Google, Profile Capturing, and User Intent
March 18, 2010
On March 16, 2010, Google nailed a patent for its invention “Profile Based Capture Component,” US7,680,809. On the surface, maybe not so exciting. Google wizard Steve Lawrence had a hand in this invention filed in early 2004. Don’t you admire the turnaround time? Here’s the abstract:
An indexing system in a computer system may include applications, a capture processor, a queue, a search engine, and a display processor. The indexing system captures events of user interactions with the applications. Events are queued and if indexable, indexed and stored for user access through the search engine. Capture components in the capture processor can include a keyboard capture component that processes user keystrokes to determine events. A display capture component captures event data from windows associated with the applications. Display event data can be captured on a polling schedule or based on state changes of window elements. To determine target applications and window applications of interest application profiles and window profiles can be used.
Google’s predictive methods revealed.
Stephen E Arnold, March 18, 2010
A freebie. I will report non payment to the USPTO which recently rejected a fax because it was “backwards”. The other pages, as I understand the event, were not backwards.
Google Wants Independent Bands
March 18, 2010
Short honk: Ask yourself this question, “Why does Google want independent bands on YouTube?” Read the news story “YouTube Introduces Musicians Wanted for Indie Bands.” One piece of the answer appears in this comment:
To get on to Musicians Wanted, you need to apply through the site’s partner programme. If you are then selected by YouTube, you will have your own dedicated page where you can add gig listings, links to buying your music and your live videos and music. To get money from the site, you will get the ‘majority’ of advertising revenue. And you will also get a share of funds if your music video is embedded on to other sites.
For the full answer, you can read my forthcoming study which will be published by Intellas in mid 2010. In the meantime, watch for other Googley actions in the non text sectors.
Stephen E Arnold, March 18, 2010
Shameless marketing of my new study. I paid myself to write this stellar article.
One UK Media Companies Pays Tribute to Two US Media Companies
March 18, 2010
Read “How the New York Times and CNN Try to Keep Up with the Tech Companies.” No, not the article. The title. First, the notion of “try”. The write up makes clear that the NYT and CNN are in try mode. Calvinistic? A less gifted pair trying to cope with Sergey and Larry? In French, “to make an often tentative or experimental effort to perform”.
Second, notice the phrase “keep up with”. Not “lead” or “change the game.” The notion is that a couple of 40 year old former high school stars are in fantasy camp with a bunch of retired pros and some college players. Yep, “keep up with”, not go “toe to toe” or “push the boundaries.” The details of the write up are interesting, but what is omitted are:
- Explanation of what Facebook has done to become one of the largest sites on the Web and a preferred source of news for lots of folks in some tasty demographics
- How these two sites plan to leapfrog Google and Yahoo
- Where the money will come from when it becomes necessary to build software, implement enhancements, and do big time research.
- What about that traffic challenge? Eyeballs are needed, then more eyeballs.
I think these questions make clear the challenges traditional publishing companies face as they try to learn new tricks, make them work, and find ways to pay for the costs that software and systems for the Web bring to the party.
Stephen E Arnold, March 18, 2010
Free, uncompensated information. I will report this writing for no money to the Jefferson County shelter manager.
The Math Club Wants Your TV
March 18, 2010
When I think of my high school math club, I remember comments about the oh-so-cool people on American Bandstand. Then it was dinner, homework, and reading under the watchful eye of mom and dad. TV was not the main event, and to this day, I don’t get too excited about information acquisition that is essentially sedentary and serial. The Google is more hip. I read “YouTube Is Huge: 24 Hours of Video Now Uploaded Every Minute,” and I realized that I am decidedly uncool. The article points out that the Google can suck in this content and “there’s no sign of it stopping.” Great.
Next I read the New York Times somewhat breathless write up “Google and Partners Seek a Television Foothold,” which recycles Google’s interest in pumping digital video from its content delivery network to televisions. For me, the most interesting comment was:
Google’s efforts to break into television advertising date back three years. Through a program called Google TV Ads, the company sells advertising on a small number of satellite and small cable television systems, as well as some cable networks. But interest in the program has been limited and the amount of revenue that Google has been able to generate from it has been small.
