Anti-Science Party

For background, I was raised as a Republican, but have been utterly disappointed by the GOP for at least a decade now. This is for many reasons, but the one that is motivating this post is that they have chosen to become the anti-science party. This is particularly evident when it comes to climate change, but it's also visible in their long-standing crusade to inject creationism into school curricula. The climate change issue alone would have been enough to force me to abandon the party, but they have actively pushed me away over the last few years with their anti-gay rhetoric, anti-intellectualism, and now thinly veiled racism, and totalitarianism.

But let's focus on just the anti-science thing for a moment. Unfortunately, Republicans don't have a monopoly on catering to beliefs that are, shall we say "highly improbable" of being true. For example, the idea that GMO foods are unsafe is a banner that the left wing of the Democratic party has been carrying for quite some time, despite plentiful evidence to the contrary. If you view politics through the (limited) view of a left/right spectrum, then the Green party is just the left-wing analog of the Tea party, and the Greens are (quite unfortunately) not highly concerned with whether science supports their positions.

Why do I give the anti-science lefties a pass, while damning the anti-science right-wingers for their views? Well, I don't give them a pass, but I do dismiss them. I can safely do that because despite appeals from some of their most passionate (but unscientific) base, the Dem party hasn't made banning GMOs part of their platform. The most the that the D's have been willing to do is establish an (entirely voluntary!) organic classification and advocate for GMO labeling. The cries for banning have largely been ignored by the party. Anti-vaccination is another non-scientific position from the left ... or is it? Turns out that it's not really a left / right thing -- there are plenty of right-wingers who are anti-vaxers, too. My most-definitely-Republican mother among them. And again, the Dem party has mostly ignored that issue (as has the GOP, thankfully).

So both sides of the ideological spectrum have some decidedly anti-science beliefs floating around. The difference between the Republicans and the Democrats is that the R's have incorporated anti-science beliefs into their DNA. And the GOP doesn't just hold views that are unsupported by science, they actively attack it. It's a big difference. Furthermore, the biggest, most politicized anti-science issue that Republicans are championing is dangerous. Remember reading about the Dust Bowl, and what an environmental/financial disaster it was back in the 1930's? It was nothing compared to the sort of problems that unchecked global warming will cause. The dust bowl was able to sort itself out after merely a decade or so. The consequences of our present-day greenhouse gas emissions will be with us for millennia. From the IPCC (SPM 2.4):

Many aspects of climate change and associated impacts will continue for centuries, even if anthropogenic emissions of greenhouse gases are stopped. The risks of abrupt or irreversible changes increase as the magnitude of the warming increases. {2.4}

Warming will continue beyond 2100 under all RCP scenarios except RCP2.6. Surface temperatures will remain approximately constant at elevated levels for many centuries after a complete cessation of net anthropogenic CO2 emissions. A large fraction of anthropogenic climate change resulting from CO2 emissions is irreversible on a multi-century to millennial timescale, except in the case of a large net removal of CO2 from the atmosphere over a sustained period. {2.4, Figure 2.8}

Machine Learning Motivations

I've been doing a fair bit of reading on machine learning lately, and I need a dumping ground for my notes, thoughts, and todo lists. A place to think-out-loud. Normally, I'd just stash it all in a text file, but I have a great track record for periodically losing everything on my computer. This blog has proven to be remarkably durable in comparison, and I think my future self may find some of this interesting.

So why machine learning? It's a silly thing, but I recently picked up a retro-style space shoot-em-up game on the cheap. I played it for a few minutes, made a little progress, and thought, "You know, what would really be interesting is trying to get a computer to play this. Certainly more interesting than playing it myself." Back when I first started learning to program, I had had a similar thought about Sudoku, and wrote a fairly simplistic Sudoku solver as my very first hobby program. I was incredibly proud of that Sodoku program, and loved the motivation that doing my own project provided. It was a great way to dive in and learn. I'll be trying to reproduce that experience with machine learning.

But still, why machine learning? Just because it sounds like a fun way to get a computer to play a game? Well no, not really. A co-worker at Insitu recently left to go work for Google and will be working with the team that's helping to drive adoption of deep learning throughout Google. That certainly got me thinking about it more, but still, that's not the why either. It did lead me to read an excellent N.Y. Times article about Google's AI work, which led to yet more thinking. In the end, the full "why" is a confluence of two factors.

