“Do what it takes to solve the problem, and no more”: a small talk with ‘Bunnie’ Huang

Follow

As a Free & Open Source Software (FOSS) geek, I was really looking very much forward to anything coming out from the linux.conf.au, better known as LCA. And for a huge bouquet of surprises, this year’s edition was one.

Along with Radia Perlman and Bdale Garbee, there was Andrew ‘bunnie’ Huang. He is a very interesting person, and his activities go beyond just-basic-nerdiness. Huang happens to be the person responsible for the jailbreaking of Microsoft’s Xbox360, something he wrote a book about in 2003. Moreover, he holds a PhD in engineering from the MIT. What I really like Bunnie for is all his great coverage of Chinese manufacturing, while sourcing suppliers for Chumby. Chumby is a gem: a totally open and hackable device designed from the ground up complete with open source hardware. This embedded computer provides internet access through a WiFi connection, and is able to run a wide range of software widgets once connected. The hardware being open, everyone willing to use Chumby is encouraged to get into the device and make it his/her own.

You may be blinking a bit thinking “OMG, how come I don’t know this guy!” I believe many actually do: you may relate if I mention the open source Geiger counter he crafted after the Fukushima Daiichi disaster hit Japan. Bunnie’s keynote at the LCA, ‘Linux in the Flesh: Adventures Embedding Linux in Hardware,’ was all in the tune as it “embraces the idea that Linux is not just for desktops and servers, but exists, in one form or another, in millions of smartphones, routers, set top boxes, and other hardware appliances that we encounter in everyday life.” This translates Huang’s push-and-pull approach to open hardware: he contributes original open designs and also liberates closed designs, for the general good.

Australian Science spoke to Bunnie, and the discussion was thrilling for me. I’m also a biology geek, so you can easily imagine my excitement when Bunnie came to talk about biology. As of 2012, Huang lives in Singapore and – beyond 100Mb fiber internet connection and various other everyday sweets life there seems to offer, – it also has a Genome Institute where Bunnie did an internship studying regulation mechanism in the interaction between bacteria and their viruses. In the video below, Bunnie very nicely explains about somewhat theoretical knowledge he already had from Harvard and how stunningly different doing experiments is. I was really excited when he started talking about the equipment in the lab: it is archaic, “as from the 1980s-1990s!”, but it can undergo loads of improvement. My guess is many will relate when Bunnie tells about the tedious series of optical density measures to follow the growth of a bacterial culture: “you cannot go for lunch because you are obliged to be around all the time. It could be so much easier if the machine was tweeting you when the culture reaches the desired value…”

I was also delighted to hear that Bunnie does not exclude a path of lab equipment development and design. One thing which fascinates him is the possibility to sequence and analyze DNA outside of the lab. According to him, such activities at home are far from science fiction which opens great horizons for health analysis and diagnostics. Bunnie suggested that a cheap DNA analyzer is amazing enough, even without anything else added, and insisted that such equipment could potentially start a new era in healthcare at home.

Noticeably, Bunnie discussed the future of the Moore’s Law. In other words, the observation that “over the history of computing hardware, the number of transistors on integrated circuits doubles approximately every two years” – the Moore’s Law – is about to hit the end. According to Bunnie, technological advancements in general, and more specifically the decrease in size of devices and the doubling of processor speed will cause the slow-down. As Bunnie pinpoints it: the next objective is the optimization what we already have.

Interview Transcript

How are you finding the conference?

Yeah. The conference was great. It’s been really interesting. The caliber of the participants here is really high. This is my first time here at the conference, and so I really didn’t know what to expect, but everyone here seems to have at least picked up a soldering iron. Some people have actually built some very impressive things. I’ve learned a lot too actually. It’s been good going to the sessions and hear other people’s ideas and inputs on different Linux related things.

This one guy who’s also building and Fpga+ ARM-based computer, and it was really interesting to hear his talk. So I mean, that’s very close to the kinds of things that I do. And then I got to meet the openstack guys who are doing the cloud computing environments, and that was an eye opener to see that it’s actually available and then all of that model rocketry stuff, and just talking to people and meeting people, it’s been really great.

Background, past projects, experiences.

In my past life, like before when I was in grad school and stuff, there’s some work I did on the Xbox that people know me well for. But since then, I’ve done a lot of other things. I went through a phase of doing chip design. So, I was designing WIFI chip sets, and then I went to a company doing nanophotonic integrated circuits, where I actually had designed some of the first high-speed modulators on bio-silicon, you know silicon drivers. And then after that, I just decided to go with more consumer electronics. I did a company called chumby, where we wanted to sort of bring the cloud internet to all kinds of consumers, all over the place, and make it very approachable and affordable. So, we’re embedding Linux in alarm clocks and picture frames and T.V.’s all that sort of stuff. And then, since then I’ve been kind of on my own. In the last year, I’ve been sort of independent, living in Singapore, trying to look around for new ideas and doing things that are interesting and good and fun.

