Powerset

Like many features at Powerset, our highlighting feature came about through the ingenuity of one of our engineers. Ian was having a discussion about circular polarized filters for camera lenses with his brother. He copied and pasted some text from Wikipedia into an IM chat, but his brother was dubious: “Where did that come from?” After Ian tried, unsuccessfully, to navigate his brother to the right passage in the Wikipedia article, he thought, “There must be a better way!”

Hence, a new feature was born. User Highlighting allows you to highlight sentences on a Powerset page. The highlighted passage persists when you bookmark or share the URL.

To use, just click the highlighter icon in the Article Outline and then click any sentences that you want to highlight. You can copy and share the URL; your modified page will automatically scroll to the highlighted section. This is made possible by the fact that Powerset semantically marks up Wikipedia and knows where sentences begin and end. It’s a great tool for sharing notes with a friend, posting a specific sentence in context to your blog, or proving someone wrong in an argument.

For example, let’s say that you wanted to share with a friend that John McCain has been a guest on The Daily Show more then anyone else, as of 2006. As opposed to just sending over that bit of information, you can send a link directly to the text. Now they can read the fact in context of the Wikipedia article, and read about McCain’s other television appearances. Check out the video below for a full demonstration.

Powerset User Highlighting demo video from Ian Collins on Vimeo.

Find a great use for the Highlighting feature? Let us know in the comments.

From October 20-22 Powerset will be attending the Internet Librarian Conference in Monterey, CA. This year’s theme is “Beyond 2.0: User-Focused Tools & Practices,” and Danny Sullivan from Search Engine Land will be making a Keynote Presentation “Search Engine Land: What’s Happening Out There?”

We’re excited to be able to speak to so many internet-savvy librarians, to show off how Powerset is a cutting-edge research tool and to gather feedback about how to make Powerset even better.

Powerset is sponsoring the Internet Café.  If you’re attending, stop by to check your e-mail, learn about Powerset, and get some nifty schwag!

Powerset is excited to be participating in Web 3.0, held in Santa Clara from October 16-17, 2008. 

Scott Prevost, Powerset’s general manager, will be keynoting in the morning on Friday, where he’ll describe The Road to Semantic Search.  Also, Mark Johnson, a Powerset Program Manager, will be hosting a roundtable discussion on Thursday afternoon called Semantic Startup 101 – Successes, challenges, and strategic decisions, featuring Alitora, Cerebra, Metaweb, and Evri.

If you’re interested in attending, you can get a full conference pass or just come to the Expo Hall and keynotes for only $50.  We hope to see you there!

Powerset officially became a part of Microsoft a little over a month ago and we’ve already completed our first few integration projects: Freebase Answers, improved captions for Wikipedia results, and new related searches using our Factz engine . 

These projects were meant to be achievable in 30 days and act as a first collaboration between the Live Search and Powerset teams.  We have plans for deeper integration in the future, but these projects gave us an opportunity to get to know our colleagues up in Redmond and drive greater understanding of our respective technologies.  All of these projects are currently being “flighted” on Live Search, which means that they are being shown only to a small percentage of users (if you get one, consider yourself lucky!).  Once we’ve gotten data back from the tests, we’ll plan next steps and decide what features will eventually roll into the product.

The first project was to expand the number of queries for which Live Search shows Answers.  Queries like San Francisco weather, MSFT, and Banff national park already produce answers.  Also, many celebrities from abba to frank zappa and even bloggers from seth godin to leo laporte show up with xRank biographies.  But, many topical queries do not show Answers today such as  musicians, albums, films, etc. For this experiment, we selected some of these categories and will return a topic summary with links, similar to the Freebase Answers we show in Powerset, using data from Freebase.  Eventually, we hope to give Answers for even more topics.
 
 

The second project was to use Powerset’s semantic technology to generate improved captions for Wikipedia articles.  Since Wikipedia articles show up in a large percentage of Live Search queries, it’s important that the captions are top notch.  These changes are transparent to the end user, but we’ll be able to analyze the Powerset captions versus the Live Search captions to see which perform better.



The third project is also transparent.  We used Powerset’s Factz extraction to generate a list of related searches for a set of queries.



We also started to use some of Live Search’s technology on the Powerset side.  You’ll notice that we now have “related articles” on Powerset enhanced Wikipedia articles.  We’re getting these directly from Live Search.

 

Powerset is excited that we’re already able to make improvements to Live Search.  Expect more announcements in the coming months, both of Powerset’s technology integrated into Live Search and of enhancements to Powerset.com.

-    Dr. Scott Prevost, Principal Development Manager, Powerset
-    Dr. Hugh Williams, Partner Development Manager, Live Search

Powerset+Microsoft hosted another Lunch 2.0 this year.  We seved delicious food, including sliders and cupcakes.  There were lots of people from many different companies present and we all celebrated the Microsoft acquisition of Powerset with our commerative shot glasses.  Andrew Mager of ZD Net did a great writeup of the event (and he tooks some rockin’ pictures). Terry Chay was there with his huge camera and took a gorgeous set of pics with his fancy camera.  In the photo is Marie Williams of SHIFT, Linda Chan of Powerset, and me.

Expect Powerset+Microsoft Live Search to be hosting more events like this in the future, so subscribe to our blog or check our Twitter feed for details. 

In fact, keep on the lookout for us at TechCrunch50 next week!

Powerset has always suspected that college students love Powerset because it’s such a great way to search and browse Wikipedia content.  However, the only evidence we had was from Powerset parents and glowing feedback e-mails from .edu domains. 

Last week, we conducted a study of 200 college students through Peanut Labs, which confirmed our suspicion: college students are huge fans of Wikipedia.  Here are some of the highlights of the study.

• Though 90% of students have used Wikipedia to complete an assignment, a surprising 73% of students have been explicitly told by their professor not to use Wikipedia.
• Also, about a quarter of students always use Wikipedia when they’re completing an assignment.
• Not surprisingly, the most common use for Wikipedia is initial research.  However, about half of the respondents said that they use Wikipedia as a link to secondary sources of information.  That suggests that the reference section at the bottom of a Wikipedia article is an extremely valuable starting point for many topics.  Also, a third of students say that they’ve used Wikipedia as a primary research source.
• Students find Wikipedia very valuable.  28% of students thought that Wikipedia was a very valuable resource, 49% thought that Wikipedia was relatively important, and 23% thought that Wikipedia was only marginally helpful.

If you have any questions about the survey or you’d like to talk to someone at Powerset about why it’s better to read Wikipedia on Powerset, drop us a note at press@powerset.com.

On Friday, August 1st at 5:15 p.m., Powerset officially became part of the Microsoft family.  Though the deal was announced on July 1, 2008, there were a few more i’s to be dotted and t’s to be crossed.  That’s all finished and the deal is done.  Our employees are excited to be a part of Microsoft and the Live Search team.

Integration is underway in force.  This Thursday and Friday, there will be a bunch of folks down from Redmond for a Symposium to share details about our respective technologies.  We’re excited to learn about everything now available to Powerset in the Live Search stack.  We’re also excited to share the cool things we’ve cooked up here at Powerset and start planning opportunities for us to integrate with Live Search.  Check back over the next month to see what we’re up to both on this blog and the Live Search team blog.

We’re excited to announce officially that Microsoft has signed an agreement to acquire Powerset.

Powerset has always been a small company with big dreams, with the ultimate goal of changing the way humans interact with computers through language. We set out to improve search by indexing Web pages based on the meaning expressed in them rather than just the literal words. Powerset licensed breakthrough technology from PARC, hired world-renowned computational linguists and search engineers, and recently released a search and discovery experience for Wikipedia articles. Our technology helps to improve search results and also makes new features possible, such as Factz, which aggregates information from many articles to summarize a topic.

With any startup, the challenge is to take the seeds of an idea and grow it into a viable company. At Powerset, we transformed our idea into a world-class semantic search platform, demonstrating the future of search with our Wikipedia search experience. But building a large-scale semantic search engine is expensive, requiring an engineering effort and computing resources beyond what most start-ups could ever imagine. Because our goals around improving search align so well, Powerset has decided to team up with Microsoft. We believe that this is the fastest way to bring our technology to market at a large scale. 

Microsoft shares our goal to improve search through deeper analysis of queries and documents, and understands that our technology and expertise will play a key role in the evolution of search. With an existing search infrastructure, incredible capital resources, unlimited data, a leading search team, and clear mission to revolutionize the search landscape, Microsoft can rapidly accelerate our progress in building semantic search technology and bringing it to full Web scale.   When we launched our first product, we heard: this is great, but when and how will we get Powerset to go beyond Wikpiedia?  Microsoft accelerates our ability to move Powerset to the entire Web faster than anyone could have imagined.

Powerset will continue to operate much as we currently do, working in the same building, with the same organizational structure, and with the same uniquely talented and growing team (apply on our jobs page). We’ll continue to tackle the hardest problems in parsing, semantics, ranking, indexing, scalable computing, user experience and all of our other specialties. But now we’ll do it with the support of Microsoft and the vast resources of the entire Live Search team. 

Over the past couple of years Powerset has made amazing progress. Starting with just a big idea, we licensed the best linguistic technology, recruited a top-notch team, built out our datacenter, engineered a world-class semantic search platform, tackled deep natural language issues, improved relevance, innovated an interface and launched a great product.  So few start-ups ever tackle such deep, scientific problems successfully and create the kind of value we’ve delivered in such short order.

For now, Powerset.com will continue to host our Wikipedia Search & Discovery and we’ll be continuing to experiment with our product, based on user feedback. But, expect many announcements from us in the coming months about how we’re integrating our technology and features into Live Search.

As always, if you’re a member of the press and would like to speak to someone from Powerset, please contact us at press@powerset.com.

A big thanks to everyone who participated in our Factz contest to win a Powerset t-shirt! It was a lot of fun to see people’s motivation and creativity in finding Factz. Here are some of our favorite blog posts.

Megha Agrawal wrote an excellent post about researching "at the speed of light." She concluded his blog post by saying, “So if you’re having a class assignment due by tomorrow and you don’t have time to thoroughly read and understand the topic you can make use of the Powerset Factz that are fun to discover about a particular subject.”

Christian Straub found answers to questions such as Who killed jfk? This query is a great example of how Factz can directly answer a question, as well as provide further reading on a topic. Christian jokingly lamented that, “Powerset has actually managed to get me to waste more time (or educate myself more?) on Wikipedia. Instead of clicking at random links in random articles (which is still awesome by the way), you can type a subject or a question or a relation, and Powerset will often find some interesting factoids relating them."

Yuanjie Liu from China wrote a great post that included the query What did Thomas Edison invent? These Factz provide insights on Thomas Edison and the light bulb, pulling relevant Factz triples from articles such as “Timeline of electricity,” “Precursors of film,” and “Edisonian approach.”

In his blog, Visish Srinivasan used a query related to his research: epigenetic. The results from Factz found the relation between epigenetic and senescence. He wrote, “Impressively enough, clicking on that actually mentions exactly what I’ll be researching - epigenetic phenomenon and their relation to the causes of senescence.”

Steven P. Sanderson was especially ambitious and had a multitude of Factz queries on his blog. Our favorite was what does Microsoft own? which returns an easy to scan list of information strewn throughout Wikipedia.

Other Factz posts include: Caitlin Bales, Catherine H van Zuylen, Anand Kishore, James Grimes, Jr., Matt McMahon, Sandeep Sripada, Raveesh Meena, Nitin Karandikar , Mardav Wala, Becca Schonberg, Darren Louie, Richard Bradshaw, Boyan Yurukov, Rajesh Balakrishnan, eric, Dan Tam, Patrick Traughber, Vilson Vieira

For those of you that didn’t get a t-shirt but still want one, we’ll probably run another contest like this in the future, so stay tuned.

Powerset has always been a fan of the iPhone.  Over a third of the company owns an iPhone (roughly equivalent to the number of PhDs working here) and a handful took the afternoon off on the day of the iPhone launch to make all of their friends jealous.

With that many iPhones ringing around the office, it was kind of embarrassing to demonstrate a handicapped Powerset mobile search product.  Therefore, after launch, one of our first major projects was to create an iPhone-friendly version of our search & browse experience for Wikipedia.  If you’ve got an iPhone, just go to www.powerset.com and it will work automatically.

The iPhone web app includes all of the functionality that Powerset features on the Web site, but is designed to be easy to use on your iPhone.  We think that Powerset is currently the best way to search and read Wikipedia on your iPhone. 

For searching Wikipedia, you can enter a topic, phrase, or question and get back Powerset results.  For simple Factz-style questions (see our blog contest about Factz for more details), Powerset will return the list that we derived from sentences across Wikipedia article.  Side note: Dr. Bruce Horn, one of the original developers on the Macintosh and creator of the Macintosh Finder (as expressed by what did bruce horn write on Powerset), is manager of Powerset’s Semantic Search Platform.  Also, when we get an answer from Freebase, we’ll display those results.  And, we’ll show regular search results for any query, whether it’s a keyword query or a natural-language style query.

Once you click on a search result, you will go to the Powerset-enhanced Wikipedia articles.  You can either browse the article or use the Outline to navigate directly to the section you want.

The Factz summary gives you an easy-to-read list of all of the Factz that Powerset was able to derive from the article, which often reads like a Cliff’s Note’s summary.  This is great if you’re on the go and looking to quickly scan the information in the article.

The search-within-the-page brings semantic search into the results page.  You can search for keywords or you can search for questions, e.g. “who stole pat’s rocket” on the Achewood page.

We’re really excited about this release and hope that, if you’re an iPhone user, you’ll use Powerset next time you need to look up something while you’re cheating at trivia night at your local bar, get details when you’re uncertain about some new fruit at Whole Foods, or just trying to convince your friends that you really do know the right answer.

If you’d like more information or an interview, drop us an e-mail at press@powerset.com.  If you have suggestions, join Powerlabs and make feature requests for the iPhone app.

Update: We created a quick demo video, for the luddites who don’t have iPhones.

Powerset iPhone Web App Demo from officialpowerset on Vimeo.