Letters from Ψ

Announcement: IQC's Newest Member

2010-02-13T16:53:00.001-08:00

If you don't mind this post being far more of the announcement variety than the standard sort of philosophizing you've come to expect from me, I'd like a few minutes to share some exciting news. One of the most stressful parts of being a graduate student is not knowing what comes next, such as wondering where one will go for a PhD program. Unfortunately, the structure of the PSI program exacerbates this somewhat by accelerating things such that we needed to apply for PhD programs not long after arriving. Well, for me, that uncertainty has been pushed back at least 3 years, as I have been accepted as a PhD student at the Institute
for Quantum Computation at the University of Waterloo. Starting this summer, I will be a member of IQC and will be doing research I love in a place I adore. As for what happens after that, well, that's a worry for a much, later day. For now, I am celebrating my participation in one of the best quantum computation research groups in the world.

Open Source: Right Tool for the Right Job

2010-02-10T20:16:00.001-08:00

I've talked before about the importance of computers in science, but that's a bit generic, and is fought with cultural problems. In particular, as the use of computational models in gains traction in fields frustrated by intractable problems, we will see more and more misunderstandings, such as the recent non-scandal with climate science modeling code. As has been discussed elsewhere, an important part of dealing with these kinds of problems is to ensure that all of your simulations are verifiable and auditable. This is part of why the ideals of the open source movement mesh so well with the ideals of academic research, and why the use of open source software in academics is something to be encouraged and promoted. Just as open access and open data are starting to be seen in fields like biology as a natural counterpart to the principles of verifiability and reproducibility (for more on this, please see the ever-insightful Michael Nielsen), it is only natural to see open source as a valuable component of good, solid research.

Of course, I don't ever expect a good researcher to put using open source software above doing good research, but these two goals are not, contrary to popular perception, orthogonal. Sure, SAGE is still behind Mathematica, and sure SciPy is still behind MATLAB, but this is something that is changing quickly. For many projects, having easy access to standard general-purpose libraries like GTK (for writing user interfaces) is a big enough gain that SAGE and SciPy are already more attractive. Where these projects tend to fall behind, however, starts to become evident when one compares the availability of user interfaces and toolkits. While SAGE's web-based notebook interface is innovative and wonderful, it lags behind the native Mathematica interface in terms of ease of use and visualization tools. Similarly, while SciPy allows you to use your favorite text editor, whatever it may be, along with the power and versatility of a modern POSIX-style command line, the lack of a cohesive integrated development environment (IDE) similar to that provided with MATLAB adds a steep learning curve to the otherwise excellent library. Part of the promise of open source, however, is that when one sees a void, they can go try to fix it. When you don't have the right tool for the right job, you can work with a community to change that. That's why I'm starting the SciCore project to develop a good IDE for SciPy. While I myself am too busy to ever get it done on my own, by working with a community of other intelligent people, we can provide the right tools for our peers to get work done. In starting this project, I join a long line of academics who, when faced with a lack of a tool, sought to change that.

A Story of Institutional Failure

2010-02-05T09:10:00.001-08:00

Today, I wanted to share with you a particularly banal example of institutional failure that I spotted while walking home recently.

So how is this an institutional failure? Look at the poster again: here is an advertisement for an official university-sponsored event for students living in university-owned dormitories, and yet the e-mail address advertised is provided by Gmail. In general, people do what is easiest, and so this means that the University of Waterloo has made it easier for its own employees to use a third-party e-mail provider than to set up a temporary address at uwaterloo.ca. Of course, to be fair, there are very, very few institutions that get this right. Moreover, I don't mean that UW has an incompetent IT staff (they don't), an overly restrictive bureaucracy (they don't), a lack of funding (they don't) or any other such handicap. Rather, what I mean is that the very way that the institution is structured decouples the provider of a service from the necessity for that service, and that the result of this decoupling is that it is easier for university employees to depend on an external service that cannot be audited and that is not responsible to the university than it is for employees to depend on their own internal services.