Living in Singapore.

Singapore has kind of got everything that a geek really needs. I mean, they have 100MB fiber connections to the home, for a very reasonable price. I have two of them. And food is quite cheap. Rent is very reasonable, particularly if you can get into public housing, and you don’t have to have a car. You know, my book here is very reasonable. So a lot of things that you would have to worry about, you know if you’re trying to do a startup or be independent are taken care of by the infrastructure there. So it’s been quite good actually. I can really focus on innovating and trying to do something new.

Passion for biology.

When I was in college, it was sort of a coin toss, whether I was going to go to biology or computers. So, I’ve always had this lingering interest to learn more about biology. So, recently, I, through the good graces of the local institute in Singapore, the Genome Institute in Singapore, I did an internship with the infectious diseases group. And there I was doing some research on the regulatory mechanisms of phage versus e-coli. Coming from a hardware background, you sort of know that you don’t really know what you’re talking about until you’ve actually done it, right? You have to really get your hands dirty, or get your hands wet, as it may be in the biology lab. And so I had read a lot about the theory of how informatics works and sort of, how you might do a transformation or how you might do a knock in clone or a knockout clone, and it was really interesting to just get in there and try to do some experiments, see them fail, see how they work, see how unexpected results can be very interesting. So, it was a really good experience.

Inchip is now finished, but I hope to someday be able to return to that lab or some other lab and do a little more work. I have some ideas for things that I would like to try to explore further.

There’s two observations. Number one is that it feels to me that biology, lab equipment is kind of caught in the stone ages, like the 80’s or 90’s or something like that, right? So, to do a culture growth, right? You want to hit a particular OD. You have to go back and check every hour, two hours or something like this when you get close. And so, when we do these things, we wouldn’t be able to eat lunch. We’d have to, like, sit there and eat outside the lab, get all suited up, check, come back finish our lunch or something like that. Seems silly that building has not connected the internet, and doesn’t just send you a tweet when it’s actually done with growing to a particular level. And all the equipment is there in the lab and all of the pieces are there. They just kind of need to be merged together into a piece of equipment. And when you look at what is behind it, it’s a shaker, it’s a heater, and it’s a photodiode with a photo emitter. These pieces are not rocket science, it’s been hard getting them all together in one box. So, I think that there’s a lot of room, low hanging fruit on just the basic equipment level, that can perhaps help science automate more and people be more productive. Although I fully understand that biologists have their protocols and they want to stick with it, it’s hard to sell them on new things. But, I think there’s room there.

When you look at what is behind it, it’s a shaker, it’s a heater, and it’s a photodiode with a photo emitter. These pieces are not rocket science, it’s been hard getting them all together in one box. So, I think that there’s a lot of room, low hanging fruit on just the basic equipment level that can perhaps help science automate more and people be more productive. Although I fully understand that biologists have their protocols and they want to stick with it, it’s hard to sell them on new things. But, I think there’s room there.

And the other thing that I find it’s interesting, because biologists start, they’re essentially reverse engineers. They have a complex system, and they have to perform these experiments on these black boxes, trying to figure out how it works, or how it breaks , how to change it through these very empirical observations. Computer scientists tend to be the opposite. They like to design from the bottom up. I do both. I do design and reverse engineering, and both processes get you into these mental traps of how you think about the world. Like when you reverse engineer, you get some particular dogmas on how things must work, because you’ve done a lot of work, and so the rejection of the biology community, like short RNA sequences when they first came out as being ever functional or having any role, sort of is a symptom of the dogma that can happen in biology. You really have to sort of work hard to overcome certain thinking’s. But, if you come from the computer science background, and you say, `ok, well you know, what can these short RNA sequences do? Well they can template onto mRNA’s and prevent [transcription], or whatever it is. That makes sense, right? I mean, that seems to be a perfectly possible thing. And so, I think maybe, applying more of the other kind of thinking to look for other potential areas could be interesting. And so, that’s a very vague thought, but I think that there could be some contribution there.

I think it’s more just getting the idea out there that more people from engineering backgrounds should get into science. It’s interesting, it’s cool, it’s definitely really fascinating and very complex systems. There’s a lot to really get into, but a lot of people from my background would think that it’s intimidating or scary to learn about a whole new system of biology. But actually, it’s another system. It has a lot of rules. It’s like learning a new programming language, in some ways. I mean, if you stay out of the lab, it’s like learning a new programming language.

