|Review: Wolfram|Alpha iPad app|
Thursday, 28 June 2012 12:00
This 'Siri on steroids' app puts supercomputing into the hands of ordinary people.
My first impression of the Wolfram|Alpha reference app was that it was just a glorified search engine. However, unlike traditional search engines that spew out links to related sites of the search query, Wolfram|Alpha provides more information about the actual topic being searched. A lot more.
For example, typing ‘coffee’ into Google brings up a Wikipedia definition, a Google Maps summary of nearby coffee shops, images of coffee and a number of Web sites of the most well known coffee companies.
Type ‘coffee’ into Wolfram|Alpha, however, and you’ll get more information than you probably ever cared to know about the beverage. The app offers up nutritional information for one cup of coffee, average daily value ranking and nutrient comparison to other foods, before going into even deeper detail with separate sections for calories, carbohydrates, fats and fatty acids, protein and amino acids, vitamins, minerals, even alcohol content (0%, in case you were wondering).
The top section of the search results offers coffee-related search suggestions including ‘Starbucks coffee of the week’, ‘brewed’, ‘espresso’, ‘decaffeinated’ and ‘prepared with tap water’. The app automatically categorises coffee as a food, but provides the option to use ‘coffee’ as a plant, a word or a general material instead. Clicking on ‘plant’ brings up the scientific name, an image, growth information, geographic distribution, taxonomy, species identifiers and other members of the same family. Like I said, probably more than you would ever want to know about coffee.
And therein lies the value proposition of the app. It’s not just a simple search tool – “it performs dynamic computations based on a vast collection of built-in data, algorithms and methods”. The concept, according to the ‘About’ section of the app, is to turn free-form natural language input into dynamically computed results. Its goal is to make all systematic knowledge immediately computable and accessible to everyone.
The thing is, the capabilities of this app go way, way past my levels of comprehension. I guess in that sense it achieves what it set out to.
Look and feel
The Wolfram|Alpha app is very clean. On first opening, the only element displayed on the screen is a search block. There’s a menu icon in the top-left corner and a camera icon situated next to the search block. That camera icon is whole other can of worms, which I’ll get to in a bit.
To input a search request, users tap on the search block, which brings up a keyboard and a scientific calculator.
The menu offers Examples, which I found particularly useful, as it highlighted just what this app is capable of – computation, analysis, comparison, estimation and information provision – for starters.
The History feature keeps a record of all search requests the user has performed, and the Favourites section stores all searches a user has marked as ‘favourite’.
Aside from providing troubleshooting suggestions, the tips tool explains to users what they can expect from the app – answers to specific questions rather than general topics, and answers about objective facts. And in case you were hoping to find out if the Tooth Fairy really exists or to get the phone number of Joe Soap, the app makes it clear that only public information is available and that “only what is known is known to Wolfram|Alpha”.
Computation for dummies
The Examples section gives the user a small idea of just what this powerhouse app can do. Remember that bit about it going way beyond my levels of comprehension? Well, after this list, you might be in the same boat as I am in.
Considering that these are just examples, it’s impressive that the app is able to compute:
* Mathematics (elementary, numbers, plotting and graphics, algebra, calculus, geometry, number theory, discrete mathematics, applied mathematics, logic and set theory, mathematical functions, advanced mathematics, mathematical definitions and famous math problems);
* Statistics and data analysis (descriptive statistics, regression analysis, statistical distributions, random variables, probability);
* Physics (mechanics, electricity and magnetism, optics, thermodynamics, relativity, nuclear physics, quantum physics, particle physics, statistical physics, astrophysics, physical constants, physical principles);
* Chemistry (chemical elements, chemical compounds, ions, chemical quantities, chemical solutions, chemical reactions, chemical thermodynamics, functional groups, chemical protecting groups, chemical formulas, 3D structure, pH indicators, orbital hybridisation);
* Technological world (communications, satellites, space probes, photography, bar codes, structures, nuclear power, nuclear explosions, carbon footprints);
* Web and computer systems (Web and Internet, notable computers, data transfer, RAID, file formats, character encodings, computer keyboards, string processing, hashing, display formats, passwords, CIDR, base conversion).
And that’s not the half of it. Other example categories include materials, engineering, astronomy, Earth sciences, life sciences, computational sciences, units and measures, dates and times, weather, places and geography, people and history, culture and media, music, words and linguistics, sports and games, colours, shopping, money and finance, socio-economic data, health and medicine, food and nutrition, education, organisations, and transportation. See the full list here.
Under each category, as can be seen from above, are a number of suggestions to users on how to search for related information. In the Health and Medicine category, for example, users are shown how to compute body statistics, body mass index, growth charts; how to analyse a weight loss regimen, how to compute calories burned in exercise, as well as maximum and target heart rates, blood alcohol content, estimated risk of heart disease. They are shown how to get information about a specific disease, how to compare life expectancies in several countries, how to get data on deaths from a specific cause. They can get information about a medical test, a test result, estimate conception and due dates of pregnancies, get information about a specified tooth, and can specify a drug by brand name, generic name or interaction information. They can get an overview of healthcare costs in a particular country, compare hospitals and – to really confuse me – “compute mean plasma glucose from HbA1c level”. Sorry, what?
And that’s just one category…
The possibilities are endless, it seems. But wait, there’s more.
Share the knowledge
Aside from bookmarking favourite searches, users can also share their findings on Twitter and Facebook. There is no option to e-mail or print the findings, however.
When sharing results on Twitter or Facebook, an Internet link is included that clicks through to a Web page with the information.
I’d be seriously concerned if I were a teacher today. This app would render homework headaches obsolete and could, ironically, be a serious hindrance to learning. By simply entering a geometry sum into Wolfram|Alpha, the app draws up a graph in a matter of seconds. The same goes for trigonometry and algebra – no left-brain thinking needed.
And as a Words With Friends fundi, the fact that I can calculate Scrabble scores, find anagrams and words matching a pattern instantly makes me a formidable player.
As if I wasn’t already blown away by Wolfram|Alpha’s capabilities, it decides to mess with my head even more with its image input/analysis function.
The review app did not have this function installed – it costs an additional $0.99 – but the tutorial gives a pretty good idea of what’s possible when users input an image for analysis. Image processing functions are able to colourise components or apply a gradient filter. They can isolate certain colours and specify the percentage of the image a particular colour makes up. Images can also be converted to greyscale or binarised.
According to Apple, parts of Wolfram|Alpha are used in its Siri Assistant. Parts of it. And I thought Siri was incredible. Turns out Wolfram|Alpha is Siri’s big, big brother.
Wolfram|Alpha brings supercomputing to the hands of ordinary citizens. The tool has infinite applications and would be useful to anyone – from high school students to scientists, hobbyists to the technically minded. It seems there is nothing this app can’t calculate or find information on; and if it doesn’t understand your request, it will make suggestions that are usually pretty accurate.
I honestly can’t fault this app. As much as I tried to trick it, I was outsmarted each time, and learnt something new in the process. There is not enough time, space or brain capacity (in my case, anyway) to try to fully explain this app’s functions. I highly recommend you purchase it and experience it for yourself.
Wolfram|Alpha is also available for iPhone and iPod touch.
Good: Extremely detailed search results, vast capabilities, seemingly endless computational abilities
Bad: It outsmarted me every time
Price: $1.99 plus an extra $0.99 for the image input and analysis tool
Download: iTunes link
Newer news items:
Older news items: