Wednesday, June 24, 2009

Abstractions

"Software is ... onionlike, a thing of layers, each built painstakingly and precariously on the previous one, each counting on the one below not to move or change too much. Software builders like to talk about laying bricks; skeptics see a house of cards. Either way, there's a steady accumulation ogoing on. New layers pile on old.

"Programmers call these accretions 'layers of abstraction,' because each time a new one is added, something complex and specific is being translated into something simpler and more general. The word abstraction comes from the Latin for draw away; here's one computer science definition of the term: 'The process of combining multiple smaller operations into a single unit that can be referred to by name.' Abstraction begins in the nursery. It is what children learn when they realize that if they want an apple, they don't have to find one to point to but can simply say the word that refers to it. They also learn that abstraction depends on a comon language, on shared assumptions of what class of objects the word apple refers to.

" 'This is what programmers do,' wrote Eric Sink, a programmer who led the creation of the Web browser that became Microsoft's Internet Explorer. 'We build piles of abstractions. we design our own abstractions and then pile them up on top of layers we got from somebody else.' And every year the piles grow higher."

--Scott Rosenberg, Dreaming in Code

For example:
Web browser are a wonderful creation. They turn the mess of networks, servers, routers, clients, network protocols, html code, and access permissions that are the reality of the internet - and let users pretend that they are working through a magical window into another world.

This quote does an excellent job of explaining what I mean when I talk about abstractions - a term I apply to linguistics and psychology almost as much as computer science - as well as justifying my current direction in life.

Wednesday, June 3, 2009

Perceptions of History - a funny story

"It might be a good idea if the various countries of the world would occasionally swap history books, just to see what other people are doing with the same set of facts."
- Bill Vaughan
(Found in the Social Studies section of the Globe and Mail)

The next day there was a reply:
"On the topic of swapping history books with other countries (Social Studies, June 1), reader E. A. Carruthers writes: "When I was a child in the States, we learned that the War of 1812 was fought between the Americans and the British, and the Americans won. Later, in Canada, I learned that the War of 1812 was fought between the U.S. and Canada, and Canada won. I asked an English friend about the War of 1812. He looked thoughtful for a moment and asked, 'Weren't we fighting Napoleon then?' "


I really, really, really want some foreign textbooks. On all sorts of topics. (If only I could read them in other languages as well...)

Tuesday, May 26, 2009

Living vs. Doing

I had a chat recently where the diffusion of ideas came up as a topic. Discussion around how best to get something done.
Is it possible to pick the best way of doing things?
Or can we let ideas diffuse and allow for natural selection of what's best?

I've recently found that I'm more comfortable with focusing on living over pushing a common best approach on others.

In a practical sense, baking bread over solving poverty.

If I do my thing and let others do their thing, will things all work out ok?
In any case, I seem to help most as a sounding board. Listening, asking questions, supporting, and critiquing constructively. Or not even being involved, just uprooting trees and eating.

Maybe I'm getting old(er). What I know of developmental psychology suggests that this transition from novel/broad/revolutionary pushing to supporting/suggesting/living is part of normal/biological/human aging.

Or I could just be lazy and unmotivated.
Where's the line between not pushing to make an impact and not having anything to do?
Do we live to try to make a difference?
Do we do to have something to live for?

(I'm still blogging existential angst late at night, so I can't be done with the developmental stage (ie. teenageryness) just yet right? :)

Wikipedia = Cities, an analogy

Disclamer: This is not my idea. This is just a reblog.

On the topic of trusting wikipedia - the great online encyclopedia where anyone can change anything, for better or for worse.

"It is this sidewalk-like transparency and collective responsibility that makes Wikipedia as accurate as it is. The greater the foot traffic, the safer the neighborhood. Thus, oddly enough, the more popular, even controversial, an article is, the more likely it is to be accurate and free of vandalism. It is the obscure articles — the dead-end streets and industrial districts, if you will — where more mayhem can be committed. It takes longer for errors or even malice to be noticed and rooted out. (Fewer readers will be exposed to those errors, too.)

"Like the modern megalopolis, Wikipedia has decentralized growth. Wikipedia adds articles the way Beijing adds neighborhoods — whenever the mood strikes. It is open to all: the sixth-grader typing in material from her homework assignment, the graduate student with a limited grasp of English. No judgments, no entry pass.

...

"[T]here is a professional class of Wikipedia skeptics. They, too, have some seriously depraved behavior to expose: Wikipedia represents a world without experts! A world without commercial news outlets! A world lacking in distinction between the trivial and the profound! A world overrun with facts but lacking in wisdom!