I would really like to be able to make a contribution, but I don’t know if I can. That’s yet to be seen. I’m still kind of looking for the right angle or direction in that area. I do look forward to, generally the industry getting more automated and having more services. The biology today, the DIY bio today is a bit like, step one get FR copper rodded and then etch traces and then drill the holes, and then you can build your circuit, right? It’s not like build your circuit. You have to do all of these little things before you can actually do the experiment you actually cared about, right?
I think hardware today is really taking off, because you can just spit out a circuit board with a design. It’s all just bits, and it comes back as atoms and you’re like hey, I can put my components on and do my thing.

Biology is getting to the point where, you can start ordering plasmids in. You can get competent bacteria, you can get all the bits and pieces that you need sort of outsourced, so you can actually just say, “I want to do this particular experiment,” and you can put them together in a single step and then actually get to the result you’re looking for. I think more of that will help more people participate.

If we can get either sequencing or programming sort of into the home environment, so it’s really cheap, and basically you don’t have to think about. Like, if your child has a sniffle, you just put some snot into the little thing and it tells you what your kid has. Or if you don’t know if your fish is like actually the fish you bought, or if you want to know where the grapes for your wine came from, you want to know how your gut is doing, you can sample your poop and put it in there and see how your bacteria is changing over time. I think there’s a huge amount of information just in the DNA of the environment around us, and you know, I don’t have any scientific evidence to prove this. I have seen and read a lot of papers that are supporting this particular point, but I think that once people can start sequencing their environments, then it enables a lot more people to sort of contribute things.

It would be really fascinating to see if you can sort of aggregate huge amounts of citizen data on sequence, so you can see epidemics and how they progress. You can get details on how people’s diets change over the seasons versus various trends and stuff. I think there could be a lot of interesting stuff that could come out from getting more of that out into the public. I don’t know if people will be designing custom bugs any time soon. Oh, I don’t want to have oil, I’m going to grow some oil bearing bacteria. Or, I want some drugs; I’ve got to make some drug bearing bacteria. I don’t know if that’s in the cards anytime soon.

It would be nice to have a personal tricorder. I think with sequencing alone, you can get a lot of information about your environment. Of course you only have one genome and that doesn’t change, hopefully. But your gut bacteria and your skin bacteria are constantly changing, and your environment is constantly changing, and those we’re finding have a greater impact on our health and our wellbeing than we previously thought. And those could be really interesting pieces of data to have.

Any other branches of science?

I’m pretty interested by astrophysics. I think every geek sort of looks to the stars and says `wow,’ the curiosity is awesome, and I was just reading about the Huygens probe and all of that sort of stuff. There’s a lot of really just fascinating stuff that happens in that area. And the amount of science that’s being done, the quality of the science and the amazing precision that we’re able to measure; the wobbles of stars and stuff like that to see if plants are out there, or whatever it is, seclusion from very distant stars is just, it blows my mind, so.

Can we read any of your papers?

I wrote a paper about the Xbox, actually. It was an academic paper, which is part of the reason that I got it relatively leniently because it was recognized by the academic community as a contribution, it wasn’t just some random yahoo doing hacks, which it’s unfortunate that it has to be that way. It should be that people should be able to do science in their own home and not have to write a paper to be approved by the, sort of priesthood or whatever it is. But, you know, it is what it is, I suppose, and that should change. And then when I did the 40 gigabit modulators, in silicon, I got a paper about that in some conferences, and I’ve written some journal articles and stuff like that.

The Xbox Hacking

The Xbox stuff was incidental actually; it was an incident that happened in a lab, right? It wasn’t really something that I had intended to really do, it just happened, because I was curious and I found a result that was cool. So, I wanted to make sure that I wasn’t just falling into the, sort of the pigeonhole of the Xbox guy, or the reverse engineer, and so I spent a fair bit of time, I mean I do un-design, but I also like to do design as well. I’d like to be remembered as someone who can do both, right? Not just taking apart but putting things together. So, you know the silicon hacking stuff was great. That was a good phase of my life. About every four years, I kind of like reinvent myself and do something completely different, and then the chumby stuff was great, because, you know, I got to give something to the world through hardware designs and I find that very satisfying to kind of share my results and bring it to the world. I don’t know what the next thing is. I mean, definitely, I have an itch for biology, right? I would love to make a contribution there, I really don’t know if I can. It’s a big hill to climb. And I also like art a lot as well. So, I’d like to do something, not necessarily technical, but maybe just something beautiful. So, there’s a few things left for me to try and play around with, but I don’t know what’s next yet.

