Tag Archives: language

Finnish Idioms

While living in Finland, I learned of some of the Finnish idioms that give a wonderful glimpse into the Finnish mind & culture. Here are some of my favorites.

juosten kustu translates literally as, “pissing while running” which, if you’re a man and have ever had to do this, results in a wavy line in the snow. It means something done half-assed. Related but different is kusee hunajaa which means that you’re happy because you are “pissing honey.”

laittaa hanskat naulaan translates as, “hang up your mittens” which means you give up or, more specifically, are quitting your job. Because we all know that if you’re not wearing your mittens in sub-zero weather, you’re not going to be working.

irtoaa kuin mummon hammas or “comes loose like a grandmother’s tooth” is something that goes very easily. This is the opposite of kiven alla which is difficult to obtain because it is, “under a stone.”

joka kumartaa yhdelle, pyllistää toisille “when bowing to one person you show your ass to those behind you” could be cynically translated as standing on other peoples heads to get to the top.

juopon napit when you mis-match the buttons on your shirt, you have “drunkard’s buttons”

lukea kuin piru Raamattua translates as, “like the devil reads the Bible” or someone who reads something carefully to look for a loophole.

Any ones that you like?

Digital Cartography

Eric Fischer takes large datasets and turns them into art. His flickr stream is a collection of fascinating time-series maps plotting data over time to draw out shapes which take on a greater meaning. Weather it’s a map of taxis in San Francisco or an overlay of flickr metadata on top of NYC, Eric’s creations are at once beautiful and informative.

Last month Eric was able to use an open-sourced version of Chrome’s language detector library to parse a week’s worth of geo-tagged tweets and identify who what tweeting in what language, where. What you see above is is the result. Note how languages such as Portuguese, Spanish, English, Dutch, Italian, Swedish, and German stick to and define the borders of their nations. Within each country, major transportation hubs are lit up like avenues. One can only imagine it is the result of people tweeting while enroute somewhere on a train or bus.

IBM Watson on Jeopardy

In a brilliant piece of PR, IBM Research stormed back on the scene matching their artificial intelligence computer, Watson, against top contestants of the popular American game show, Jeopardy. On February 14, 15, and 16 Watson’s competes against two humans on live television.

According a piece on Wired’s Epicenter blog, 25 IBM scientists spent four years building Watson. “Powered by 90 IBM Power 750 servers, Watson uses 15 terabytes of RAM, 2,880 processor cores and can operate at 80 teraflops, or 80 trillion operations per second”  The database of content upon which Watson draws its “knowledge” is from over 200 million pages from reference texts, movie scripts, entire encyclopedias, as well as, Wikipedia.

How did Watson do?

After the first day, Watson is tied with Ken Jennings for the lead at $5,000, beating out  Brad Rutter who has $2,000. For details on the game and some behind the scenes of what was going on, listen to the interview with Stephen Baker on Arik Hesseldahl’s post on All Things Digital.

Footage from a practice round back in January below.

More details at the IBM Watson mini-site and a documentary on Nova, Smartest Machine on Earth.

Singularity and Sentience

I tuned into the launch of Echo’s social media mixer, the StreamServer, which they describe as a platform for the activity streams-based economy. As the saying goes, in a world where the amount of information is ever-expanding and time remains constant, attention is what is of value. As your phone and computer beeps and buzzes with the latest urgent notification, the ability to monitor, much less take action on a signal becomes impossible. All information is approaching real-time in the constant battle to be first.

The volume of this “me first” wave of data increases causing the half-life of information to get shorter. Steve Rubel quoted a study that found that 92% of retweets happen in the first hour. If you can’t get your point across so that it resonates with someone else within that first hour, that thought is gone for good. Scrolled away, below the fold, decayed away.

So we wire things up to make things faster and we put systems in place to help us make sense of all this information flowing around so we can pick up a signal that we can use in a meaningful way. Something that will hopefully make our life better than it was before we had to deal with all this information that gets pushed at us.

