04.30.08

Woes of a freelancer…

Posted in General at 4:26 pm by Michael

Understands the math, understands the programming, understands the client. Pick any two.

Idea: Collapsing directions in Google Maps

Posted in Ideas at 10:28 am by Michael

Google Maps gives turn-by-turn directions to get from one point to another. This is very handy, but most people do not require turn-by-turn directions on all legs of their trip. I’ve often found myself thinking things such as “Just get on Route 1, take exit 5, then follow the directions”. Google should make this sort of thinking representable by allowing users to collapse parts of the directions that they do not require. It would save space and make the directions more intuitive.

But given my past experiences giving ideas to Google, I no longer do so directly.

04.29.08

Classification of population density by satellite images

Posted in Ideas, Research at 4:59 pm by Michael

People have sort of brushed on this idea before at http://www2.cr.chiba-u.jp/symp2005/documents/Postersession/p002_Ketutwikantika_paper.pdf, but it doesn’t look like there’s any significant progress towards actually having done it yet. I bet you can use vector quantization to do this – feed in some urban, suburban, and rural areas, extract texture descriptors using VQ, and feed them into a classifier.

This is one I’m going to pursue. The barrier to entry is very low and fits nicely into my area of expertise.

Of all the evils in the world…

Posted in Ideas, Philosophy at 12:12 pm by Michael

Of all the evils in the world, the greatest is the resistance to any action which would help correct the others.

04.28.08

Finally, a CBIR implementation in a major search engine.

Posted in General at 10:24 pm by Michael

Google is finally about to roll out a CBIR system using something similar to the “Vantage Objects” concept. It’s about time – I’ve wondered why no major search engines adopted this sort of thing for a long time.

ASD really does make for difficult communication.

Posted in General at 5:08 pm by Michael

I have nothing against people with Asperger’s syndrome or other autism-spectrum disorders – in fact, they tend to have some great ideas and are among the smartest people I’ve ever met (true savants are outright intimidating, even to me). However, the communication difficulties aren’t overhyped; they really are very difficult to talk to, even when I make a conscious effort to communicate in spite of the difficulty. It seems that they either focus incredibly intently on a single topic to the extent that everything else, including the person they are speaking to, is shut out, or they exhibit the complete opposite behavior, jumping from topic to topic so quickly that it’s difficult to retain coherence amidst the clutter (but is this actually Autism?). It seems to have an impact not only on their real-time verbal skills (i.e., those used in a conversation), but also on their writing skills in general.

I don’t think I quite understood the nature of the disorder until I began communicating with some individuals who possessed it.

The progress of science could be visualized on a circle.

Posted in General at 10:05 am by Michael

And it would make an awesome poster :)

The vast area in the center of the circle would be solid, representing “common knowledge”. The area outside would be filled with variable-length linear spokes projecting outwards from the surface of the circle, representing current work in various scientific fields. A breadth-first approach to knowledge would then be represented by traversing the circle, while a depth-first approach would be represented by moving in the direction orthogonal to the circle’s tangent line.

04.26.08

Short-term conditions influence long term decisions

Posted in Ideas, Psychology at 7:42 am by Michael

I call it the Grocer’s Hypothesis, because it’s probably something supermarkets have intuitively known. Customers who are hungry when they go shopping probably tend to buy more food than their satiated counterparts. This can be generalized; short-term conditions influence long-term decisions.

(Update: This has now been validated by a study at the University of Pennsylvania. Perhaps it should now be called the Grocer’s Theory :) ).

The Status Quo is a Local Minimum

Posted in Ideas at 7:41 am by Michael

If we treat society as a neural net, the current status quo represents convergence on a local minimum. That’s why things work, but not necessarily well.

04.25.08

Around what does society swing?

Posted in Literature, Sociology at 12:01 pm by Michael

This isn’t all that great a poem, but the idea it expresses is something I’ve been struggling with for a while.

Around what does society swing?
What is the pivot, the crux?
What great idea, what a thing,
sparks and ignites such a flux?

Arbitrary though it may seem,
underneath it does have a cause,
a root buried deep in some meme,
a weight against all of the flaws.

Chaotic it is, and profound,
enigmatic puzzles abound,
it won’t ever seek to perfect,
so all it can do is direct.

« Previous entries Next Page » Next Page »