Future Predictions

The most significant thing that I see happening is that Morre’s law is ending, and probably will come to an end in about a decade or so, and so what will happen is that these things won’t get any smaller or faster, and your computers won’t get any faster. You won’t get anymore RAM. You won’t get bigger hard drives, all that sort of stuff. So that, everything that has driven the internet ecosystem so far, we push crappy software. We wait two years and it runs fast.

We enter a different area of optimization. Where if you want something to be better, you actually take the same code base and the same feature set, where you make it more reliable, you make it run faster, and that’s completely different sort of mentality from the just pure sort of new features and sort of information bloat kind of age. So I think that it will be a lot more room for individual innovators in that space, because it’s a lot easier for a very bright person to take an existing product and really polish it and make it beautiful and heirloom quality and add value; versus trying to create this whole new, like new phones. Like, one person can’t build an iPhone, right? But one person can build a beautiful, well optimized app or make a huge contribution to a web browser, or something like that, so that would be a big change.

I wouldn’t say it’s a brick wall, it’ll be a little bit softer than a brick wall. But like every exponential curve, like you go through any growth curve of any population, it’s an s-curve, right? It eventually, has to level out, right? No exponentials are forever, and we are hitting physical limits. Anyone who designs chips will tell you, we’ve been hitting them for a long time, it’s getting much, much harder to make these changes. I think that people who are into computers right now, who want to have a viable business in the next ten years, should be thinking, maybe where they want to land, right? We’re so busy right now of sort of where we want to go, and we’re just, the next one, the next one, the next one, the next one. And then what will happen is someone will reach for that next rung on the ladder and there will be nothing there, right? And that guy will, maybe, fall of the ladder, right? And sort of maybe slow down a bit and will sort of ask, where do we want to sort of stop and hold on and really you know, put some pegs down and hold on and figure out what’s next.

There’s a huge biological analogy, every populations sort of classic, fox versus rabbit island stories, you, like know where the rabbit population explodes and then flattens out and then collapses and then the fox population does the same. These kinds of things happen.

I don’t see, like for example, graphing technology or quantum technology or superconducting technology catching up fast enough to really fill that gap. When those things come about, there will be another little jump, right?
But even so, we’re already at atomic scale lithography, right? A lot of these things, I mean atoms, we haven’t really found a way to really beat those yet I think.

Humans just want new things because we’re humans. We like beautiful new things. There should be more of an emphasis on that end of technology as opposed to just cramming more features into a little box.

Thoughts on Search Engines

When Google started, I loved using it, because it’s simple, it’s just search as compared to Altavista, or yahoo, there was like all this stuff and portals. But, as every service evolves, it seems inevitable that they kind of clutter up and start to add a lot of features. I think I like my search engines simple, right? I don’t use any of the Google cloud services, because I don’t really like them. I mean, it’s neat that they can sort of customize your life to what they know about you. It’s also kind of freaky, and it weirds me out, so I think also, there’s a huge amount of, kind of linguistic there, I mean you mentioned Baidu. I kind of, just for shit’s and giggles, I’ll often times sort of put one search into Google and one search into Baidu and just look at just the amazingly different results, and a lot of times I can’t find it in Google and may find it in the other search engine. And that’s also partially just because of linguistic bias. There’s this other half of the internet that no one ever sees inside this country called China with this huge firewall. And it really is a different world on the inside of there. There are whole services. The firewall not only keeps people on the inside from getting out, it also keeps people from the outside getting into certain services, is what you’ll discover. There’s a lot of really interesting things that happen on the internet in China that we just can’t find out about because we’re outside the firewall.

If you could time travel…

The question depends on part, if you believe that travelling to the past can actually have an impact on the future that we know now. If I’m just going to go to the past as a tourist, then that’s not as interesting, because that’s history, right? If I can actually make a difference, then I would pick a point in the past. I would think that past time travel may not impact the future that we know today. So given that windy posit, I would want to go into the future, because I’m going to die someday, and I’m really curious where this planet ends up, right? Do we actually get to the point where we actually manage our population growth, that we sort of settle in? Do we ever reach for the stars? And do we ever get to these points? It’s sort of the story that I really want to know the ending to, so a lot of times, in books, I’ll sort of cut to the end and peek at the last couple of pages, it would kind of be something like that. I’d pick something fairly far out, like maybe even millions of years out, right?
You know once the continents have shifted substantially, and I always wonder what would happen to these mega-cities that are in places that you know will be underwater or will raise up another thousand feet, and how do humans deal with these kinds of things that we’ve never had to deal with before? You know, I would be really curious to see how we handle these challenges.

Make a wish!

All I really wish for is that I can help people in some meaningful way. I don’t know what it is. I think it’ll find me when the time comes, and I hope to be ready to do it. And maybe I just wish that my purpose finds me someday and I can fulfill it adequately, when it comes.