Wolfram|Alpha is by no means a mainstream search engine, yet it has managed to establish something of a beachhead and has a growing fan base. It owes some of that to Apple’s Siri, which chose Wolfram|Alpha as one of two structured data providers that tie directly into the personal assistant. The other is Yelp.

Wolfram|Alpha gets about 3.7 million daily page views and 17.3 visits per month in the US, according to Alexa data the site itself provides when you ask it about its own traffic.

According to an article in the New York Times tied to the Wolfram|Alpha Pro launch, “Siri accounts for about a quarter of the queries fielded by Wolfram Alpha.” However the company receives no branding on those results. Here are some examples of how Siri presents Wolfram|Alpha content:

Apple and Siri’s relationship with Yelp and Wolfram|Alpha are a kind of template for how the personal assistant will likely work with other third party data providers in the future.

Since Wolfram|Alpha launched in 2009, I’ve often wondered if the company would allow users to use their extremely powerful computing infrastructure to analyze their own data along with data sets available on the open web.

Well, they have and that’s what Wolfram|Alpha Pro is all about. As I wrote a few week’s ago, I’m a frequent W|A user and long-time admirer of the company and it’s founder, Stephen Wolfram. I think they’ve scored again.

The service has an introductory rate of $4.99/month with a reduced rate of $2.99/month for students. The company offers pricing for the enterprise. So, what does this initial launch offer?

Wolfram|Alpha Pro: Key Features

• Ability to Analyze Your Own Datasets Input of upload a data set and let Wolfram’s technology automatically recognize and analyze the data. For example, a timeline of your email, your credit card expenses, or a data set obtained from Data.gov.

• Data in more than 60 file formats can be uploaded and analyzed. From an XLS spreadsheet to an HTML document to the audio in a WAV file. Image files (in many formats) can also be analyzed. As you would expect, many science and math formats are also included.

• Access to extended keyboards like the one available with the Wolfram|Alpha mobile apps.

• Visualized and interactive results using the CDF (computational data format) that Wolfram launched last July.

What’s not available today is the ability to publish and compare, comment on or share results. According to Stephen Wolfram, these options are forthcoming.

I’ll can say now that while some of what Wolfram|Alpha Pro has to offer might not be for everybody today it very likely will be in the future as the amount of data available to analyze (personal data and publicly available datasets) continues to skyrocket.

In other words, Wolfram|Alpha pro is getting in on the ground floor of personal data analysis with a powerful but at the same time very easy to use tool that will appeal to everyone as data analysis moves out of the lab and office and into just about everyplace else.

Postscript by Barry Schwartz: On February 8th, Stephen Wolfram posted more details about this Pro offering on their blog.

So, queries like:

• Tom Brady Joe Flacco Passing Yards
• 49ers 2011
• New York Giants field goal attempts, field goals made
• Chicago Bears Record 1985
• Ray Lewis Tackles
• Detroit Lions vs. Denver Broncos Passing Yards

all return results that often include visual elements like a graph.

Statistical data for every NFL team and every NFL player back to 1985 are available and the number of statistics and the combinations you can create are endless.

What’s available? This page lists all of the categories of stats available. A guide like this is something new for Wolfram|Alpha and I’m very happy to see it. Hopefully, more of these guides are in the works.

Where Is The Data Coming From?

It’s one thing to provide access to statistics, but it’s quite another to make sure they come from a reliable source.

Wolfram has partnered with STATS based in Chicago’s northern suburbs. STATS is the leading provider of sports data. It’s you follow a sport or sports, play in fantasy leagues, or just look at the latest scores, you’ve seen and used their services.

STATS has been around since 1981 and I’ve been watching them grow ever since. The STATS first office was only a few blocks from where I grew up. Currently, the company is owned jointly by the Associated Press and News Corp.

The one caveat (at least for now) is that NFL statistics are NOT real-time or near real-time. In other words, if you’re watching the games on Sunday, the stats accessible via Wolfram|Alpha will not be updated after every completed pass, made field goal, etc. Currently, NFL data is updated weekly.

Hopefully, this will change by next season and real-time data will be available. Also, I would love to see all-time statistics included in the database. For example, all-time NFL rushing leaders, all-time Chargers passing leaders, all-time Bears vs. Lions record, etc.