Then we build filters. And what are filters but a search query that swims mid-stream. Not a respective search like what you would type into Google to search an archive but a prospective search, one that looks forward in time. And each of these queries we type are nuggets of intelligence. We fine-tune them to get exactly what we want and filter out what we don’t.

Follow all my LinkedIn contacts (that have twitter accounts) that are in the Mobile Phone industry and have over 500 followers that are saying anything with a hashtag of #MWC and has more than 5 retweets.

In plain English (kinda), that is what we want our filter to do and a smart system will look at how we respond to the results of that filter and try and automatically make it better. More like this, less like that, etc.

So we teach the machine how to think. We tell it how we connect the dots and draw conclusions.

So I dig around the aboutecho.com wiki and scroll down to the Philosophies section (I dunno, sounds intriguing) and click through to read this post on http://synapticweb.org/.

Social profiles are becoming real-time streams. If the old profile was a neuron, the stream is a neural pathway or pattern. It is the connective tissue between applications and people that feeds information from one node to another. Profiles come and go, people express themselves using countless tools and technologies – the stream, however, is the consistent and persistent channel that matters. It is the new presentation metaphor that increases the level of information we can consume while reducing our sense of overload. Just like synapses, they fire, and like synapses, it is the collective patterns of multiple firings – multiple signals or re-tweets – that creates a pattern. Patterns create meaning. Tune in, tune out, it doesn’t matter. The information will find you if it matters. Implicit information derived from content and gestures is one of the great opportunities of the Synaptic Web. To observe a set of gestures and connect them together creates a dynamic profile of interests, intentions and friends that can be used for discovery and filtering.

This is heady stuff. Yeah, I read Kevin Kelly’s book too but we’re going to have to evolve quite a bit beyond brute force keyword filters. How do you encode a vibe, a hunch?

Don’t get me wrong, Echo StreamServer looks like an interesting idea and I’m sure we’ll hear something along these lines from Facebook soon too. Big minds are at work on this. But let us not fool ourselves into thinking that a bit of hacking is going to solve our information overload problems. We’re just taking the tools out of the toolbox and learning what we have.

Soylent, the macro made of people

More on this question of crowdsourced writing, (plugging wet-ware into software) comes Soylent, (yes, it’s made of people) a research project which has written a add-in to Microsoft Word.

Hooked into Amazon Mechanical Turk, this project offers three services:

  • Shortn – trim document length without changing meaning
  • Crowdproof – distributed human proof-reading
  • The Human Macro – open-ended tasks such as changing the tense of document, “natural language crowd-scripting”

Think of how something like Soylent and other outsourcing services change the game. I used to work with someone that would farm out the preparation of his expense reports to Man Friday. How soon before someone, in a moment of bureaucratic weakness, whips up a script to to outsource preparation of the weekly TPS report to management from inside Corporation X. Chances are, it’s already happening.

Textbroker.com – low cost, bespoke content

So the results are in from our little experiment and the view is . . . you get what you paid for.

Just to review. Textbroker.com is a service which will write for you on the topic of your choosing. They have a four levels of service and pay by the word. For our experiment, we tried two levels on the upper end of the scale. The topic was “The History of Cream Cheese,” as a control, I added a third sample for the poll in which I copied an article from wikipedia and used some software to shorten the text using an algorithm.

For those of you who read through the samples on the earlier post, here’s what we paid.

  • Sample ONE, copy/paste from wikipedia, shortened using software algorithm
  • Sample TWO, 2.2 cents/word, 24 hour turnaround
  • Sample THREE, 6.7 cents/word, 5-day turnaround

The overwhelming choice (over 80%) was for the most expensive#3  sample. It’s pretty clear that the most thought was put into this text and at a total cost of almost $20, it was by far the most expensive sample to commission.

Textbroker is a pay-as-you-go version of companies such as Demand Media which are content factories that focus on, “optimizing high-quality content” for domain squatters and publishers looking for fill to generate pageviews for their advertising partners. What was interesting about to my Finnish colleagues is the thought they could use the service for preparing rough drafts. English skills are very good here but the hardest part for many is just getting started. Many that I talked to thought Textbroker would be a great way to jump start a first draft to get beyond the blank, white page.

