Big Bucks to NetBase for Semantic Search
April 13, 2010
Semantic search continues to fire probes into the information sector. The most recent laser flash was the $9 million that charged the batteries of NetBase. I read “Semantic Search Startup NetBase Gets $9 Million” and learned:
With the new cash, NetBase will have now raised $18 million since its start in 2005. The funding comes from Altos Ventures and Thomvest Ventures. Cash will be used to “aggressively develop and serve new markets.” NetBase says it will build out its sales, marketing, business development and product management functions.
Interest in semantic search continues to percolate. The idea is that indexing words is often not enough. In order to go beyond key word indexing, semantic search innovators employs a broad range of techniques. These methods can degrade into marketing buzz. Casual observers of search hear “natural language processing” and “entity extraction” and leap to the conclusion that smart software can understand text.
Well, sort of. But the point is that most people who have used search engines for information know that a key word can generate information wide of the mar. One example is the use of the word “terminal” in a query. If you are looking for a bus station and get hits to an untreatable cancer, you realize that the word “terminal” has insufficient specificity to get you what you want. On the other hand, if you enter five or six key words, you may get a handful of hits but these may be too specific.
The fix is to assemble a number of methods to make fuzzy boundaries sharper and make too sharp boundaries fuzzier. The idea is to create a recipe that yields results that are just right. Google has some formidable intellects working on the semantic challenges. A Paul Allen start up Evri, bought the high profile Radar Networks, to beef up its semantic offerings. Most search vendors assert that their systems use semantic technology.
NetBase describes its approach as delivering the “next generation of search.” The company says:
We are the insight discovery company. Founded by innovators Jonathan Spier and Michael Osofsky, NetBase develops and markets a next-generation semantic technology that reads and understands the English language. This technology is the basis for solutions that help our users answer complex questions faster, more accurately, and with greater confidence. And we do this at scale. NetBase finds and extracts the most relevant information from billions of public and private sources of online information. Our advanced technology combines with patent-pending lenses to provide context for search results and intelligently guide users to highly relevant answers.
The company’s technology “focus searches by organizing the patterns detected by our next-generation semantic technology in the context of a specific set of questions relevant to a specific discipline or audience of researchers.”
Hakia, another semantic player, has been successful in attracting investment as well. The challenge for any semantic vendor is to find a way to generate sufficient revenue to keep the investors in vacation homes and new BMWs. There is a race underway among a number of interesting companies with semantic solutions. Looming over the entire sector are giants like Google and Microsoft which are keen to use semantics and any other context generating technology to give their services an advantage.
Which semantic vendor will break out and deliver a solution that delivers a hockey stick growth curve. Azure chip consultants praise most semantic technologies. But the proof is not PR. The proof is a sustainable base of revenue, sufficient revenue to continue technology investment, and the agility to dance around the very big players in the semantic game.
NetBase may be focusing on a vertical strategy and it will be interesting to watch the story unfold.
Stephen E Arnold, April 13, 2010
No one paid me to write this.
OneRiot Identifies Challenges to Monetizing Real Time Info
April 13, 2010
OneRiot’s Kimbal Musk has identified the three challenges to monetizing real time information. The reasons appear in “Monetizing the Realtime Web” in the company’s blog. I agree that there is interest in real time information. Even SAS, the analytics giant, wants to hop on this fast moving content train. Law enforcement has long had an interest in knowing what’s going on, particularly in certain fast moving situations when mobile devices are used to pass messages. The challenges are, however, formidable. Mr. Musk identifies these hurdles:
- “Real time targeting”; that is, knowing what message goes to whom at a particular point in time. Advertisers want to fire info rifle shots, not shotgun blasts in my experience. However, real time targeting can be computationally expensive.
- “Data is everything”; that is, individual messages must be processed and converted into meaningful information. Google has had this challenge gripped in its teeth for more than a decade. Many organizations are struggling with this issue. There are costs and precision issues in addition to technical challenges to resolve. Better metadata are needed to make some real time information useful to an advertiser.
- Advertisers have some learning to do. Missionary marketing is important and some old expectations and habits can be difficult to change.
Mr. Musk provides some color about OneRiot’s successful approach provides a useful case.
The challenge is not just OneRiot’s. Google continues to tweak its presentation of real time results. I noted that our research suggests that users skip over the real time results. Some topics don’t have real time results; others do. Traditional searchers, therefore, don’t see information consistently in result sets. Consistency is important.
The larger issue, in my opinion, is that some real time results lack context. Additional information may be needed to make sense of some real time results. These injected content wrappers provide the user with the information needed to make sense of an otherwise cryptic or out of context item of information. If you run a query on a current event such as updates to the PGA tournament, the user presumably has context. But even these messages may need framing.