Wolfram will also be working with STATS to provide the same type of info for other sports including baseball and basketball.

Wolfram|Alpha’s Largest Hurdle

I said even before Wolfram|Alpha launched that the company’s biggest challenge is not the technology or data sources. They’re experts in this area and continue to do a great job.

The biggest obstacle Wolfram faces is marketing. Marketing Wolfram|Alpha in a time when most people live in a Google or Bing-centric world, is a serious challenge. Of course, this is challenge is one for many companies.

Luckily, after many years, I’m slowly starting to see more people open to trying something different and adding more tools to their search toolkit.

Of course, what search tool you use and when you use it all depends on what you’re looking for and your information need. More and more often, Wolfram delivers not only the data you need but also visualizations and related material.

Comparing Wolfram|Alpha to other search tools can be unfair. It’s not a general purpose search tool but rather a “fact engine” or interactive reference resource that you can query using natural language using data from curated and trusted sources.

I speak from experience that once people understand what Wolfram can and cannot do then see practical examples of what Wolfram has to offer, they’re often hooked.

By adding NFL statistics (and soon more sports statistics) Wolfram|Alpha has powerful ammo to reach out to new users with and not only provide them authoritative American football statistics but also show them all that their product has to offer.

Exciting!

To access this literal, real-time flight search, just type “flights overhead” into the Wolfram Alpha search box, and voilá! If you’re using a desktop computer, Wolfram Alpha does its best to figure out your location automatically and — if there are flights in the sky above you — lists the commercial flights in the area. You’ll also see the planes’ altitude and angle against the horizon, how far away they are and the aircraft type.

Clicking on any of the flight titles gives more details — I can learn, for example, that American Airlines Flight 621 flies between Seattle and Dallas/Fort Worth. (This flight is probably a lot closer to Dallas right now than what Wolfram Alpha is showing me. I live in a rural area and we don’t get a ton of overhead flights, so it may be defaulting to show flights that have been overhead; those of you in more populated areas will probably have better real-time results than me.)

Chances are good that you have a solid roof over your head, so desktop search probably isn’t the primary use case here. You’ll probably be somewhere outside with a smartphone or mobile device and want to know where that airplane above you is going. And thankfully, the same search query works just fine on the mobile version of Wolfram Alpha. Here’s an iPhone screenshot, and I’m not using the Wolfram Alpha iPhone app — just the regular Safari browser.

Obviously, a couple flights left the skies above me between the time I did the desktop and iPhone screenshots. Not shown, but further down the page (on both desktop and mobile) is a sky map showing the same flights in a more visual way.

Wolfram Alpha’s blog post says you can also use this feature to find out things like what you were looking at when you shot a photo out the window of your airplane in mid-flight.

This super awesome, literal, real-time flight search is available for U.S. searches only. And since the data comes from the Federal Aviation Administration, it only works on searches for flights that have at least one endpoint in the U.S.

Seriously, when I was a kid, every time I saw a plane in the sky above me I wondered where it came from and where it was going. I’ll probably start wondering the same thing again now … and getting answers. Finally.

Questions? Tweet to @Wolfram_Alpha and include the hashtag #WAChat.

Today, Google is opening up the Public Data Explorer for anyone to upload, share and visualize data sets. To facilitate this, Google has created a new data format, the Dataset Publishing Language (DSPL) (see the documentation on Google Code for more information). To upload data, click on the “My Datasets” link on the left hand side of the Public Data Explorer. Once imported, a dataset can be visualized, embedded in external websites, and shared with others like a Google Doc. If you have questions or feedback, you can post a message in the DSPL discussion forum.

The Public Data Explorer uses technology Google acquired when it bought Trendalyzer in March 2007. The datasets that Google has curated are from authoritative sources such as Eurostat, the International Monetary Fund and other sources—data that until Google made it available would have been thought of as part of the invisible web.

Opening up the Public Data Explorer is similar in many ways to what Factual has been doing for a couple of years now. Factual is a self-described “open data repository.” Like Wolfram Alpha, a “computational knowledge engine,” Factual seeks to create order from chaos by allowing anyone to share and mash open data on any subject, structuring information in database-like tables.

Google is also inviting official providers to request that their datasets appear in the Google Public Data Directory; these providers should contact Google to discuss this process.