So that means that the NYT has determined that Google and TV has been in the works for 36 months. I am not so sure I agree. Perhaps the NYT researchers elected not to examine Google patent documents. The rich media locomotive has been in development for somewhat longer. The purchase of YouTube.com took place in 2006, and some of the technology referenced in the NYT write up dates back beyond that point in time.
What else is in the Google train yard? Lots more.
Stephen E Arnold, March 18, 2010
No one paid me to write this. My reference to a train means that I need to report non payment to Amtrak, a heck of an outfit.
Fabasoft Mindbreeze and Its Lotus Connector
March 18, 2010
I was able to read a white paper prepared by Fabasoft Mindbreeze about its updated Mindbreeze IBM Lotus Connector. The document is “Configuration of Mindbreeze Enterprise Search for IBM Lotus” and is available from the company. When I worked at Ziff Communications in New York City, I had an early exposure to the product. Since that time 20 years ago, Lotus Notes has found its way into many commercial and governmental entities. Those who love the product cannot live without it. People like me tolerate some of the system’s peculiarities exemplified by this question, “Why can’t you restore my email?”
Fabasoft is the successful Austria-based enterprise software and integration company. Mindbreeze is its search, content analytics and content processing subsidiary. The Mindbreeze engineers have developed a solution for organizations with Lotus Domino/Notes as well as a lot of other types of content systems. You can get Mindbreeze and its Lotus Domino/Notes support, snap it into your environment, and search for Notes content, even in mobile environments, including the RIM Blackberry, Apple iPhone, and Google Android devices.
In January 2010, I got a preview of the system and I received a copy of the white paper. I followed up with Daniel Fallmann, founder and managing director of Mindbreeze. Here’s what I learned in an email exchange on March 14 and 15, 2010:
What is the main focus of the IBM Lotus Domino/Notes support you offer?*
What is very important for us is that that the Mindbreeze Connectors run with a minimum of required configuration, even to very large scale. So Notes items and even complex Lotus Domino object models are very easy to adapt to fulfill the need of the customer/users. So Fabasoft Mindbreeze Enterprise makes it easy to search-enable any line-of-business application based on IBM Lotus Domino within a minimum amount of time with great results for the knowledge workers, even with their mobile information needs. Of course our customers get all the needed social search and federated search features built-in. We have a lot of Lotus partners that love the ease you can now search-enable IBM Lotus line-of-business applications.
As we offer an appliance as well you can buy the Fabasoft Mindbreeze Appliance or you can install Fabasoft Mindbreeze Enterprise and run the IBM Lotus Connector on a Linux environment which totally saves you the money of the operating system and enables you to even support our customer’s users with IBM Lotus line-of-business-application in the cloud with a modest and easy to calculate investment.
What is the method for indexing Lotus Mail which has been moved to an archive?*
Fabasoft Mindbreeze Enterprise follows the “link-information” (for example, a link to an archived mail for example in a Fabasoft iArchive for IBM Lotus Notes) left in the remaining item stub and index the archived information by applying the rights based on the stub object that’s left in the IBM Lotus installation.
How are emails across Lotus Notes installations indexed so that only the authorized person can see a single email or a group of emails?
Fabasoft Mindbreeze Enterprise uses the rights based on an IBM Lotus object information to evaluate if a user has the right to read information based on the document level or even extensible to the field level. Things like inherited rights and user name fields are as well taken into account. As Fabasoft Mindbreeze Enterprise is based on a modern distributed service architecture, it is easy to spread queries against several instances and respond to a user’s query. Thanks to our innovative technology and architecture we typically are up and running at customers in between 30min and 2 days, of course this highly varies on the customer’s needs.
When Lotus Notes is used with an IBM collaboration tool like Lotus Notes Traveler, how are the indexes federated so a single query retrieves the content across the Notes’s components?