At this time, injection and wrapper technology is available, based on our research, just not deployed. Real time information is likely to benefit when more than the terse message is presented. Smart software may be able to shoulder the burden, converting isolated items into mini news stories.
Whoever cracks this problem will have an edge in monetization because the machine generated wrappers can have ads attached which may offer more advertising hooks.
Stephen E Arnold, April 13, 2010
Unsponsored post.
Magazines May Not be Rescued by the iPad
April 13, 2010
If you are a magazine, don’t expect the iPad to be your knight in shining armor.
While there is a lot of hubbub in the print industry about the many possibilities of the iPad to rejuvenate dying magazine sales, things don’t look so rosy upon closer inspection. TBI Research recently posted an interesting story, “Here is Why the iPad Won’t Save the Magazine Industry” listing the many ways this new technology will let down the magazine business.
For starters, magazines are planning on charging higher prices for digital issues than print, in some cases $15 a month. But an even bigger hurdle for this industry is the massive amount of free, searchable information available. For any topic imaginable, iPad users will be able to seek out low cost and free resources and directly skip costly magazine subscriptions. Doesn’t sound much like a savior to us.
Patrick Roland, April 13, 2010
A freebie.
If Books Become Apps, What about Regular Reading?
April 13, 2010
My view is that the era of text is drawing to a close. Words won’t go away, but the future is video and interactivity. Even InDesign CS5 allows moving stuff to be inserted into text. Text is becoming a sidebar. The real action is jiggly wiggly content. You can get a glimpse of the future in “The Amazing Media Habits Of 8-18 Year Olds.” The article is based on a study funded by the Kaiser Family Foundation and you can view the PowerPoint highlights on the Business Insider Web site. For me, the most interesting item in the Kaiser study is summarized in the screenshot below:
If books become multimedia, then book consumption may go up. The reason is that books will be more like games or TV, two popular pastimes for the sub 18 year old set. The problem facing any traditional publisher is shown in the slide below:
Looks like the sub 18 crowd is moving beyond text. Apple may be better positioned that either Amazon or Google in this sector. Which horse will win the rich media derby? A favorite or a contender off the radar at the moment?
Stephen E Arnold, April 13, 2010
Google to Newspaper Industry, Experiment
April 13, 2010
I found the article “Google CEO: ‘Journalism Will Triumph’” darned interesting. I am sitting in the Rome airport and reflecting on the suggestion to newspapers to experiment. I found the following passage particularly fascinating:
Speaking to the American Society of News Editors’ annual convention at the J.W. Marriott in downtown Washington, Schmidt showered praise on the industry, calling journalism an “art.” Schmidt said he reads three newspapers, and called their work indispensible. And he blasted blogs, saying that any questions about the value of newspaper editors can be answered: “Look at the blog world.” “High quality journalism will triumph,” he said.
According to the article, the Google speaker was not “adversarial”.
Three thoughts:
- Experiments are underway with the iPad and all manner of innovations to generate revenue. In fact, the experiments have been underway for decades. How much time and money remain?
- Google offers a solution. I call this “surf on Google.” Problem is that newspapers may not be adept surfers.
- Google is building bridges, but is it too late? The focus seems to be shifting away from text, where Google dominates, to the Facebook approach to content.
Read the article. Decide for yourself.
Stephen E Arnold, March 12, 2010
No one paid me to write this.
Google Strengthens Visual Search Team with Plink
April 12, 2010
The fastest way to get technology and staff is to buy a company. Google, according to AFP, grabbed UK based Plink, a search system for “artwork” and pictures. You can read “Google Buys Visual Search Start-Up Plink” and get some basics. (Yahoo News links often go dead, so you may have to do some hunting for the story.) I am at an undisclosed location east of Italy and I don’t have my full Overflight service available. Some basics:
- The url for the company is / was http://www.plinkart.com
- The company offers a mobile phone app. Take a snap and the system identifies the object
- You can see the company’s news release at http://www.plinkart.com/blog/?p=77
The company will keep its mobile app, but:
we won’t be updating the app and will instead focus our development efforts on Google Goggles, so you’ll see new functionality appearing there in the future.
Google has a clutch of patent documents and technical papers that address visual search, image recognition, and video segmenting. Google is moving beyond text.
Stephen E Arnold, April 12, 2010
A freebie.
Does a Google Manager Have AOL News at Sea?
April 12, 2010
If AOL is going to be the 21st century media giant it wants to be, someone better tell the management.
The media behemoth has been suffering an identity crisis that doesn’t seem to be getting better. A recent Business Insider story, “AOL’s Content Strategy is Still a Mess,” showcases a company with its heart in the right place, but its execution failing at every opportunity. One of its main objectives is to create high quality online content by hiring seasoned journalists and editors. The logic being, if everyone has a voice in the new web, those with proper writing and editorial skills will speak loudest. This sounds great, but swallows great gobs of time and money. Meanwhile, the company is also bum rushing search engines with a glut of SEO-friendly content from practically anyone off the street. The result is a two headed monster going nowhere fast.
Patrick Roland, April 12, 2010
No one paid us to write about AOL.
Trade Association Defines Search Its Way
April 12, 2010
I don’t know much about the Technology Services Industry Association. Most associations serve the narrow requirements of a select membership. Some “associations” are not really associations. I learned that the outfit called the “National Association of Photoshop Professionals” is a company that owns an association, a magazine to which I subscribed once, and a string of expensive “how to” conferences. TSIA may be like the American Bar Association or it could be like the NAPP outfit.
What caught my attention was a news story that we snagged in the Overflight system. The headline was “TSIA’s “Intelligent Search Market Overview” Report Identifies Innovative Criteria for Search Technology Selection.”
Reports about members are bread-and-butter activities in some “associations.” I don’t have a problem with a member profile write up but I did stumble on this passage in the news story:
The following search specialists participated in the study: Attensity, Baynote, Clarabridge, Consona CRM, Coveo, InQuira, KANA, nGenera CIM, Q-go, and RightNow.
So what’s the big deal? Well, for the addled goose, this listing of companies as “search specialists” is one of the most egregious examples of confusing an enterprise procurement team I have encountered. Tossing in the word “intelligent” just plain flummoxed me.
Let’s look at this line up of “search specialists”.
First, there’s Attensity. This is a deep extraction content processing firm. Recently the company has moved from the intelligence market sector into advertising, sentiment analysis, and other markets. The company’s technology processes content and generates a range of outputs that can used to figure out whether email is positive or negative. The firm provides basic search functionality, but the company is a vendor that adds metadata to content objects. Those metadata can be manipulated in a number of ways. One of the uses is to locate documents tagged in a consistent manner by the Attensity system. This is impressive technology, but it is a component of search, not a search system along the lines of the Autonomy, Exalead, or Google offerings. This is an error of confusing the parts with the whole, and it is a serious logical flaw in the TSIA write up.
Second, there is Baynote. This is a company that offers a “recommendation engine.” Think of Baynote as a more robust, more configurable version of the Amazon system. The idea is that the firm’s technology can process information about a Web site visitor and then generate outputs that reveal intent and context. Again, this is powerful technology, but it is not search. Baynote supplements more comprehensive search-and-retrieval systems. Baynote is what it says it is, a recommendation engine. Why label it a search system? (I think it is to create a report for which inclusion was advertising and revenue perhaps?)
Third, Clarabridge is a company that, at one time, had some of the good old MicroStrategy DNA. The system can process the type of data collected in a traditional structured business intelligence system and perform additional functions. Instead of coding a report, a client can use the Clarabridge system to frame a Google-style query and get a report out. Recently Clarabridge has embraced the Attensity approach of pushing into customer support and other allied market sectors. There’s good business logic behind this shift, but Clarabridge is not a vendor of search-and-retrieval technology. In fact, one might need both Clarabridge and a more robust text processing system to get most users comfortable with the outputs in a business application. This is a repositioning of Clarabridge from business intelligence to a specific vertical application. Okay with me but misleading in my opini0on.
Consona CRM is just what the name says. Customer relationship management. The company includes a basic search system with its software, but the core competency of the company is in supporting a call center application. Try to extend the system over the full spectrum of potentially relevant content in an organization, and you will find the need to look for other bits and pieces. This is a naming error because CRM is not search. Search is a utility within CRM in my opinion.
Coveo is a vendor of search and content processing. Unlike the other firms, Coveo started with a search system and has now created a solution that fits into a customer support application. Coveo’s platform makes possible more than customer support. While it is important to explain how Coveo’s customer support solution delivers call center features, it is a disservice to Coveo to position the company narrowly.
InQuira is a company formed by fusing two other firms. The company has natural language processing technology which is sold as an engine for self-help systems. The firm can deliver a broader search solution, but I think of the company as a niche player in the customer support sector. I don’t think of InQuira in the same way I perceive the Microsoft Fast type of solution. In my experience, there are some interesting parallels in the trajectories of the two firms that merged to create InQuira and the fusion of Microsoft and Fast Search & Transfer. InQuira, therefore, is a search system but it is one that has been shaped to somewhat special purposes.
KANA is a help desk vendor. In a meeting with the firm years ago, I was told that KANA had state of the art search technology. The demo showed that a customer support rep could enter a product name and see information about that product from different repositories. This is indeed search. In my opinion, it was primitive but it worked. Since that demo, I have not considered KANA a search vendor. In fact, I have resisted KANA as a vendor of knowledge management solutions. The firm builds and maintains customer support system for a large number of companies. Some of these companies have multiple search and retrieval systems plus KANA.
nGenera says that it is a vendor with systems that power “the collaborative enterprise.” One function of some nGenera applications is search. Search is like the hubcap on a Hummer, and I am not sure that nGenera itself would describe the company a search vendor. The company says, “Our solutions combine strategic insight, onsite services and the most comprehensive suite of collaborative applications on the market.” I have no problem with nGenera, but I think that describing this firm’s products and services as “search” is just misleading.
Q-Go says that it delivers “relevant online answers, better customer service.” I suppose I could interpret this phrase as meaning enterprise search or an Intranet and Web combined search, but I think that would be a real stretch. The company, like others in this list, focuses on customer support. Search is one facet, but it is not the complete system the firm delivers. In fact, the company asserts, “Q-Go guarantees a six month return on investment. Not many search vendors can make this type of statement in my opinion.
RightNow, a TSIA silver partner, is a customer support platform vendor. The company has moved into cloud computing and includes a search and retrieval function in its products. As one of the leaders in call center and related functions, search is important, but RightNow is not a vendor of enterprise search solutions. Maybe the company is moving into this sector? I know that when I hear “RightNow”, I think of the company’s push for “customer experience.” In my files I had a clipping that addressed the function of indexing a Web site with RightNow. The answer in the 2007 item here was that the Web indexer was a separate component. But since 2007, I haven’t seen much about the RightNow search system in the enterprise. Labeling RightNow as a search vendor seems to be a stretch. In 2007, a change to an indexed article required an index rebuild to pick up the change. Not exactly what I prefer.
My view is that the term “search” is used as an umbrella to cover a report about customer support vendors. Some of the vendors deliver full service solutions with search as an after through. Some deliver a specific type of content processing. Some deliver a package search solution tailored to the needs of customer support.
It is confusing to me and probably some potential customers to slap the word “search” on these vendors. Perhaps the report would be more compelling if a more informative title and description were used? Perhaps some of the vendors are stretching their own capabilities to cover this lucrative market for reducing the cost of providing customer service?
Stephen E Arnold, April 12, 2010
A freebie.
Google and the European Telcos
April 12, 2010
The Financial Times’s “Google Accused of YouTube ‘Free Ride’” documents and fans the flames for a Google and European telecommunications dust up. You will need to read the FT’s article in full and make up your own mind about this issue. From my vantage point, if the contention increases, Google will have another legal matter to manage. Like the Viacom situation, high profile legal hassles create public relations problems and cost money. Accumulating legal problems sends a signal in itself. Depending on one’s relationship with Google, the signal can be good for business or an inhibitor for business.
For me, the most interesting passage in the write up was:
To increase the pressure on Google, the telecoms groups are interested in finding common cause with content owners such as media companies, which get little or no money from the technology company when it aggregates their content on Google News.
Facing anti-Google forces that act in concert could, in itself, become a distraction for Google’s senior management. Google has the technology to deal with almost any challenge. We will know, possibly by the end of 2010, if Google’s senior management has that same capacity. Even Google’s billions may not be enough to meet the cash needs of the many folks who want Google to pay for access, services, risk, or any other reason that enables a cash life support system from Google to them.
Stephen E Arnold, March 12, 2010
An unsponsored post.
Google Cloud Deals: Are Some More Equal than Others?
April 12, 2010
“Frustrations with Cloud Computing Mount” runs down a number of issues with cloud computing. The hassles wander all over the map. But the article contains a quite interesting point about Google’s cloud computing agreements. Here is the passage I found fascinating:
The big cloud customers, such as the City of Los Angeles, which reached an agreement for unlimited damages with Google when it contracted to use its Google Apps services , should it ever violate its nondisclosure agreements, can negotiate terms that may give them a transparency and enforcement leverage. But many other users don’t have that clout and, and in a lot of cases cloud providers may not even provide the logging information needed to prove a breach, said Jim Reavis, the founder of the Cloud Security Alliance.
I think this means that some are more equal than others.
Stephen E Arnold, April 12, 2010
Unsponsored post.
-
- Subscribe to Beyond Search
Feature archive
News archive
-