The first is climate change. It is extremely important to me that we do everything within our power to address the issue. I believe it's the "World War II" of our generation -- something for which the whole country should throw all of our efforts towards for as long as necessary, so that our children, and grandchildren, and great, great, great grandchildren will have the possibility for a bright and happy future. It's an issue that has weaved its way into many of my personal choices, such as becoming vegetarian (mostly), buying a Nissan Leaf, added insulating to our house, vacationing close to home, and even our choice not to have a second child. Climate change hasn't been the only consideration, or even necessarily the first, but its always been on the list.

I think personal choices matter, but am well aware that they can only accomplish so much. The nature of climate change requires that in the long run any solution must be achieved through global politics, with the whole of the planet agreeing to work together. But in the short run, there's still a great deal to be done technologically to help grease the skids -- to ensure that the path to a better future is as frictionless and easy to move towards as possible. These are things that groups of people working towards a common goal (i.e. a company) can accomplish. So I've been starting to search for a new field in which to apply my skills, and for ways to amplify the (positive) climate impact of my future work.

The second factor in "Why machine learning?" is that currently, I'm just a programmer and a team leader/manager for a small team working on a relatively basic product. I think I've become a fairly good programmer, but I am at best a big fish swimming in a small pond. I am not amazing or outstanding (or egotistical) enough to think that I will ever be at the true cutting edge of technology, solving mindbogglingly difficult problems for the very first time. I've known enough genuinely brilliant people to know that I can not count myself among their number. Still, even knowing that I may not have what it takes to be a leader in a field, I know I'm smart enough, and capable enough to be a fast follower. I'll give myself that much credit, at least.

I think machine learning, particularly deep learning has advanced to the point where it needs a healthy crop of "fast follower" types. It's very powerful, and has really advanced in the last few years due to algorithmic improvements, but also due to improvements in GPU computing, and in large-scale cloud computing. The learning phase of machine learning is quite data and compute intensive, yet once trained, the resource requirements are much, much lower. That, shall we say, "elasticity" in resource requirements can really benefit from the cloud cost model. Cloud computing has granted machine learning the sort of scale where it can accomplish things that traditional programming has extreme difficulty with, particularly speech recognition, translation, and image recognition.

I'm fairly convinced that ML is poised to revolutionize a number of fields over the next decade or so. I think it's a huge, powerful hammer of a technique, yet relatively few people currently know how to wield it. Yet the tools are there. The ground's been laid. It just hasn't occurred to most programmers that they could pick it up.

I've not dived in enough to know with certainty, but first impressions (i.e. a few days worth of reading) suggests that getting machine learning to work reasonably well is perhaps just a few times harder than getting a RDMS to work well. You have to learn a new paradigm, and there's lots of important details to learn before you can achieve anything ambitious, but you can put the tools to work without needing to learn anything too mind-bendingly difficult. Where the comparison breaks down is that ML will often need large amounts of carefully labeled training data, which is quite a lot harder to acquire than just a bunch of hard drives and a high-end server. Still, if you can overcome that requirement, I think that ML may be the sort of technique that eventually most programmers will have in their toolbelt, rather that always requiring a Ph.D'd specialist. That "eventually" may even be now, as far as potential goes.

So my goal is to learn enough about machine learning that I can confidently apply it when suitable, ideally without needing to become such a specialist that is the only tool I know how to wield. At the least, I want to learn enough that I can explain in detail why or why not various ML techniques can be applied to a given situation, and then bring in a specialist to help as appropriate.

Then, with the shiny new hammer in my belt, I want to go looking for domains where there really are a bunch of nails just waiting to be struck -- particularly domains which could potentially have a high impact on climate change. From where I stand right now, I think there's likely to be a lot of them.

Kindle 2

I rediscovered pleasure reading recently. I'd never stopped reading, but for a time I had drifted away from reading books for entertainment and had only allowed myself the time for stuff with a practical angle. Mostly, this meant reading books about programming, or robotics, or Personal Rapid Transit (PRT), or what have you.

A few weeks ago, a pair of books arrived in the mail unbidden. Not totally unbidden of course--they were old additions to my wishlist on Bookins--but still, unexpectedly. One of the books was Self-Made Man and looked far too interesting to set aside, even though I didn't really have any of that elusive 'time to spare' stuff lying around. That weekend, when I probably should have been 'doing stuff' (ideally, 'doing useful stuff, productively'), I read it.

I loved it.

