariate optimization – power, performance, reliability – in programming models, new system software resource management challenges and a plethora of new applications.

More generally, with lots of silicon real estate (dark silicon), chip designers and architects have the space to place functional blocks that are used only part of the time. However, at some point (and we are arguably close to that point now), we will have deployed those functional units that cover the common tasks and adding more will have possibly deleterious effects due to larger chip size. That potentially drives the economics to smaller chip sizes, which, given foundry capacity, leads to lots of small chips. That is the collision of the big chip server world with the small chip mobile and Internet of Things world, a battle already raging.

To address these challenges and battle dark silicon, we need new ideas in computer architecture, system software, programming models and end-to-end user experiences. It's an epic struggle for the future of computing. Giants are battling, and not all will survive, nor will the survivors emerge unscathed. Heroes will also emerge, sometimes from unexpected places.

I think I left my hammer, Mjölnir, around here somewhere. Do you feel your spider sense tingling?