Searchwiki: A Social Search System
September 5, 2009
A happy quack to the reader in Europe who alerted me to Searchwiki. I navigated to http://www.socialsearch.com and ran some test queries. The content processed by the system comes from a metasearch across Google and Yahoo, among others. The system also indexes Twitter. My query for my current favorite Brazilian musician, Calinhos Brown, returned a number of useful hits. I located sites, videos, and an entry to an international artist database.
This is a revamp of an older service called Swiki.com. The interface features tabs that allow the user to slice the indexed content by Web, Local, Twitter, Images, and Videos. The engine provides an icon that shows where the hit originated, which I find useful. I also like the international flavor of the system.
Results pages include buttons for rating a site, sharing the link via Facebook, and adding an a comment.
Other test queries returned equally useful results with minimal latency. Let me highlight one other query for images: Expert Systems Cogito. Searhwiki returned a system diagram that I chased for days using other search systems. I was delighted with the 42 mostly on point results.
The company was founded in 1997 and you can get more information about the firm from the company’s About page. Looks quite useful and the name change is a very positive step.
Stephen Arnold, September 5, 2009
Google Financial Services: Another Wobbly Step
September 4, 2009
I described in 2005 several market sectors where Google wanted to probe and then (if the clicks arrived in sufficient numbers) enter. The idea was that for Google to sustain rapid growth it had to jump into new market sectors, not hive off different chunks of the search and ad business. This notion is now clearly evident, although in 2005 even the banks who paid me to describe Google’s long term strategy thought I was a stand up comedian on Hee Haw. Now the telcos, publishers, and new media fund managers are not finding my comments quite so entertaining.
Back to Google Global Bank or GBB: I articulated this idea at the 2006 BearStearns’ Internet conference and gave some examples; namely, the financial plumbing that hoses around Google’s Mississippi River of cash from advertising. This is cash in and cash out, score keeping, and a number of other automated services that allow Google to keep certain overhead costs associated with money under control. I also pointed to the then little used Google Check Out service. One brilliant MBA in the audience told me that PayPal was in a “game over” position. I replied, “Well, maybe.” The final example I gave was Google’s pretty lousy service at http://finance.google.com. I reported that I had heard Google had hired at least one person who had worked on the popular Yahoo Finance site. Another whizzy young thing now working as a Wal*Mart stock clerk pointed out that Yahoo Finance was a killer service and Google had a long way to go. I agreed.
With baby steps, the Google has been moving forward in the world that I sum up with the phrase Google Global Bank. You will have to spend money to get more pieces of the puzzle and descriptions of the technologies Google has disclosed to provide turbo chargers to its core revenue machine of advertising.
Google revealed after I had left to take my technical advisers Tyson and Tess to the park to chase rabbits and squirrels. You can read the somewhat disingenuous description of some new financial data services that Google made available for free in “Google Search Volumes and Economic Activity.” On the surface, Google’s resident economic wizard has dipped into some of Google’s data to make it possible to see how search volume relates to certain economic indicators. There’s the “aw, shucks” understatement and some Google-style graphics that are primitive compared to the shiny world of Microsoft’s user experience demos. Don’t be fooled.
Here are my observations:
First, countries issue economic indicators. Government agencies issue economic indicators. Ad companies usually issue proposals. Google is falling into the country and government agency camp, not the ad camp. This is important because I asserted in 2007 that in order to understand Google’s vision, the organization has to be seen as a new type of enterprise, one that operates in many ways like a nation state or at least like Monaco or Luxembourg. (Feel free to laugh and snort. I have come to expect these reactions.)
Second, the notion of an indicator is that only an entity with a super view of data can generate them. When blue chip consulting firms like Booz, Allen & Hamilton undertook SWEC (Study of World Economic Change) in the late 1970s, which I worked on. We had to enlist the support of William Simon (the charming former Secretary of the Treasury) and a number of major (at that time) financial institutions. Today is not 1977, of course, but it is interesting to note that Google is happily using its own data and no blue chip line up of commercial and governmental entities to produce the data., Google is just using its own data. There’s an important point in that self sufficiency. I think it is evident that Google is operating from a super view position. This is another important point in one’s analysis of Google.
Finally, the outputs are canned. Google has robust analytic routines and has disclosed in its public writings analytic tools that would make DE Shaw’s remaining quants and the founder of Financial Engines writhe in calculating excitement. Yes, the Google is * that * good in analytics, probably one of the top financial analytics outfits in the world.
The point is that this new service is one more component of Google Global Bank. I am of the opinion that more Google financial services will wobble out of Mountain View in the next 12 months. A Google credit card with cash back for AdWords purchases? A Google personal finance service that is intuitive and mint flavored? Interesting in my view.
Stephen Arnold, September 3, 2009
Endeca and Sigma Aldrich
September 4, 2009
Update: September 3, 2009, 8 am Eastern: A happy quack to my associate who located “Sigma-Aldrich and Ingenuity Systems Announce Partnership to Create E-Commerce Search Solution Based on Biological Knowledge“. A tie up focuses on search. With Endeca’s congratulations, it appears that Endeca licensed (maybe partnered with) Ingenuity and / or Sigma-Aldrich. Ergo: Endeca is congratulating itself and using the news release as a way to signal “We are in this health care game too.” Stay tuned because I think the “congratulations news release” introduces a new spin on marketing. Pfizer and other phrama companies have been creative in generating sales. Now search vendors may be learning from those pace setters.
Original Post below:
I reviewed a Reuters’ story / news release “Endeca Congratulates Customer Sigma-Aldrich for Award Winning Web Research Tool, Your Favorite Gene.” The story / news release said that “Your Favorite Gene [is], powered by Ingenuity. A powerful biological search tool, Your Favorite Gene provides scientists access to dynamic gene-based content that they can use to model and evaluate prospective experiments in the context of previously published scientific literature.” Ingenuity Systems offers systems and methods that “researchers to model, analyze, and understand complex biological and chemical systems foundational to human health and disease. The Ingenuity products include pathways analysis software and knowledge bases for life scientists and bioinformaticians, and enterprise knowledge management infrastructure, content and services for leading pharmaceutical and biotech companies.” The story had strong pick up, but I could not find a direct connection between Endeca and Ingenuity. I did find a reference to Endeca’s having some connection to Sigma Aldrich. In short, as I chased down the companies mentioned in the story, I found the item somewhat mysterious and different from the routine company news releases I scan each day.
Stephen Arnold, September 3, 2009
Microsoft Fast ESP Architecture
September 4, 2009
Short honk: I was riffling through the Overflight inputs and I came across “EMC ‘Stitching’ Its Stack With Kazeon”. Kazeon is a vendor of eDiscovery systems. What interested me was this statement, attributed to an unnamed software engineer at EMC, so take the comment with a dash hoi sin sauce:
EMC sees e-discovery as a strategic tool allowing it to sharply differentiate itself from Microsoft. (Microsoft acquired erstwhile EMC search partner FAST Search and Transfer in early 2008. A software engineer at EMC told me that FAST has a “fundamentally flawed architecture. Microsoft is welcome to have it.”) Use our stack, EMC is saying, and we will provide you with tools allowing you to find and manage all the information in your organization, not just for litigation purposes, but to help stimulate innovation.
What struck me was the statement “fundamentally flawed architecture.” Harsh words. Could Microsoft have spent $1.2 billion on a company with “fundamentally flawed architecture”? I don’t know the answer, but I put the statement in my “Quotes” folder. When the “new” Fast ESP becomes available, more information will be available. Perhaps some of this information will prove or disprove this statement from Bnet.com’s story.
Stephen Arnold, September 3, 2009
Autonomy and BB&D Tie Up
September 4, 2009
You will want to read the Reuters’ version of this story: “BB&D Group Selects Autonomy as Preferred Partner for Pan-Enterprise Search and High End Information Processing.” On the surface, Autonomy sings a deal with a South African software development company. Africa is the apple of Google’s eye, and Autonomy is quick to recognize opportunities as well. What interested me was the wording of the headline. The idea is that the South African company “selected” Autonomy. Most of these tie ups are a bit like a high school dating ritual. The software vendor needs a partner. The partner needs a product or “big name” to add impact to its sales activities. In the most sophisticated partner match ups, the partner has to pursue the larger company. Google makes some of its partners demonstrate Googlephilia. Microsoft requires love and a dowry. Autonomy’s language shifts the work to BB&D. Autonomy, in a rhetorical sense, adopts the posture of a Google or Microsoft. I find this as interesting as the use of the phrase “pan enterprise and high end information processing” as a synonym for search and content processing.
Stephen Arnold, September 3, 2009
IBM Software License Search
September 4, 2009
I do not know which vendor or software system provides the search plumbing for the IBM Software License Search system I stumbled upon today. I own a couple of ageing NetFinity servers, and I wanted to know if there were license agreements for the pre-loaded software on these once-high value servers. I located a software license “vertical” search service that contains only IBM licenses. This is a mixed blessing because I wanted to skip search from the licenses to the referenced component. I had to do this by opening another browser window and using the general purpose IBM search service which is not too hot in my opinion.
I ran a number of queries for NetFinity servers and discovered four hits from year 2000. I did not recognize any of the software descriptions as what came on or with my servers. My recollection was that I had Lotus Notes, a version of DB2, and several other products including the often flakey ServeRAID software. No joy.