That seems to be one of the conclusions you can draw from a fun blog post today by Wolfram|Alpha founder Stephen Wolfram.

Writing about the Jeopardy battle between past champions Ken Jennings and Brad Rutter and IBM’s Watson computer (the computer won a dry run earlier this month), Wolfram decided to see how several search engines would fare when asked to answer Jeopardy clues.

The results? Google scored highest in both having the answer show up anywhere on the first page of search results and having it show up in the first result on page one.

Ask.com was a close second in getting the correct answer anywhere on page one, while Bing was a close second in getting the correct answer in the first search result. Wikipedia’s search was last in both tests.

But Ken Jennings, Jeopardy‘s all-time champion in wins and earnings, answered about 79% of his questions correctly — a lot more than any search engine. Which suggests two things: Either he’s an immediate acquisition candidate in Mountain View and Redmond, or he should just start his own search engine and answer queries as they come in.

There are also some interesting tweaks to the results pages. One is the addition of linked data inside the search results, which you can see on this screenshot.

I did a search for “space needle,” and in the data that Wolfram Alpha returns, the words “Seattle,” “Washington,” and “United States” are all links that run searches on each term.

The data sets also now have dog-ear reveals that can be used to get access to links for copying the text or saving the images that appear in the search results.

Wolfram Alpha says the changes were needed to “allow expansion for upcoming consumer, enterprise, and developer products” later this year.

It is expected about a year after they introduced the Wolfram Alpha iPhone app. The launch date is October 6, 2010 and they are working with T-Mobile to launch the app. The app will be available for download on the T-Mobile Channel for download on the G2 and of course on the Android Market. It will work with devices running Android 1.6 and later.

This time, Wolfram Alpha set the initial price to $1.99. Back when they first launched the iPhone version the price was $49.99, but they lowered it to $1.99 months later and offered refunds.

The purpose is to enable publishers, bloggers, professors, teachers and so on to easily build quick widgets to add to their content and materials. You can build widgets based off of Wolfram|Alpha’s large database of scientific information and data. Widget creation is all based off the initial query, which then can be made into a simple form that can be used by anyone. I’ll explain this in more detail below.

First, you can browse the widget gallery to get ideas on some of the powerful widgets you can create. For example, you can build widgets that show your heart disease risk level, compare company sizes or give you the Thermodynamic properties. You can browse through the gallery to come up with new ideas and even modify someone else’s widget in order to build a new one.

One of my favorite uses of Wolfram|Alpha is as simple as typing in a date and getting information about that date. Sometimes, I simply want to know how long ago that date occurred, while other times I want to know what time sunrise is on that date or who was born on that date. I decided to create a very basic widget that let’s people quickly plug in any date and get a quick data sheet on that date.

For now, all anyone can embed a widget but to create a widget yo need a widget developer account, which is free. Wolfram|Alpha will release a commercial version of this widget creation tool, which will not have the requirement of users sharing all widgets. Let’s say you wanted to create an internal widget for company use only.

Widget Creation Process:

(1) Start off by typing an example query, in my case, I simply wrote a date. But your query can be a lot more complex than just this:

You can then preview this query to make sure it looks right:

(2) The next step is to template your query. By template, I mean you need to create variables for the parts of your query that should be variable. In my case, I want the January, 25 and 1985 to be all variables that can change in my widget.

It really is as easy as clicking on and highlighting part of your query and selecting the variable text, options and so on.

(3) Now you can customize the colors, the position of the boxes, the default variables and if you want to define any specific variables to be selected in a drop down or radio box. You can add titles, change the alignment, add spacers and much more. This step is about the layout of your widget:

Clicking on a cell gives you the ability to set specific changes to that variable:

(4) The amount of information given for any particular query can be tremendous and sometimes you want to only show a small portion of that output. Wolfram|Alpha gives you the ability to control the output type on this next step:

(5) You can then select the method you want people to view the results of the widget. Do you want the “lightbox” affect, which overlays the results on the page? Do you want a pop up window for the results? Maybe you want to embed the results using an iFrame. You can even adjust the width and height of the output box and preview the look:

(6) Finally, you name your widget, give it information for the widget gallery and get your embed code:

(7) Published widgets show up under the “My Widgets” section and it gives you the ability to see some basic statistics on the widget. Here is a picture: