Must. Overcome. Fear. Of. Posting.

Over the last few months I’ve been thinking a lot about the ways in which I communicate with my friends, family and colleagues.

Looking now at my clock widget, I have clocks telling me what time it is in Singapore, Brisbane, Los Angeles, New York and Berlin covering time-zone difference of 18 hours. The reason for having so many clock widgets is that the people that I want to communicate with on any given day are spread across a pretty wide region and it’s good to have some idea when they’ll be awake.

For the most part, I communicate with people via quick skype/gmail chats, tweets/facebook, or short emails. I’ve been keeping this up for a few years now and I’m realizing that many of those that I’m close to really only ever get to see glimpses of half-thought-through snapshots of the ideas in my head. It’s becoming clear to me that this is a pretty terrible way of keeping those relationships that I really care about.

One fix to this problem is, I guess, to hit that little green call button more often. Unfortunately, the times when I can freely call someone rarely fit with the times that they can take that call.

Of course, I could spend a lot more time writing long emails to people. I actually did this for a while. The problem that I found is that as the time between emails drags on, those emails get longer and longer and I get less and less likely to actually ever send it. On top of that, the absolute social awkwardness that is me makes me feel weird whenever I email someone out of the blue. Especially when you haven’t spoken to that person in years.

One of the things that I find very strange about becoming an increasingly old physicist, is that I regularly butt my head up against some new problem for a while only to realize that I knew the solution years ago and had completely forgotten that I knew anything about it.

When I lived in Austria I really struggled with my poor German speaking skills. My inability to speak the language meant that many of the conversations that I had were either short, or I couldn’t convey full essence of whatever it was that I was trying to communicate. I guess people probably felt the same way whenever they spoke to me in English.

I found that while I was in Austria I increasingly turned to the internet to get my daily fix of communication. For the most part this began through my old blog, Quantumbiodiscs, and then later as part of the LP hivemind.

While blogging didn’t always help me to keep in touch of others, those that new about the blog could easily keep tabs on me.

So, why did I (effectively) stop? I guess there were really 2 things that, together, cut down my blogging activities.

The first was that I moved to Bristol. In Bristol it was a lot easier to communicate with people on a daily basis – so much of my ranting was done to my colleagues and others in the lunchroom as opposed to random folk on the intertubes. Also, as time went on in Bristol I became much more involved with my work. Increasingly I found that I was more likely to read a paper or to think through some whacky QI thoughts than to spend time putting together a good blog post.

The second thing was that I began to develop a bit of a fear of posting to LP. Now, don’t get me wrong interwebs, I really enjoyed writing for LP and working with all those involved. They really are a fantastic bunch of people. I also think that the LP collective worked wonders during 2007 Oz election campaign. Unfortunately, it seems that this was clear to many other Australian political tragics.

By the time I joined LP it was already getting pretty big. That didn’t bother me so much, in fact I thought it was a great thing (hell, I still do). But at some point after the 2007 election I made that horrible mistake that Wile Coyote always made, I had run out over the edge and then looked down.

After the 2007 campaign there was a lot of academic analysis of the effect of the new interwebish media. It became clear from this work that LP was very widely read and also had a large influence on the media at that time. While this might seem like good news, to my addled brain it made me think that maybe I should spend longer than 15 minutes writing a blog post and actually put some serious work into my blogging.

I guess gradually, the realization that I was throwing some fairly raw thoughts out there spooked me. The net effect of the spooking, work and life in general pushed my output down to zero.

Now, as time goes on and I realize that moving home every couple of years is actually taking a toll on my relationships with people that I care very much about I realize that, possibly, my writing can help the situation. I’m reluctantly realizing that I’m rarely ever going to be able to spend long afternoons chatting with my dearest friends over (good) coffee. But maybe if I keep this blog up to date then at least they might have some chance of keeping track of what’s going on with me.

Advertisements

Matt Hastings is guest posting over at Tobias’ place

For those who are into the adiabatic model of quantum computation I suggest going here and having a read.

No more Surfdom

It seems that Tim Dunlop has pulled the plug on Road to Surfdom. One of the first and the best Australian political blogs.

For a long time Road to Surfdom was one of my favourite blogs and it played a huge role in dragging me into the Australian political blogging scene. I’m sure that I wasn’t the only one that was heavily influenced by this blog.

All of those who contributed to Surfdom deserve a lot of thanks for all the hard work that they put into it. Believe it or not, running a blog and consistently coming up with high quality material is really hard work and I am ever so grateful that they made the effort.

Christian Kerr takes a swipe at Larvatus Prodeo

I just thought I’d throw up a link relating to the clumsy swipe that Christian Kerr took at LP on The Australian’s website.

I condemn Larvatus Prodeo and their so called “influence”

I just thought I’d try to get in there to be the first to denounce LP for being an influential blog.

Damn you LP, you tall poppy you!

/sarcasm off

Indiana Jones IV trailer hits the nets!

Gizmodo has just thrown up the unreleased trailer for the new Indiana Jones movie!