First: It typically makes sense that Fabasoft Mindbreeze Enterprise crawler, filter, index services are located where the information is, so the best practice is to use the distributed architecture of Mindbreeze Enterprise Search to bring together information from several IBM Lotus databases. Second: As Fabasoft Mindbreeze Enterprise connects against the IBM Lotus web services it is even trivial to get information from different locations and index it in a central location.
How does your system’s pricing work?
We have a per named user pricing model as well as a concurrent use model that is very easy to calculate and use. Moreover the Mindbreeze IBM Lotus Connector supports custom object models as well and you can host the whole product on a Linux platform. As far as I know there is no other IBM Lotus Connector and search product, that can so easy adapt to IBM Lotus Domino object models for your specific line-of-business application. This of course has to be taken into account.
Do you support the Notes – Cisco Unified Meeting Place?
Fabasoft Mindbreeze Enterprise allows you to index all the calendar information for all meetings that you are invited to by Cisco’s Unified Meeting Place. This information can be updated in the index during the meeting place notification mechanisms. You could even index the audio content streamed via a Cisco MeetingPlace Audio Server by using speech to text functionality.
If you want to index Lotus content, we think the Mindbreeze solution warrants a test drive. Contact the company at http://www.mindbreeze.com.
Stephen E Arnold, March 16, 2010
When I am next in Linz, Mindbreeze promised me a pastry. Until then, this is an uncompensated post.
Google and Italy: More Hassles
March 17, 2010
Short honk: “Italy Goes after Google over AdSense Payments” revealed that Google has more hassles in Italy. The story in SEO Roundtable pointed out:
Italy is “going after Google for how they pay Italian AdSense publishers.”
The issue is that Google does not provide much detail about how it figures payments. SEO Roundtable noted that there has not been much coverage of this dust up. Interesting? The addled goose’s has noted that his AdSense payments have traditionally drifted down when a quarter is closing. Now there appears to be other concerns about payment predictability. Could the effectiveness of online ads been decreasing and Google is just a barometer of the economic downturn? Could an algorithm be stumbling? No solid information at this time.
Stephen E Arnold, March 16, 2010
No one paid me to write this. Since it is an international post, I will report non payment to the Department of State.
Perhaps Facebook Is Real Trouble for Google?
March 17, 2010
A happy quack to the readers who sent me links to the story, which if accurate, is big deal. I read “Facebook Tops Google As Most Visited Site in U.S. Last Week.” For me the most interesting comment in the write up was:
We saw hints that this milestone would come. Facebook beat Google for the first time in single day traffic on Christmas, followed by New Year’s, Valentine’s and some recent snow days. The social network also was the top place to go to discuss all thing Winter Olympics. Facebook also saw a 10% increase in search volume query in the month of February, a month when most search engines saw their query volume decline.
From intermittent spikes to a steady climb, Facebook appears to be out Googling Google in the traffic department. Will Google regain its lead? If the Google continues to suffer pecks from legal eagles and self inflicted wounds like the Buzz service, the odds may be getting longer.
Facebook makes finding information easy. Just ask your friends. The search box may be inappropriate for some Web users. Is this a trend? Nope. I want to wait for several months of usage data to accrue. Maybe the math club is losing out to the more popular folks?
Stephen E Arnold, March 17, 2010
Free. i mention data. I will report non paid writing to the Census Bureau.
Pew Documents What Some Info Vendors Will Learn the Hard Way
March 17, 2010
There are some tricks to learning. To memorize a list, put each item in a room of your house and walk the rooms, recalling each item by association. One of my classmates remembered the names of the Great Lakes with a mnemonic word. I prefer to look at survey data and let the numbers do the talking. The write up “Pew: Readers Prefer Ad Supported News to Pay Walls” provides me with some evidence that the dreams of traditional publishers to make yesterday’s revenue from gizmos like the iPad and the Nook might be just a figment of the imagination.
According to Pew, the oh-so-reliable research outfit, the article reports:
when it comes to online news, getting people to pay for content they otherwise value is “like trying to force butterflies back into their cocoons.
Yikes. People must not know this factoid which is pretty well understood among the savvy, but ageing commercial database publishing crowd.
