Posts Tagged ‘tech’

April 13, 2011

Skymall is the worst thing in our culture. Not in an insidious way — just in an absolutely depraved way.

Becky Russoniello


Tumblr Likes

April 11, 2011

Look at just the first digit and the number of digits.

  • science: 32914, 11566, 4989, 3743, 968, 814, 673, 482, 286, 2811
  • black and white: 1694, 1167, 1108, 988, 919, 639, 596, 591, 580, 544
  • lol: 22627, 18100, 17688, 14374, 13459, 12045, 4711, 3779, 3670, 3393
  • fashion: 955, 581, 486, 435, 402, 303, 279, 279, 278, 275
  • architecture: 1426, 461, 433, 251, 230, 219, 194, 194, 175, 167
  • art: 7492, 2965, 2761, 1316, 544, 435, 413, 331, 307, 296

Snapshots taken between 9:30-10:30 on 5 April 2011.

To the victors, go the spoils. (popularity?)

R code:

> science <- c( 32914, 11566, 4989, 3743, 968, 814, 673, 482, 286, 281 )
> bw <- c( 1694, 1167, 1108, 988, 919, 639, 596, 591, 580, 544 )
> lol <- c( 22627, 18100, 17688, 14374, 13459, 12045, 4711, 3779, 3670, 3393 )
> fashion <- c( 955, 581, 486, 435, 402, 303, 279, 279, 278, 275 )
> architecture <- c( 1426, 461, 433, 251, 230, 219, 194, 194, 175, 167 )
> art <- c( 7492, 2965, 2761, 1316, 544, 435, 413, 331, 307, 296 )
> require(RColorBrewer) > accent = brewer.pal(8, "Accent")
> leg.txt <- c("science", "black & white", "lol", "fashion", "architecture", "art") > leg.col <- c(accent[1], accent[2], accent[3], accent[4], accent[5], accent[6])
> par(bg="#fafaff")
> plot(science, type="s", log="y", lwd=2, col=accent[1], xlab="x-th most popular blog post", ylab="# likes", main="distribution of LIKES on tumblr", cex.axis=.8, col.main="#444444", col.axis="#333333", fg="#332211")
> points(bw, type="s", lwd=2, col=accent[2])
> points(lol, type="s", lwd=2, col=accent[3])
> points(fashion, type="s", lwd=3, col=accent[4])
> points(architecture, type="s", lwd=2, col=accent[5])
> points(art, type="s", lwd=2, col=accent[6])> legend("topright", leg.txt, fill=leg.col, title="TAG", text.col="#393939", title.col="#222222", border="#f0ffff", box.col="#666666")

Question: is there lag-1 autocorrelation in the likes per tag over time? If I were scrolling down from the top, I’d be more likely to quit (or skip to the bottom, not AR[1] or AR[3]) if the last few pictures were boring.

I wonder how this distribution of Likes compares under the Explore vs Directory régimes. I would guess Likes are more focused under the new régime.

February 26, 2011

Linear combinations of eigenfaces — images like the above — are the cheapest way to store and search photos of faces. Like if you want to computer analyse the faces of everyone at the Superbowl and see if there’s a terrorist.

All of the terrorists’ faces are saved in a database as like {11% Eigenface_1, 6% Eigenface_2, 1% Eigenface_3, …}. So the real face breaks down to just a list of percentages {11%, 6%, 1%, …}.

Articles on Eigenfaces: 1, 2, 3, 4, 5, 6  (sorry to reinforce the google hierarchy)

If you break the faces in the live Superbowl crowd CCTV’s into such a list of percentages, then you can have a computer do a search against known terrorists. (Imagine having the computer search pixel by pixel. It would be like, Hey, the top left of your head looks like the top left of a terrorist’s head! Whoops, that was just the soda machine.)

Final point. There is more than one way to mathematicise a face. Surface normal vectors are one way; the manifold limit of polygons is another; projection is another; and eigenfaces are another. Each way has you conceive of a face differently.

It’s only been about 150 years that humans have had a unified vision of physics and chemistry. It’s only been about 100 years that humans have had a unified vision of biology and chemistry. It’s only been about 50 years that humans have been able to travel by airplane. Pretty cool, living in the modern age. Hooray, science.