But:
"It’s all reminiscent of the longstanding accusations made against cities: They don’t produce anything! All they do is gossip! They think they are so superior! They wouldn’t last a week if we farmers stopped shipping our food! They don’t know the meaning of real work!"
--from: Wikipedia: Exploring Fact City, NYTimes


And so, says scott, wikipedia is to be trusted as you would trust a city.
And so: Do you trust wikipedia more, or cities less?

Thursday, May 7, 2009

all I want...

All I really want is a home. With people coming and going. Some group habits that allow me to make exist in a regular way, and to bake bread and have a garden. These things take time to build into habits, and I want to do a good job of making them into habits, so it will take time.

I want to know that when I clean the kitchen or tidy up the trees outside, that friends/family who are important to me will benefit from it. Plural people. (Where is the line between family and friends? Could I break it up a bit?)

Pancakes in the morning. I get up early. Early enough to go for a run and then make breakfast, early enough to enjoy myself and to help others wake up. *Poke* "Breakfast is ready!"

Where are the people?
I'm lonely.

Wednesday, April 22, 2009

Human Cognition

The current approach of trying to figure out what goes into making something intelligent doesn't seem to be having much success.

Instead of asking what it takes to make a mind, what if we asked what is missing from the human mind?
Such as:
+Constant attention
+Perfect memory
Other things that computers have that we don't?

If we want to replicate humans, we need to copy their flaws as well as their strengths.

Wednesday, March 4, 2009

Re: expectations in my CPSC course

Context: The lecture style is a bit different from usual, and people have generally found the course harder than they expected (it being a first year CPSC course). The midterm was very hard for most people and many people got lower grades than expected. The instructors have admitted that it was a bit harder than they aimed for, and have offered some makeup questions on the next assignment.

"I think more questions is better than fewer questions as it assesses more parts of the course topics, and not biasing those who are more stronger in one area vs another."
--from the class discussion board

hmm, this is an interesting point. Could we as students just be very used to finishing all questions on an exam? Is this a reasonable goal to have?

How much of the course material should the exam aim to cover? If the coverage is limited to what can be completed in the given time, then there will be some parts of the course that aren't tested. This means that students who are better at the content not covered will be at a disadvantage.

Perhaps there is an expectation inflation in the same way as grade inflation?
I personally feel I learn more from classes that cover some content that is beyond my current ability (only expecting me to just pass the course, and not (necessarily) to know everything). If I don't get it right away, it will often make sense later in the class or even after the class is finished.

Friday, January 23, 2009

Group estimates

The average of many peoples' estimates is better than an expert's estimate.

Interestingly, the many estimates by one individual is almost as good.
(See: The Economist, from a long time ago)

So I'm happy not writing down every good idea that I come accross or dream up, because I feel that if it's really good, it'll come up again, and in other places and contexts.

So people collaborating on {books, projects, etc}
should produce better work than individuals.

People all together sometimes do things that none of them aim to, or can.

Again, collaboration leads to new places.
Now we have books written by large, un-regulated, amorphous groups. Will this be better than the alternative?
Books | Global Text Project

I belive so.

Friday, January 2, 2009

family tree

My grandfather:
http://www.thecanadianencyclopedia.com/index.cfm?PgNm=TCE&Params=A1ARTA0005220
Aeronautics and cognition, neuropsychology.

My grandfather's father:
http://en.wikipedia.org/wiki/Melvill_Jones
The importance of fluid resistance (aerodynamic drag) for aircraft.

My grandfather's godfather:
http://en.wikipedia.org/wiki/Geoffrey_Ingram_Taylor
Quanta, Fluid Dynamics, and Wave Theory.

My grandfather's godfather's grandfather:
http://en.wikipedia.org/wiki/George_Boole
y'know... that logic dude...

My youngest brother:
Want's to fly planes.

Me:
Cognition and machine-human interaction (??)

And fun times were had by all.

Thursday, January 1, 2009

Things that were learnt in conversation:

Separation of work and work-environment. There are many aspects to any activity/learning/work, you could enjoy the content of what you are doing but hate the people you are working with. In the vast majority of cases it would be possible to find a similar situation that changes just that part that you don't like. It is possible to learn from activities you do and don't enjoy in an attempt to build up a better understanding of what works well for you in various areas of life.

That some people (related to me) prefer to have some starting point or an existing project to build on in self-directed work of their own. For example, the uncle-that-lectures prefers to build on an existing curriculum (no matter how terrible it is), than to come up with his own. I remember thinking something similar before about myself, but I forgot it. I do feel that there is something similar in me.

The best ideas are those that come up on a regular basis, so don't worry about writing brainwaves down immediately, they'll come back if they're actually any good.