I found this passage fascinating:
First things first: Pew notes that last year, online advertising saw its first decline since 2002. Numbers from eMarketer said that revenues fell by a total of $1 billion between 2008 and 2009. Still, a full 81 percent of Internet surfers say they’re cool with online ads if it means the content remains free, although “much of that is because they find them easy to ignore.” Further, 21 percent said they click on ads “at least sometimes”—much higher than we expected—and that number goes up when the user is more active. For example, among daily Internet surfers, 28 percent reported clicking on ads. For people who visit at least six sites per day, the click rate is as high as 37 percent.
Where’s the revenue going to originate? In my opinion, the former country club owners will be looking for regulatory help in the form of a “news tax” or some financial piece of the online revenue action from the new owners of the information country club. I caddied for peanuts and I don’t think the new country club proprietors will be too keen to give up too much cash to run “real news”.
Stephen E Arnold, March 16, 2010
Free, free as a goose. No one paid me to write this article. My reference to a goose reminded me of the Bethesda Country Club member who bludgeoned a swan to death decades ago to much fanfare. I will report my killing of this story to the new manager of that country club in suburban Washington.
The Open Source Card in Apple HTC Dust Up
March 17, 2010
The goslings and I are not attorneys. We are addled geese, and we do not understand the ins and outs of litigation. The article “Apple’s HTC Patent Lawsuit Is a Bluff” presented some angles on the Apple HTC patent matter. HTC seems to be a more convenient target than Google because HTC is manufacturing Android phones. These devices have some shared DNA with Apple’s iPhone. The write up suggests that Apple is engaging in some saber rattling. Litigation is expensive and risky, particularly if the parties cannot reach an out of court settlement. Juries can be darned exciting. For me, the most surprisingly passage in the write up was:
Android’s open-source status creates all kinds of logistical and legal problems for Apple. The company really doesn’t want to be labeled with a big Scarlet Letter as an open-source opponent. Apple has benefitted from open-source community development. It’s a vocal group Apple doesn’t want to piss off. Then there are all the nasty legal issues and potentially damaging precedents should Apple make a frontal open-source assault.
Android, as new Googler Tim Bray pointed out recently, is not about open source. The idea is that Apple does not want to be hoisted on the open source noose.
My view is that open source is becoming a highly charged phrase. Marketers, programmers, and investors have their own view of the concept. If the write up is correct, will Google use its open source approach to create more problems for Apple? There are some advantages associated with open source. But there are also some advantages associated with the proprietary approaches taken by companies like Apple.
Open source “plays” have a dual nature. Viewed one way, open source decreases the “lock in” that most vendors covet. Viewed another, open source could be a kinder, gentler form of getting people into a more spacious walled garden.
One thing is certain. The Apple patent matter will be with us for many months and there will be twists and turns as Apple builds out its server centers, gets serious about search, and expands into cloud services for its chain of devices.
Stephen E Arnold, March 17, 2010
Free. The bane of real journalists. I did this without compensation or the hope thereof, and I will report this to the event manager at the National Press Club.
Globalbrain Version 5
March 17, 2010
My feedreader sent me a story via “Your Story” tagged “Globalbrain v5 Provides More Customizable and Flexible Search Functionality.” According to the write up,
Globalbrain is highly scalable and enables access to vast amounts of structured and unstructured data (including e-mails, e-mail attachments, OCR-ed images, PDF files, word processing documents, spreadsheets and hundreds of other formats). Organizations and their users can simply search for information by using queries comprised of phrases, sentences, paragraphs or even entire documents of text rather than complex Boolean logic or complicated taxonomies.
One of the highlights of the new version, according to the write up, is:
At query time, the engine can decide when to use an inverted index approach for common keyword type queries or the fuzzy contextual based search. The engine can even use a combination of these search approaches. This approach, along with faceted groupings, allows users to further drill down and fine-tune results.
More information is available at www.brainware.com.
Stephen E Arnold, March 17, 2010
A freebie. No one paid me to write this item. I will report non payment to the Dulles Toll Road authority. Brainware can be reached by that route if you wish to visit the firm.
-
- Subscribe to Beyond Search
Feature archive
News archive
-
