home
on exploration, introspection and creation

Archive for the ‘changes/cyclical’ Category

Side View Mirrors

Monday, January 3rd, 2011

There are a couple of things I’ve always done my own way (mostly because I had the luxury of not having anyone tell me the standard way to do it) that ended up being a better way of doing things. My favorite example is how I arranged my car’s side view mirrors. I would turn them outwards as much as possible to minimize (or in most cases eliminate) the blind spot. That way as soon as the car disappeared from my periphery, it appeared in my side-view mirror. As soon as it disappeared from the mirror, it appeared in the rear view mirror. This was particularly useful in what I considered to be the primary purpose of the side view mirrors — to help you switch lanes.

Only a couple of years after I started driving my friends told me that people set their mirrors to show the side of the car to have a sense of perspective (a reference frame) and to make parking easier. I didn’t feel that I needed to see the side of my car in the side view mirror — parking is not a common use case and the frame isn’t necessary once you get used to the setting.

Apparently a couple of years ago the DOT began recommending the way to set up the side view mirrors the way I’ve been setting them all along. It felt good…

Our Unchanging Voice

Sunday, January 2nd, 2011

Our voice may undergo slight variations all the time but it remains pretty much unchanged for the majority of our adult life. Is it just a fluke, or are there good reasons for that? For example, would a voice that’s easily recognizable help our offspring find us more easily?

Either way, it’s a good thing for the music industry; I can hear Roger Waters sing at a concert thirty years after he recorded an album and have an authentic experience!

A Nerd Generation

Friday, December 3rd, 2010

I was a nerd (I still am). I went to great lengths to carefully and painstakingly create an image of myself as a nerd. It took a long time and it took many sacrifices, mostly in popularity in elementary and middle school.

But what I’m seeing now is the emergence of an entire generation of nerds: people are becoming more and more computer-savvy. They blog. They laugh at xkcd comics. Technology is no longer this inaccessible cloud of mystery.

People like me don’t like that precisely because of the amount of effort we put into being different, into being special.

A Recession Special!

Tuesday, November 30th, 2010

I’ve seen these around a year back or so in the windows of some shops. This reflexivity of the retail industry struck me as strange. Was the industry just as aware of the previous recessions as they were happening? Or was it simply not yet so exploitative of them?

It’s probably the latter: companies find themselves having to find more and more elaborate and complex ways to draw in the customer, and taking advantage of the overall mood of the consumers seems just another in the series of such tricks.

The Philosophy of Reductionism

Wednesday, November 17th, 2010

I’ve been longing to write this post. It describes most closely how I make sense of the world.

Throughout many posts, I’ve shown examples of some simple concepts that I think are fairly universal. Those concepts are related in a kind of hierarchy. For example, here are three concepts related to one another:

  • Change: change is good; it’s fundamental and more powerful than any of us; it happens all the time (despite our perception bias related to viewing things, for example history, in a very narrow way)
  • Cyclical behavior: a lot of the change the happens is cyclical (in fact, if one is to take an Occam’s razor view of things, the simplest change in nature is a cyclical one because of the balance that is held between many competing factors; or in math, the simplest function that changes all the time is a sine wave). New ideas are just permutations of old ideas; we often take diametrically opposite views and switch back and forth many times
  • Equivalence: things are instances of higher concepts; those who can see it (we call those people conceptual thinkers, can make more out of the world because they can take the specific things they learn everyday, convert them to learnings about the concepts, and then apply the concepts back to the specific.

If you take the concept of equivalence to its logical conclusion, you will realize that everything is related in some kind of hierarchy. In fact, this idea of recursively reducing concretes into concepts is a very powerful one — you can build an entire life philosophy on it. Let’s call it reductionism.

According to reductionism, you begin understanding that everyday complexity can be lessened by relating things to one another. In other words, by taking concrete things, creating equivalence classes of them by grouping them by which concepts they represent, and then grouping those concepts together, you can travel up that ladder where the concepts are few, simple, and very fundamental. The feeling of understanding the fundamentals of the world is a very satisfying feeling. It can also help you make decisions: start with the fundamental concepts, derive the consequences, and keep going until you get to the level of specificity you require. In a way, reductionism is a wonderful framework for knowing what to do, and it’s a wonderful way for you to feel connected to everything.