Of course, one might object at my choice of example. After all, do we really need auditability in the sending of Valentine's Day candygrams on campus? Almost certainly not. That said, I am a physicist, and part of the methodology of physics is to build "toy models" that illustrate a physical effect or concept, but that is small enough to study in some nice and analytic fashion. In that sense, looking at the failure of a candygram promotion to exploit internal services gives us a clue as to how larger and more pervasive failures occur in large and competent institutions. When the various parts of an institution become decoupled from the purposes that they are intended to serve, then intra-institutional economics can cause problems.

For a bigger example of this at work, consider Microsoft. As an ex-vice president recently noted, the various departments and organizational units had become fiefdoms unto themselves that obstructed, rather than served, the individuals within the organization, and thus ran counter to the goals of the institution as a whole. That a corporation as large, as profitable and as influential as Microsoft is so prone to such organizational failures should give us pause. Our global economy is dependent on every minute detail of how these institutions conduct themselves, and that there are so many ways that organizations make it easier for employees to do the wrong thing than the right thing should, frankly, scare the hell out of us.

On a more personal note, that these kinds of failures can occur in competent, world-class academic institutions is an important thing to be aware of. I firmly and deeply believe that in academics lies the means to ensure that the future of our society is a future that we can be proud of, and so it falls upon me, as an academic, to do what I can to ensure that academic institutions work as well as they possibly can. To that end, toy examples like this candygram failure lead me to recognize that in every action that I do, I should be aware of the economic incentives that I help to set up. When I see cases, even small ones, in which our technology is an obstruction rather than a tool, when a bureaucratic process incentives doing things in a shoddy fashion, or any other case where incentive structures have become perverted, I must act to try to correct that, even if only in small ways.

Brand New Me

2010-02-03T13:41:00.001-08:00

OK. I promised a while ago that I wasn't dead yet, and it's true: I am alive. The title to this post, however, is significantly less accurate, but that's another story. So where does this new year (and new month, even) find me? PSI courses are going quite well, thank you. In fact, so well that it's time I get back to sharing stories about my life with the rest of the world. To that end, I've restarted the blog, and will be posting more (hopefully interesting) things soon.

Not Dead Yet

2009-09-29T05:42:00.000-07:00

Just to let my dear readers know, despite the lack of regular updates, I am not yet dead from homework and lectures. Rather, I am pleasantly occupied learning as much as I can about physics. We have now moved into the second round of courses, Quantum Field Theory I taught by Dr. David Tong and Statistical Physics taught by Dr. Leo Kadanoff. Please stay tuned for a more substantial update, coming soon!

Weekly Highlights: These Are Really Highlights

2009-09-12T19:31:00.000-07:00

My previous two attempts at Weekly Highlights have been notably free of actual discussion of the weeks that they're supposed to be highlighting, so I'm going to try to fix that problem by talking much more briefly about several different items of interest.

Start of Relativity and Quantum Theory. This week marks the first week of the Relativity and Quantum Theory courses. These lecture courses are the first two in the set of Core courses for the PSI program. In one week, we've gone from Lagrangian mechanics to the Schwarzchild metric, and from spin as a group construction to momentum-space representations of Gaussian wavefunctions. The lecturers, Malcolm Perry and Robert Myers, are both wonderful instructors that have spared no time in challenging the heck out of us.

Homework. As we've started into our core classes, it's only appropriate that we have homework. So far, we've turned in two assignments, gotten one returned and started working on the next two. Fun times ahead, as these homeworks aren't throwaway. Our instructors make us work for our answers. While there is lots of help available for when we get stuck, that doesn't make it easy at all.

The Company We Keep. It's not just the lecturers who are world-class around here. It's also the company that we keep in between classes. Last week, we attended two talks by Leonard Susskind, and we spent meals talking to the likes of Lee Smolin and Neil Turok. Even outside of the realm of physics, the Perimeter Institute attracts interesting people; for instance, we spent Thursday's lunch talking to Jorge Cham, PhD and author of PhD.

Student, Teach Thyself. Naturally, for such a curious bunch of people as ourselves, we aren't satisfied with just getting world-class instruction in physics from leading researchers. No, we need to teach ourselves. Through reading groups and student-run lectures in topology and Lie groups, we have been taking an extremely active role in making sure that we are getting the background needed to get the most out of our PSI courses.

Looking Forward. Of course, I can't highlight the week that was without mentioning what's on the horizon. We have much to look forward to in the next week, including things both academic (more wonderful lectures, several new collquia), and social (the U2 concert in Toronto). Not to mention the inevitable surprising events that occur around here.

So that's what our week past, and what we look forward to. Maybe not so philosophical as normal, but hopefully seeing a week laid out like this helps one appreciate what life is like for a PSI student.

Weekly Highlights: On Computers in Science

2009-09-05T18:25:00.000-07:00

This week, we did two major things in lectures. First, we learned the basics of using Linux, LaTeX and other such technologies. Second, we started the unit on general relativity with a review of classical mechanics and special relativity.

For now, I want to dedicate my attentions to the first part. As a part of the funding and support package for PSI, we've been given laptops preloaded with Ubuntu. This may seem like an odd decision for a program focused on theoretical physics, but it is my contention that this decision is incredibly logical and responsible. Computers play an important role in society, and in particular, in the sciences. Even for those scientists that do not use computers for numerical simulation, computers enable many profound mechanisms of communication and collaboration.

In particular, computers offer scientists the use of tools ranging in complexity and technological sophistication from e-mail to instant messaging to Google Wave. These tools provide users the ability to share and discuss ideas at unprecedented speeds. To turn away from such tools is naught but folly for those tasked with delving into Nature's most fundamental puzzles.

Computers also enable us to publish our findings faster and to wider audiences than ever before. Tools like LaTeX allow us to format our findings without having to worry with the minutiae of every formatting decision, instead focusing on the structure of our papers, articles and books. These publications can then be distributed, reviewed and discussed via the web, using tools and sites like the arXiv and SciRate. Even social networking sites like FriendFeed have proven to be wonderful for discussing research, as is witnessed by the many feeds dedicated to scientific research.

Of course, I slipped a bit of a fiction by earlier, though. We are moving into an era of science where no one is exempt from the need to resort to computational models, even if only indirectly. Even those of us who shy away from writing MATLAB toolboxes for our problems still depend on the models tested by numerical simulation. Moreover, there are many opportunities for saving time by exploiting tools like computer-aided integration using Mathematica, Maple or SymPy. Obviously, these tools can be dangerous if one over-relies upon them, but that in no way detracts from their potential usefulness.

The arguments and rationales laid out here are not lost on the designers and directors of the Perimeter Institute or the PSI program. PI uses the web to distribute videos of all their seminars, and uses MediaWiki to coordinate and organize research projects, PSI lectures, outreach events and even building maintenance. Our PSI-provided laptops offer us a platform from which to use these powerful collaborative tools. Perhaps in the future, I will take the chance to discuss why it's important that this platform is based on open source software, but for now, suffice to say that PI gets it— they understand how to provide scientists with the tools and environment to get research done. Their use of computing technology is just one more facet of that understanding.

Weekly Highlights: Deriving the Universe

2009-08-29T14:04:00.000-07:00

Physics isn't done from an armchair. Since physics is ultimately about the real world, there can be no development without experimental verification. Thus, we cannot only use logic to guide us to how the world should be. We must also verify how the world is. In fact, our use of logic is only justified at all by the empirical observation that it works.