My father, a former editor at Random House, is weary of this trend toward mass produced content. “This way lies madness,” was his one line reaction. The business model exists, of course, at the other end of the spectrum. For those that cannot afford their own ghost writer, there is a fellow going by the name of Charles Kinbote who is offering Bespoke Art Commentary specializing in critical analysis of your child’s artwork. For 190 pounds you get a beautifully framed original.

Here’s how he pitches himself.

And you are also busy, no? You may be keen to know what modernist artists your two-year-old son is referencing in his playgroup art, or perhaps what Renaissance works your three-year-old daughter mimics in her scribbles, but you nevertheless don’t have the time, or, let’s be honest, the talent to critique your children’s artistic endeavors the way a real critic would. Explaining any work of art is not easy. Explaining why a young, immature artist — a child, if you will — chooses to be influenced by Renoir rather than Richter is an almost impossibly complex maneuver, one for which you need an expert. I am that expert. The exploration of the meaning contained in the artwork of children is my life’s pursuit. I offer you my service. My name is Charles Kinbote…

– from Kinbote’s Bespoke Art Commentary Service

Thanks everyone. It was fun but I think I’ll stick to my own writing. I enjoy it too much.

UPDATE: Just for giggles, I copy/pasted the text from the third sample into Xtranormal to get the narration.

Outsourced Thinking

Last week a colleague of mine tweeted about a new service that offers “exclusive written articles created to your specifications”

I replied that Udo might be stretching the definition of “professional” if we look at the quality of output. Cranking out “content” on demand for pennies a word, one has to wonder about the thought that goes into such work. Cheered on by the folks at @textbrokerUS we decided to test the results.

My kids came up with a topic general enough for anyone to judge and we figured it’d be fun to see what others would come up with too. The pricelist on textbroker.com shows a range of prices depending on the quality required. Three paragraphs follow:

  • 2.2 cents/word – 24 hour turnaround
  • 6.7 cents/word – 5-day turnaround
  • Copy/Paste from wikipedia, use Microsoft Word AutoSummarize feature to shorten

Below are the results, use the poll below to pick what you think is the best quality.  I’ll collect the results for a bit then share the source of each entry in a future post.

The topic was 300 words on the History of Cream Cheese.


ONE

Cream cheese (also called soft cheese) is a sweet, soft, mild-tasting, white cheese, defined by the US Department of Agriculture as containing at least 33% milkfat (as marketed) with a moisture content of not more than 55%, and a pH range of 4.4 to 4.9.

There are French references to cream cheese as early as 1651. According to the American food processing company Kraft Foods, the first American cream cheese was made in New York in 1872 by American dairyman William Lawrence. It can be a base to some spreads, such as yogurt-cream cheese topping for graham crackers, (10 oz cream cheese, and 1 cup yogurt, whipped.). It is sometimes used in place of butter (or alongside butter in a ratio of two parts cream cheese to one part butter) when making cakes or cookies, and it is also used to make cream cheese icing, which is similar to buttercream icing, (using a ratio of two parts cream cheese to one part butter) which is used to ice carrot cake. Furthermore, because cream cheese has a higher fat content than other cheeses, and fat repels water, which tends to separate from the cheese, stabilizers such as guar and carob gums are added to prolong its shelf life.

There are French references to cream cheese as early as 1651. It can be a base to some spreads, such as yogurt-cream cheese topping for graham crackers, (10 oz cream cheese, and 1 cup yogurt, whipped.). Cream cheese is difficult to manufacture.

TWO

Cream cheese, the most popular ingredient for cheesecake, is a smooth, white cheese made from heavy cream and milk.

Specifically the soft, unripened cheese is made from cow’s milk and by law must contain at least 33 percent milk fat and not more than 55 percent moisture. Light or lowfat cream cheese has about half the calories of the regular style.

Cream cheese is frequently used for spreading on bagels, raw vegetables and crackers and as an ingredient in appetizers, snacks and dips.