Of course, there is a trade-off implied in reductionism. The higher up the hierarchy you go, the bigger the distance between your thinking and everyday life. This means that to make specific decisions (and, operating in a very concrete world, we have to make specific decisions every minute of every day), you have to do a lot of thinking: derive a lot of information from the few highly conceptual ideas. While some people I know can do it very well and almost automatically, it seems to me that nature prepared us to deal with the concrete very well — by giving us relatively more scratch space (a kind of cache to keep the details in) than computational ability (there’s only so fast that we can derive these concepts). It probably makes sense, evolutionarily — when you’re chased by a predator, you want to be able to trust your intuition rather than re-derive the idea to jump on a tree from the concepts of survival, physics, and the physical characteristics of the predator.

There are other caveats too. There is more than one way to create a hierarchy of concepts, to reduce a set of things into a much smaller set of more abstract things. There is no right answer when it comes to the most fundamental concepts (after all, those are the different philosophies that, just like apples and oranges, cannot be compared) despite what people tell you. There is no canonical arrangement of all things in the known Universe in a hierarchy of concepts, although a poster that shows one example of such a thing would be a wonderful idea.

In other words, reductionism reshuffles the risk: from millions of tiny errors you could make in the realm of the concrete, to one humongous error you could make in the realm of the super-conceptual. A small difference in the definition of the concept at the very highest level propagates down the ladder in a nonlinear way and can produce an entirely different picture of the world (and thus can easily make you pick a totally opposite view to the one you had before the correction).

What if, despite these caveats, we want to reduce everything that’s around us to as few concepts as possible? At first it seems pretty easy. We reduce a lot of behavior to human nature; we reduce nature to evolution; we reduce the fabric of the Universe to a small set of rules. We reduce the different religions to one concept. Then we reduce the concept of religion and science. We reduce art to feeling (synthesis) and science to understanding (analysis).

In fact, I believe that we can reduce anything to a set of two concepts that are opposites. Above, synthesis and analysis are opposites. Many things can be reduced to good and evil. Other good opposites which things can be reduced to are change and stasis.

All these concepts are themselves an equivalence class. Let’s conveniently call them yin and yang. So we can reduce the infinite number of objects, ideas, thoughts, words into just two.

Then what? Can we reduce Two to One?

We can, but reducing Two to One is infinitely more difficult than reducing Infinity to Two.

Programming Unix/Mac circa 2006

Tuesday, November 2nd, 2010

The computing world changed tremendously between 2000 and 2006… in 2006 I completed a second version of the reference guide, this time focused on Unix and Mac superuser experience (less about programming, more about terminal usage). Most of the stuff here is timeless and still works beautifully, thus yet again proving that good technology never becomes obsolete.

Programming Mac/Unix circa 2003-06

Work Interactions of the Future

Monday, October 25th, 2010

We’re so used to the current model of getting work done that we rarely think about what it used to be like. Nowadays, we spend most of our people at our desks, staring at computer screens, where we interact with the entire world. The computer is our window into everything.

We used to talk to people in person more, and we used to have more meetings. We used to write notes with our hands. We used to deal with paperwork more. Office supplies–which feel so obsolete these days–were king.

But what will it be in the future? I think it’ll be a combination of the two. We’ll interact with people more–with computers having automated a lot of the repetitive tasks, we can focus on where we, human beings, add the most value: in being human. But we’ll need to be connected in order to get the leverage we need. We need something that we’ll keep handy that will keep us connected and serve as a tool for seeing, hearing, and taking notes. The iPad is actually a great example of a device that fits in very well with that model. I can absolutely imagine all of us carrying one with us at all times, just like we used to carry our notepads.

School of the Future

Sunday, October 10th, 2010

What will the school of the future look like? What should it look like?

Rilke in one of his letter wrote that “Each person ought to be guided only to the point where he becomes capable of thinking by himself, working by himself, learning by himself” and I think that in the future, the educational system will embrace that philosophy, together with — also from Rilke — “Schools ought to think about all in terms of individuals, not in terms of grades.”

What would it mean to embrace these statements? Well, first, schools need to be personalized. Through teaching various people in wildly different circumstances (teaching math to a seven-year-old; teaching college students; on-boarding new hires) I realized that different people learn differently, and the element of feedback is crucial to continuing progress; in other words, if the specific lesson someone took (as opposed to being given–which implies passivity) from a session isn’t reflected on by the teacher and built upon, the teaching is not going to succeed.