On the other hand, once we have confirmed something via experiment, it is often a highly non-trivial task to figure out what our observations actually imply. Since we believe that the universe makes sense (that is, described by a logically consistent set of rules), our observations never exist in a vacuum, but instead constrain which sets of rules nature can be described by. For instance, we know that any description of the world must reduce to quantum mechanics in those cases where we have experimentally observed quantum mechanics to be a correct description of the universe. Similarly, we must demand that any correct description of nature reduce to general relativity in any of a wide range of different regimes where we have confirmed behavior predicted by GR. Before quantum mechanics and general relativity were developed, we likewise knew that in order for any theory to replace Newtonian mechanics, it must reduce to an approximation of Newtonian mechanics in, for instance, describing most of the properties of the orbits of planets.

Sometimes, this process of figuring out the limitations and restraints placed on future theories leaves us with only one choice for how to proceed. In those cases, we truly can work out things about how the universe works by looking at past data and theories. As an example, in his lectures at PSI, Dr. Nima Arkani-Hamed demonstrated how combining general relativity and quantum mechanics necessarily forbids any theory which gives rise to massless particles having greater than spin 2. More astoundingly, he demonstrated that the only possible theory allowed for spin 1 massless particles within a combination of QM and GR is Yang-Mills theory.

While these kinds of arguments cannot be used to solve all of physics from a blackboard, they serve as a very striking argument as for the importance of theoretical physics: with a combination of math and experiment, we can derive much of the dynamics of the universe. What more could we hope for?

It Takes a Multi-Talented Villiage

2009-08-28T16:11:00.000-07:00

I think I shall make a Friday recap series a regular feature here, but since I haven't formalized that yet, I first want to take a few minutes and speak to something I've only danced around before: scientists are people, too. Taken together with the fact that science is social, it shouldn't be a surprise that I spend a lot of time socializing with my fellow PSI students, forging friendships and connections that not only help us out here and now, but that should continue to be useful for the rest of our careers.

What I find truly astounding about my peers, however, isn't just the academic abilities and backgrounds for which we have all been selected. Rather, it is also how vibrant the community is. At any given time, there's likely someone going around his room in a weird kind of skates, playing his electric guitar, practising on his violin, maintaining a blog, taking pictures, going on a jog, watching and discussing films, learning a language, or any number of other things that aren't strictly academic in nature. Over the coming weeks, I hope to shine a bit more focus on these activities, but for now it suffices to say that we are a diverse and vibrant group as any you'll find. Even more, we all work hard not just to learn how to calculate scattering magnitudes of three-graviton interactions, but to cultivate our own projects and skills.

It's beautiful. The sounds of my roommate's gentle guitar tunes playing and filling the space around me. All the while mathematicians walk around juggling, and kites go flying in the breeze.

I find it to be an inspiring place to be: the physics, the math, the llamas and the people. All the splendour of learning mixed with the joy of exploring one's hobbies and projects. I am inspired not only to learn about and study QFT, GR, quantum info and string theory (to name but a few topics), but also to engage in my own side projects. For instance, my girlfriend and I have been working on writing some background material for what we hope to one day turn into a series of graphic novels. (Don't get too excited-- the project is still in its infancy.)

All in all, Waterloo is a stimulating place, all the more so with the wondrous Perimeter Institute enabling and supporting us. I promise to talk more about the science here soon, but for now, thanks for letting me guide you through some unexpected ways in which PSI is awesome.

First Week: What Makes a Scientist?

2009-08-20T11:56:00.001-07:00

Now that the first week is almost over, it's time for some quiet reflection about what the Perimeter Scholars International program is all about. On the surface, there's a very easy answer: PSI is about making great students into great scientists. As we reflect on this answer, however, it becomes more opaque and problematic. After all, isn't that the same goal as any other good Masters course? What makes a scientist "great," anyway?

For now, I want to restrict my focus to the latter question. If PSI is all about making great scientists, then we first have to understand what a scientist is and is not. In some sense, that has been exactly what the first week of lectures, delivered by Dr. Kari Dalnoki-Veress, has been all about. We have discussed the typical tasks of an academic researcher, such as publishing papers and giving talks, as well as the motivations underlying these tasks. To be sure, being a good scientist must imply at the least a competence with the typical tasks of a scientific researcher. But I digress.