Update: Now it’s on YouTube.

Weakly Abstract: February 4 – February 8

An idea that I’ve been toying with for a while now is to write a weekly post on my favourite quant-ph arXiv paper which has been released in the last week. I’ve decided after literally minutes of consultation with my office mates that this new regular post should be called "Weakly Abstract".

I expect that the Weakly Abstract will appear on Mondays, though this first week is going to be an exception because I traveled a couple of thousand kilometers yesterday.

I’m not really sure what format the "Weakly Abstract" is going to take. I plan to experiment with a few different styles over the coming weeks. Ideally I’d just like this post to be a bit of an open forum for people to discuss the offerings from the previous week or so on the quant-ph arXiv. I’ll probably get the discussion started by pointing to my favourite paper, or the most scited paper on SciRate each week and then people can take it from there.

I doubt that I’ll always take this super-seriously and I’d like it if people keep the discussion as light and humorous as possible. Be polite but don’t worry too much about keeping the tone too professional, I want this thread to be relatively accessible to all but, of course, if you want to throw down some serious math be my guest. In short, a chatty and jokey atmosphere is highly encouraged.

Oh, and if anyone has any suggestions for papers they’d like discussed can feel free to email me or to let me know in the comments.

So, here goes…

For the first Weakly Abstract I’d like you all to point your browsers towards "Hamiltonian Quantum Cellular Automata in 1D" by Daniel Nagaj and Pawel Wocjan (arXiv:0802.0886):

Hamiltonian Quantum Cellular Automata in 1D
Authors: Daniel Nagaj, Pawel Wocjan
(Submitted on 6 Feb 2008)

We construct a simple translationally invariant, nearest-neighbor Hamiltonian on a chain of 10-dimensional qudits that makes it possible to realize universal quantum computing without any external control during the computational process. We only require the ability to prepare an initial computational basis state which encodes both the quantum circuit and its input. The computational process is then carried out by the autonomous Hamiltonian time evolution. After a time polynomially long in the size of the quantum circuit has passed, the result of the computation is obtained with high probability by measuring a few qudits in the computational basis.

This result also implies that there cannot exist efficient classical simulation methods for generic translationally invariant nearest-neighbor Hamiltonians on qudit chains, unless quantum computers can be efficiently simulated by classical computers (or, put in complexity theoretic terms, unless BPP=BQP).

I really like this paper. It is well written and with 11 scitations, Nagaj and Wocjan’s paper is currently the most scited paper of last week on SciRate.

The key point of this paper is that the authors constructively prove that it is possible to perform efficient quantum computations by time evolving relatively simple, separable, low dimensional initial states by a fixed time-independent 1D nearest neighbour translationally invariant Hamiltonian and measuring in a fixed basis. As is pointed out in the second paragraph of the abstract, this result sets up a compelling dichotomy – either BPP=BQP (that is classical computers can simulate quantum computers) or the folklore belief in the condensed matter physics that the time evolution of all low-dimensional 1D translationally invariant nearest neighbour systems can be simulated classically is wrong.

A similar, yet more complex, construction that proves the basically the same concept was presented in an earlier paper by Vollbrecht and Cirac (arXiv:0704.3432v2) which I also highly recommend.

In Nagaj and Wocjan’s constructions all of the control required for the computation is programmed into the initial state of the computer. In a sense, all of the description complexity of any "circuit" that is implemented is pushed into the initial state of the system. The compelling point is that this can always be done for efficient quantum algorithms with a separable input that is in a computational basis state. All of the superposition and entanglement that is normally associated with the "power" of a quantum computation is performed via the time evolution by the one, simple, time-independent entangling Hamiltonian (and for efficient algorithms the amount of time you have to wait for an answer isn’t very long).

This result seemed really surprising to me at first, but I guess in a lot of ways it isn’t really that dissimilar to the 1-way model of quantum computing. In the 1-way model all of the quantumish part of a computation is thrown into the initial state (normally thought of as a 2D cluster state). The circuit complexity is pushed into the measurement stage of the computation (albeit with a bit of feed-forward control).

Wheras in the model of Nagaj and Wocjan’s the complexity of the circuit happens, mostly, in the initial state. Though there is of course the tricky point of having to wait enough time for the computation to be able to give the correct result, this is essentially something that can’t be sped up.

In summary, I think this paper is tops because it gives us a real target for resolving the problem of quantum versus classical computation. If we can demonstrate a method for simulating the time evolution of the Hamiltonians involved in Nagaj and Wocjan’s constructions then BQP=BPP (and I’m out of a job). Though it should be noted that a recent paper by Schuch et al (arXiv:0801.2078v1) suggests that given current simulation methods in computational physics this is very unlikely.

These results, to me, gives one of the more compelling arguments for pushing ahead with development quantum computers. While we now know that simulating the static properties of generic quantum systems (like finding the ground state) is really hard both classically and quantumly, all of this recent work is suggesting that when it comes to simulating dynamics we may be able to use quantum computers to do a lot more than we ever can with classical devices.