More than that, I loved the experience of reading it. It's a fascinating book--you'll enjoy reading it too--but I loved reading it for a couple reasons that really shouldn't contribute, but do. First there was just the joy of procrastination. Never let anyone tell you that they don't enjoy procrastinating, if they didn't, they wouldn't be doing it. Second there is a definite sense of accomplishment when the last page is turned and the book is finished. Placing a just-read book on a shelf in your home is like placing a trophy in a display case. Yes, from now on it will just collect dust, but every time you look at it you'll be reminded of the experience.

When we moved to Santa Cruz, we moved to a smaller apartment, and many of my books were sold in our yard sale. On some level that was easy to do. I'm not materialistic, I'm not a hoarder, and having strangers give me money for the privilege of reducing my clutter is a delightful pairing. On another, more sentimental level, it was hard. Although I rarely reread books, having them line my shelves was a reminder of the books I've read--the works that I've loved (books that I merely 'liked' tend to make their way out of my home rather quickly). Irregardless of the occasional pangs, it was necessary, and I don't have regrets.

Finally, I come to the topic of this post. I bought a Kindle 2. I'm still not sure why, but maybe I'll figure it out as I write. Compared to Book, it's ... well it's really expensive is what it is. But if you can swallow the $360 price tag, then it's pretty nice. Being able to adjust the text size is great, as is the ability to read without either holding the book or wedging it open somehow. The eInk display is very legible in all sorts of lighting, and it's no heavier than most paperbacks. On the downside, it's a little too big to slide into a coat pocket, and way too expensive to not worry about it getting lost, damaged or stolen. Back to the plus-side, it has free (lifetime?) wireless, a web-browser and built in Wikipedia shortcuts. I'll save some money on books, since ebook prices are cheaper than dead-tree versions, but I don't have any delusions about recouping the device's price tag that way. Having books instantly delivered is ridiculously cool. But perhaps its biggest plus side isn't going to be realized for another year or so.

Next time I move, I'll get to keep my books. I think that alone was enough to make me click, "Confirm Purchase."

I also have some hope that it will allow Katrina be more comfortable when reading, thus allowing her to read more. Furthermore, I hope that just owning the Kindle will push me to read more full books, and spend proportionally less time surfing the web. Either of those alone would probably have been enough to seal the deal. As is, I'm considering buying a second one.

But probably won't. $360. Sheesh.

The past never existed -- it's purely a fantasy of your mind. In fact, you didn't really just read this.

Time will tell. She's a gossip, really.

Genetics and natural selection are Nature's programmers and design team, respectively. They're awful at their jobs. After four billion years, they still can't make animals that will reliably flush a toilet in a public restroom.

I used to think I would stay young and fit forever. I realize that's not possible now, but at least I will always stay young.

Another Craigslist Post

Face it. Your clothes have been scattered all over your room for the last year -- no matter how many more New Year's resolutions you make, you're not going to start hanging them up now.

So why not break down and just get a cardboard box to shove them all in? Even better, a bunch of boxes! You could put some of the smaller boxes in a bigger box, and the whole thing would start to look pretty slick. Organized, even.

But wait! Instead of cardboard boxes you could do it up right and build the whole thing out of wood... maybe add some little metal handles and paint a wicked flaming skull and crossbones on the side. Bitchin'.

Or you could skip all that carpentry shit (Damn, man, what'd you do to your thumb?) and go straight to airbrushing a half-naked chick on the side of this FREE dresser.

Call 530-750-1000 (or email) to lay claim to this awesome stack of boxes in a bigger box.

Dimensions: 24" x 14" x 41"

I don't have a picture handy for the blog, but did for the Craigslist post and the picture was part of the punchline -- it's a little girl's dresser.

Woot inspired

Let there be Light!

Now you no longer have to live like a Caveman, squatting in the darkness and groping your way to the fridge by the feeble light of your cellphone. With this powerful 3-bulb floor lamp, you'll take one giant step up the evolutionary ladder and you too will start enjoying the most wondrous invention since fire.

Features:

• Three separate switches! Giving you 2^3 = 8 possible lighting configurations!
• Swivel heads! Great for chasing cockroaches around your apartment without ever leaving your easy-chair!
• Rugged construction! We haven't killed this thing yet, and I doubt you'll be able to, either!

All for the low, low price of absolutely free*!

Call 530-750-1000 or email!

* No, we won't pay for your gas to come out here.

The lamp's already gone, but I enjoyed writing that far too much to just let it wither and die in Craigslist's database.