Fortunately, technology can make this possible, giving the teacher leverage he or she could never have dreamt of. However, historically the teaching industry has been going through much slower release cycles and so it will probably take a long time before any change is apparent.

This is one part of the “thinking in terms of individuals, not grades” philosophy — instead of standardizing on the outcome, think about how the teaching is internalized by each student. Another part has to do with the goal of education itself — that goal is to increase the intellectual capacity of an individual, not to produce a society that achieves high grades. The latter is just a construct of the educational system and just as any system that starts overly relying on its metrics, it runs a significant risk of the educators losing sight of the goal (not to mention the reality of any standardized system being game-able by those who have spent a lot of time in it). So in addition to focusing on the individual, we will have to come up with more meaningful success metrics, probably more qualitative ones (since the tuition will be so individualized, we will no longer be able to come up with a single number to describe an entire population).

“Guiding each person up to a point” is just as important. I’ve always thought that the purpose of school is to teach you to think, not to teach you anything specific–the specific may be a side effect, a necessary outcome of a particular educational design (and may be required no matter what design, although we don’t know that), but should definitely not be a goal onto itself. This also means specific knowledge should not be used as a determinant of how well someone has been taught.

In a way, nobody should ever “fail” an education — the point of education should be to determine someone’s potential and enable them to achieve it by themselves. Of course, an individual may choose not to fulfill that potential, but that is not a failing of the educational system (or, at least, not a primary failing of it); it’s probably a failing of the value system instilled by the parents and the society. Today the educational system also plays a role in providing these values; it would be interesting to decouple the two in order to focus better on the thing an individual has a problem with — and possibly use different techniques for either.

My friend E.P. had a particular design for a school of the future: it should teach the concept of a concept, and hopefully at some point the students will understand that this meta-ness is a fundamental block of reasoning and intelligence. I think while it’s an elegant design, it’s impractical — the students need to be bootstrapped first, before they can understand what meta-ness is. Focusing on the concept of a concept for its own sake will probably not lead to a good internalization.

I think back to my education. How much time did it take me to get to the point where I could think for myself? Specifically, when did I internalize that things are related in hierarchies, and that there are different kinds of relationships between the objects in hierarchies, for example an “instance-of” relationship. It took a while, and by the time I understood these concepts viscerally, I could say that the education satisfied an important objective. But the way I got there was certainly complicated and had many diverse and uncorrelated paths — trial and error, learning by example, learning by rote, learning by thinking (surprisingly not much of it!).

We will not get to the school of the future overnight. It probably needs a revolution just like many other industries did. But there is little economic incentive for this to happen — schools are monopolies and profit is not usually correlated with efficiency (which reminds me of the DMV). Teaching by definition takes longer. Unlike e.g. the financial sector, it’s very difficult to come up with good metrics for success. And the barrier for entry is huge (can a startup really revolutionize an educational system?).

There have been instances in the past of people overcoming similar obstacles. So I am hopeful. And while I wait, I may come up with my own syllabus…

Disposable Software

Monday, October 4th, 2010

So far we have been optimizing our software development practices for the increased lifetime of our software — yes, simple features, especially around the beginning of the project will take a long time to develop because we want to structure our code in a way that will reduce tech debt in the future. We tend to keep tech debt at a steady low number — something like 10% — precisely because it’s debt, and we don’t want to pay it off over a long period of time.

What if we make it easier for software to be thrown away, and start optimizing for rapid release? That would be a huge paradigm shift: our users would have to get used to the feature set changing rapidly, even losing feature they may like. There are some cool things we could do, for example write code that literally expires (and a countdown clock for the user so he/she is not surprised when the feature disappears).

Something to think about…

The Change of Seasons

Monday, October 4th, 2010

It’s this time of year. Summer is turning into fall, in a process that’s both beautifully gradual (every day gets colder than the previous; every day gets darker than the previous) and beautifully binary (the summer has features that distinguish it very clearly from the autumn).

This past week this change of seasons seemed to be most pronounced. I realized this was because the day began shortening around the time I finish work / eat dinner (and it began shortening the fastest). This creates a sense of uncanniness, something being not quite the same. Two weeks ago the day didn’t seem to get too short. In two weeks I’ll get used to a short day (even though it will continue decreasing). Around now is when I’m noticing the change the most. I love that change; even though it’s going to get colder I am reminded of the warmth of the home, of warm apple cider, of pumpkin flavors, of social gatherings.