February 22, 2011

Were Atoms Real?

Dead ex-Factory Towns

February 17, 2011

What we have here is a failure of the imagination.

I know of more than one town in the USA with at least 500 properties in foreclosure. Whole neighbourhoods could be bought for O(10⁴) dollars. But what could you do with such places that would make money?

Laws prohibit some kinds of things.

  • You couldn’t charge people admission to smash and burn the things in the houses.
  • In fact every house you buy would saddle you with responsibilities that would require at least a few administrative people to watch over and help you comply with.
  • You couldn’t grow cash crops like pot inside the houses.
  • It would be too expensive to demolish the houses to justify the meager returns on outdoor crops.
  • You couldn’t sell or give the houses to Haitians, Hondurans, Nicaraguans, or other people who would be super happy to just have a US quality house over their heads.
  • If you gave the houses to homeless charities, the charity would be saddled with upkeep and accounting/legal costs.

It would also be difficult to get enough momentum behind an artists’ colony. B&B’s or retreats would take way too much money & effort for the probable payoff (who is going to do a retreat in Detroit?).

But there are some real resources that are under-utilized in these places.

  1. Already constructed houses.
  2. Hooked up with electricity.
  3. Power lines and roads around them.
  4. Land — cheap land.

There’s got to be something that someone could do with these dead spaces.

Just need some good ideas, willpower, inspiration, vision.

New York

February 15, 2011

Think about this. “Wall Street” refers to investment banking, even when it’s done outside New York. “Broadway” refers to musicals, even when they’re done outside New York.

New York City lives in our minds.


January 17, 2011

Artists, engineers, and entrepreneurs all build things.

But art majors, engineering majors, and business majors seem to hate each other.

That sucks.

High-Frequency Trading

December 15, 2010

The point of high-frequency trading is to smooth markets over time.

Imagine that you know a big block order is coming in. Some huge pension fund needs to send 70,000 checks out next week so they’re selling some asset.

The asset’s price doesn’t really deserve to go down. The fund just wants to cash out. So you buy the huge block order and dole it out in smaller pieces as regular buy orders ebb in over the next few weeks. Or, you know, minutes.

You just smoothed the asset’s price, as well you should.

That’s not ultra-high-frequency trading — which is more about having the fastest technology — making it possible to liquidate a position RIGHT, RIGHT, RIGHT now.

Read Before You UpVote

December 10, 2010

This is a suggestion for Reddit, Hacker News (ycombinator’s), and other Digg clones.

  1. Only count up-votes from people who have clicked through to the article.

The flourishing of voting communities is a blessing to the Internet. Members post links and everyone gets to vote up or down. It’s an improvement on Google’s PageRank, which has been maliciously exploited by spammers for a decade now.

In my brief time watching these communities, I’ve seen snappy titles with terrible content get up-voted. Things like:

  • 7 Ways to Amp Up Your Productivity
  • FOX News Has Done It Again…
  • Resources for Busy Professionals
  • Books You Should Read But Aren’t Really Going To But You’ll Upvote This Anyway Because You Feel Guilty For Not Reading Them

I’ve even sinned myself — up-voting or down-voting before I read the content. Sometimes I feel bad once I realize I’ve done it and go skim the article. But that’s not the point. Somebody should have stopped me before I voted on something I knew nothing about.

I diluted the votes of the informed people and that’s bad for the site.

PS I don’t think many people would game the system, do you?

Sci Fi Now

December 2, 2010

People over-estimate what can be done in the short term, but under-estimate what can be done in the long term.

Weight loss is a perfect example. People think they can get in shape in a couple weeks before going to the beach (hard to do). But they don’t realize how much weight they could lose over a year by eating just 100 calories under their daily metabolic usage (answer: 10 pounds per year, i.e. 100 pounds per decade! from 1 tweak).

Futurists also over-estimate the magnitude of short term changes and under-estimate the long term magnitude. I remember there were so many prognosticators in the 90’s, during the Internet boom — it was all New Economy this and paradigm shift that.

So now that everybody is (ahem, Americans are) logging into Facebook on their iPads — who was right and how? I mean, who are some futurists who said we would be in roughly the place we are now — and what did they think of “us”, the “future people”?