"It is a fearful thing to love what death can touch."
-Unknown


The worry that arises when contemplating my own mortality is dwarfed by the true fear that the comes when considering the mortality of those that I love.

I rediscovered a genre today. Interactive fiction. That is, text-adventures, like the games Infocom used to make. You know, Infocom. Zork.

. . .

If you're seriously trying to tell me you've never even heard of Zork, how the hell did you find my Blog?

What do you mean you don't even know me? Get the hell out of here! Sheesh, can't even blog in private these days.

Anyways, if you already know what I'm talking about, check out this. It's pretty good, and it's online so you don't have to mess around with a downloading an interpreter. If you really don't know what I'm talking about and are still here reading, then still go there. You type what you want to do. Keep your sentences real simple, a couple of words long -- you're dealing with computers and they're dumb as rocks. Think verb noun level of simple and you'll do alright.

The written word is an amazing thing. Play with it.

It's Veteran's day weekend, and we're spending our three days poking around Yosemite and are spending our two nights at a perfect little hostel called the Yosemite Bug. It's off of highway 140 nearish to the park, and I can't recommend the place enough. It's picturesque, quiet, friendly, and at least in the off season, pretty cheap. Great food and a warm fire at the lodge, complete with a lovable dog and cat wandering the premises. Every so often the dog, Yin Yang, will roll over on his back and patiently wait for his belly to be rubbed. The cat's taken a more direct approach and just hopped straight into our laps. Careful, like Pfeiffer, she drools when she very happy.

Tomorrow, we'll be heading into Yosemite park and doing some day hiking. We haven't chosen a particular trail yet, but last time we were here with my brother we hiked up past Vernal falls on the Mist trail. It's autumn right now, so the waterfalls aren't likely to be so extrodinary as then -- maybe this time we'll see what other bits of spectacular this place can offer up besides flowing or falling rivers. Half dome is an unlikely destination as the ladder has already been taken down for the year. Climbing to the very top isn't necessary to make it an amazing hike, of course, but it's far less climatic to end the trip at the Saddle.

Enough writing for now. Heading to bed and will decide our destination in the morning.

I've started using Skype for my long distance telephony. Skype's a VoIP service, similar to Vonage except that they're more PC centric, charge per minute instead of monthly, and don't advertise nearly so much. So far, I like it -- good voice quality, and cheap -- which are just about the only things I care about for outgoing long distance calls.

But here's an interesting thought about Skype -- I think it's going to lead to an explosion of Nigerian 419 style scams, except voice-based and coming out of India instead of Nigeria. Like Nigeria, India has an exhange rate that will make any sucessful scammer into a millionaire. Unlike Nigeria, it has a large population with fluency in English and training in how to "close a sale" provided by all the corperate call centers there. Skype's network in particular will be appealing for scammers because of the low cost, excellent sound quality (sounds professional) and virtually untappable network.

Hope I'm wrong.

There's a great website that I peruse often called Engadget. There's not much to it, they collect and summarize news stories about tech stuff, including about robots. Their schtick for the robot section is that they're trying to warn off the upcoming doom caused by the rise of the machines -- didn'cha see the Terminator movies? Ain't it plain to see that it's comin'? Better treat 'em nice while they're still weak and incapable of dominating us all.

Today I got a firm reminder of just how much power the 'bots on display here have, and just how quickly the reins can be yanked from our hands -- maybe the Engadget schtick isn't such a comedy bit after all. One of the bots, "Tommy" was meekly making its way through the tunnel when it went went stark raving mad. It floored the throttle and kept it there, accelerating out of the tunnel, swerving to miss a stack of tires and finally crashing into the barrier wall at a angle, going somewhere in the range of 45-60 miles per hour. Two people standing one the far side of the wall barely got clear before the 'bot hit it and pushed it inwards by about a meter.

On a more technical level, the people next to me had a good guess as to what had happened -- the inertial navigation system was giving bad or no data, and when the bot lost GPS signal it thought it wasn't moving. It increased the throttle to try to start moving... and increased it some more, and more... confident that it wasn't moving at all as it barrelled towards the wall.

The hardware of a robot senses the world around it -- vision, velocity, position, pressure, whatever. The software tries to mold that data into a coherent and true model of the world around it. When that vision is warped, or when its model is drastically askew, the result isn't what we should call a "bug".

It's lunacy.

My blurry little friend is back and is running laps around the perimeter of the car's cabin. I'm pretty sure he wants my head