So, the question of the week remains: what makes a scientist? Curiosity? Integrity? Discipline? A mathematical mind? All of these answer some small part of the question, but none are complete. Besides that, one would expect that a great student already possesses these and many other traits desirable in a scientific researcher. No, to make a great scientist, we must add experience and knowledge. Moreover, a student must be armed with the social support needed in order to perform research. After all, no interesting subject of study is so small that one brain can contain it, regardless of its quality. A scientist must also have the opportunity to succeed and to study.

Putting it all together, then, the PSI program gives us, its students, the opportunity to become experienced and knowledgeable members of our academic communities. To do so, PSI has encouraged an atmosphere of cooperation: competition as it is known in traditional courses distracts one from the accruing of experience and knowledge, and can poison potentially useful social connections before they have the opportunity to bear fruit. Even in the few instances where competition is inevitable, it is not the focus. This shall be one of the primary hooks that we use to understand how PSI is different from other Masters courses, but that's a subject for when there is less homework due.

The Path Forward

2009-08-17T18:54:00.001-07:00

Who Am I?

2009-08-16T11:18:00.001-07:00

This blog isn't about me, but about the Perimeter Scholars International (PSI) program offered at the Perimeter Institute starting this year. On the other hand, since the PSI program is very much so an interactive program, I can no more describe it from a detached perspective than I can measure some observable of a quantum state without projecting that state down onto a measurement basis. But I digress.

So who am I? As with so much, the answer depends on one's perspective. From a computational perspective, I am any device capable of predicting my actions with perfect accuracy. From the perspective of national governments, I am a citizen of the United States. More interesting to you, likely, is that I am a student. In particular, I have recently completed an undergraduate degree program in physics, computer science and mathematics, and sought to join the PSI program in order to further learn about quantum computation and information. To my nearly indescribable delight, I was accepted. Thus, I am now continuing my studies in fair Waterloo, Ontario.

As a student, a large part of who I am is tied up in my research interests such as quantum computation and quantum information. I first became interested in quantum computation upon attending a lecture by Dr. Robin Blume-Kohout, and have since followed that interest to a plethora of different venues, the most recent of which being PSI itself. Moreover, I have let this interest lead me on a tour of the deep interconnections between physics and computer science. Indeed, it is fair to say that this is precisely why I am interested in quantum computation: it illustrates so well that computation is a physical process, and thus that the seemingly disparate fields of physics and computer science are not distinct at all, save for those arbitrary divisions made to assist in our taxonomy of the universe. Though pictures painted from the perspectives of physics and computer science will no doubt differ, they are at the end both pictures of the same universe. As such, I cannot help but be fascinated by this interplay between various perspectives on the universe and how they converge in quantum mechanics, one of the most beautiful and successful constructions of theoretical physics. (For what is more beautiful than a theory that works?)

Of course, I am not strictly a student, but also more generally a human being. Thus, I am also characterized by the company I keep and the manner in which I socialize. In leaving my prior home of Fairbanks, Alaska, I left behind many wonderful friends who have helped me to realize that it is not enough to simply describe the world-- one must live in it. This is precisely the same recognition that seems to have been made by those at the Perimeter Institute, for we are provided with many opportunities to explore the recreational sides of our lives. Even something so basic as a deck of cards becomes a tool for socialization, and thus for becoming a more well-rounded person. It is for these reasons that I am happy to find my peers at PSI are a jovial and intelligent bunch that share in my passion for learning. Moreover, I have been very glad to find that my peers are a wonderfully multi-talented bunch, showing that music, sports, writing and many other pursuits can be part of a fulfilling life as a student of science.

Once classes begin on Monday, I will have much, much more to say. For now, though, an introduction shall have to suffice.