There are many different types of cream cheese such as Mascarpone, which comes from Italy and Quark which is a cheese from Germany that has a sharp flavor. In parts of Europe, cream cheese is called “white cheese.”

While there are French references to the origin of cream cheese around the 1650’s, the first American cream cheese was made in 1872 in Chester, New York by American dairyman William Lawrence, according to Kraft Foods. Lawrence distributed his cheese under the brand Philadelphia, now a trademark. The Kraft Cheese Company bought Philadelphia cream cheese in 1928 and still owns it today.

To make your own cream cheese, combine 2-3 cups of whole milk and 3 cups of heavy cream in a stainless pot and stir regularly. Mix 2 tablespoons of buttermilk thoroughly into the warmed milk-cream mixture and cover. Then stir in a quarter teaspoon of mesophilic starter culture, which preserves the cream cheese.

Add a quarter teaspoon of calcium chloride liquid and 2 tablespoons liquid rennet to the pot. Cover the pot and allow it to sit overnight at room temperature. The mixture will have gelled by the next morning, at which time line a large strainer with a sterile handkerchief and gently pour the product into the cloth and let drain for roughly 30 minutes. Transfer the cream cheese into a separate container and mix until smooth and creamy and then store in a refrigerator.

And while cheese can be traced back about four thousand years, the first recipe for cheesecake wasn’t recorded until 230 AD.

THREE

References to cream cheese can be found in France dating to the mid-1600s, though surely it existed long before that. The Greeks contend that cheesecake, made with a soft, creamy cheese, was served to athletes participating in the first Olympic games over 2000 years ago. From Greece, this delicacy spread to Rome and throughout Europe in the following centuries.

Neufchatel, the French soft, white cheese, was the inspiration for an American dairyman who developed what is recognized as cream cheese, the familiar, foil packaged uncured cheese. In an attempt to make a version of the French soft cheese, William Lawrence of Chester, New York, stumbled upon a unique creation in 1872, higher in fat due to the addition of cream to the recipe. The true French Neufchatel cheese is made only with whole milk; its fat content is a little over 20% while cream cheese’s is over 30%. Although controversy surrounds the invention of American cream cheese, with some saying a neighbor of Lawrence’s independently developed the same cheese, it was Lawrence’s version that evolved into today’s Philadelphia Brand Cream Cheese.

Lawrence began commercially distributing his new cheese in 1880. Wrapped in foil and named after the city renowned for quality products, Philadelphia, the cheese was manufactured by Lawrence’s own cheese factory in Chester and by C.D. Reynolds’ Empire Cheese Company in South Edmeston, New York. Philadelphia Brand Cream Cheese’s ownership passed to the Phenix Cheese Company of New York in 1903 after a disastrous fire reduced the Empire Cheese Company to ashes.

The J.L. Kraft Company merged with the Phenix Cheese Company in 1928, obtaining the rights to Philadelphia Brand Cream Cheese, still manufacturing it today. There are hundreds of other brands, but none as famous as the one originally invented by Lawrence.

UPDATE: the results are in! I’ve posted an overview.

Hecka nice

“He’s an extraordinary player who makes every team a lot better. This guy is hecka nice. . .”

High school coach for Jimmy Rollins who has just been named Baseball’s National League’s Most Valuable Player. Jimmy Rollins is from Alameda, CA where I live. Encinal High School, where Jimmy played, is in Alameda. We’re all really proud and every time the Phillys come to town to play the Oakland A’s at the coliseum, the stands are packed mostly to see Jimmy play.

What really cracked me up about this article in the local paper where I saw this quote is the phrase “hecka” which is a uniquely Bay Area term. My kids have started to use it now so I guess we’re settling in.

More Julia-isms

When Julia has a stomachache, she says she has a “stomach egg.” It took us a while to pick up on this and it wasn’t until she complained to Izumi about an “egg” on her knee that we finally figured it out. Applied to other areas of the body it kind of makes sense. I think we all sometimes wake up in the morning with a “back egg” or a “neck egg” but we eventually shake it off and it’s gone.