Monday, January 12, 2009

The First Rule of Smell Club is . . .



I’m a fan of smell maps.  In What the Nose Knows, I described Gawker.com’s reader-generated interactive smell map of the New York City subway system. When you moused over a specific malodor icon at a given station, a pop-up window displayed the verbatim entries (“old outhouse poop” or “stinks like puke” at 34th Street and Eighth Avenue, for example).

Today, various media sites have picked up an AP story about a Smell Club in Japan.  The club has a website--Nioibu.com--where its two hundred or so members can register a smell and place it on a world map. Roll over the locator pin and the popup balloon displays the member’s entry.

In the screen shot above, I rolled over a yellow locator placed on the West Coast by a someone named “Yaya” and got this
By yaya
作りたてのホットサンド
発生源:キッチン
which Google-translates to
By yaya
Fresh hot sandwiches
Source: Kitchen
Zoom in and it appears that Yaya was sniffing about on the corner of Folsom and 3rd, across from the Moscone Center in San Francisco.

Pretty cool.

Any Japanese readers?  I’d like to know the meaning of the different colors of locator pins.

UPDATE January 14, 2009
Chicago Tribune reporter Monica Eng catches up with the story and asks
What does Chicago smell like?” Her commenters mention the Blommer’s Chocolate Co. factory, which gets some pretty good aromatic reviews on Yelp.

6 comments:

Anonymous said...

I studied Japanese, and just remember a little. The literal description from yaya (besides the kanji for "fresh" which I can't read) says "hotto sando"--sando being short for sandowichi. Japanese borrows a lot from English.

Other than that, I'm afraid I don't understand: wakaremasen--shitsurei shimashita.

-Lucinda

Anonymous said...

Avery,

In the slight chance that you don't have an answer to your query yet.

I looks like the pins are an intensity scale. The comments transliterate as:

red - Extra bold
blue - bold
green - medium
yellow - light

When you click through to the details it also looks like there is another scale used to describe the qualities on the scent. The pentagonal form on the far right contains subjective metrics of relative "taste" and not odors. I'll leave it to the expert to speculate on the reasons cultural or otherwise for this formulation.

From the top vertex proceedind clockwise the descriptions are sour, bitter, sweet, salty and ???.

The fifth uses the base kanji for pain/hurt. The form of the verb is "injured" or "bruised". The kanji also appears in the word for pungent although not in the form shown on the page. Can we just say pungent?

Sorry, my good usage dictionary is hor's de todler in a box somewhere in the chaos.

GW

Avery Gilbert said...

George:

Thanks for the nice sleuthing. So pin color codes for odor intensity and the vertices of the pentagon code for aroma quality using taste descriptors. The mystery fifth element could be either umami (monosodium glutamate taste) or irritation (chili pepper burn). Wonder if these codes are only applied to food smells?

What's cool about Smell Club is it forces you to think about how to construct a smell classification that covers all odors of the world and yet is intuitive, user-friendly, and cross-cultural. Not so easy.

ECaruthers said...

There's got to be good money in selling the software that would let each of us set up our city's smell map. I'm afraid most of the people who set up sites would want to document the sources of smells they don't like. But if the codes include good as well as bad smells, everyone can play.

Avery, could the categories from the 1986 NG Smell Survey be used or are they subject to copyright? They seem pretty general, but I never see detailed discussion of that work.

Avery Gilbert said...

ECaruthers:

The National Geographic Smell Survey descriptions (which I co-authored) are there for the taking. If someone decides to adopt them lock, stock and barrel I'd hope they'd cite the original work.

The trick to assembling a useful set of odor descriptors is to cover the entire operating space (bad to good, indoors to outdoors, whatever), avoid redundancy, and keep them all at the same level of generality (pine trees vs Douglas fir).

As you know, I like odor maps. So I'm curious--who do you think the audience is? What would motivate people to add their two cents?

ECaruthers said...

Avery,
Thanks for the info on the NG survey. I knew you were the co-author and thought that made you the right person to ask.

I can see obvious sources of ad revenue for the people who set up and maintain odor maps. I'm not so sure about motivating people to enter and update the data. If I owned a pizza shop I might be motivated to enter, "strength=7, odor=spicy, value=+5." If I lived by a tannery, I might enter, "strength=2, odor=foul, value=-5."

Probably a small minority of interested people would initially provide most of the input. The key is giving them some feedback, so they don't lose interest. If there is enough data every day, then the map looks different from day to day and people who are just casually interested have a reason to check in. E.g., in the summer, you might want to check the smell readings for public beaches before planning a picnic.

So, now that I've talked myself into buying a server, how do I get the software?