I ran queries for proprietary, confidential, and software. I got zero hits on the word proprietary and zero hits for the word confidential. I ran a query for the word software, assuming I would get a set of results that would give me an idea about how many documents were in the vertical file. I got 193 hits. That seems small to me.
I ran a query for OmniFind and got 30 hits. The most recent was August 2008. This fact suggests to me that OmniFind like Oracle’s SES10g is getting ready for the rendering plant. I selected a hit about OmniFind. I did not get the license document. I got this page:
I made more selections and finally I got an unformatted text dump of the legal text. I found it very difficult to read with my 65 year old eyes. Instead of delivering a readable version of the document, IBM assumed that I could copy the document from the browser window, open Word or Framemaker, and insert the text and format it to my liking. Wrong.
I wrote this review. I can see that IBM has tried to deliver a subset within its almost unworkable full IBM wide search system. But good effort doesn’t earn an A in my search class.
Stephen Arnold, September 4, 2009
More on the Confluence of Wave, Android, and Chrome
September 4, 2009
I wrote a column for KMWorld a couple of months ago. I learned the other day (September 1, 2009) that the column has become one of the most frequently downloaded on the KMWorld.com Web site. I was surprised because Google’s Wave, Android, and Chrome technologies are not new. Chrome is now one year old. When it was rolled out as a beta, Chrome was described as a browser like Firefox or Internet Explorer. Chrome moved from beta to real product quickly and since its début. Now people see Chrome as an operating system, which again is not 100 percent on the money. But the OS tag is more accurate than the browser label some azure chip consultants slapped on the product.
The Android services are also labeled an operating system. In fact, you can find this description in the Wikipedia entry for Android. The description is accurate up to a point. Android, like Chrome, is difficult to pigeon hole. Google itself does not do a good job of explaining what a particular product or service does. The idea is to provide some functionality and allow developers to bang on the technology. With enough people exploring, something will be discovered. At least, I think that is the general theory. Android, on close investigation, is neither fish nor fowl. The word that comes to mind is polymorphic. Once can do quite a bit with the Android suite of technologies. As computing moves to mobile devices, Android blurs into other Google services and relies on others. The hardware / software facets of Android are interesting as well.
Finally, we have Google Wave. Later this year (2009), Google will allow more people to fiddle with the Wave innovation. I don’t buy into the various popular descriptions of Wave. For me, Wave is an early use of some of Google’s dataspace technology. I don’t want to dig into the notion of dataspaces in this short item, but I think it is important that Wave is to a dataspace as a backyard weather station is to NOAA’s atmospheric monitoring systems.
The point of my write up for KMWorld is that the confluence of Wave, Android, and Chrome is now taking place. The impact is likely to be significant across a number of market sectors. The sector which interests me is the enterprise software market. Received wisdom says that established vendors like IBM, Microsoft, and Oracle have the enterprise market nailed. Google lacks the chops to penetrate the most lucrative sources of revenue in this particular market. For Google in the enterprise, it is game over.
I don’t agree. I think that Google will “WAC” the enterprise sector with Wave, Android, and Chrome. The first tap will be gentle. Over time, Google will put more and more muscle behind its “WACs”. The incumbents will have to have the stamina and the resources to withstand the blows and defeat Google.
What’s become increasingly evident to me is an observation I formulated in The Google Legacy, which I wrote in 2004. The study was published in 2005, and it is gratifying to me to see that four years after its publication, many of the research findings I disclosed are finding their way into the tool kits of consulting firms and independent pundits. Here’s the observation:
Assume Google goes away. The legacy of Google is that its disruptive technologies will become the baseline from which other companies start their software and service development.
What I discovered in my research for that 2005 monograph is that Google is important because it represents a learning laboratory for other developers. In short, Google let the genii out of the bottle. The genii cannot be put back in the bottle. As a result, the technologies embodied in the early stages of Wave, Android, and Chrome are inherently disruptive. Incumbents must adapt or find themselves marginalized.
You can read the original column on the KMWorld site.
Stephen Arnold, September 4, 2009
McKinsey and Web 2.0
September 3, 2009
McKinsey’s team of thought leaders has taken the bold step of quantifying the benefits of Web 2.0. You can read the six segments of the article by navigating to the McKinsey Quarterly Web site and registering. The spine of the article is a survey of 1700 executives. The data have been crunched and the principal findings summarized. The article also contains a first for the McKinsey firm—interactive graphics.The idea is to make it more Web 2.0-like to compare data across the 36n months of the study. Blue chip consulting firms charge hefty fees to delivery direct and immediate benefits to their clients. Not surprisingly, the front-and-center data indicate that those in the survey report “greater knowledge and better marketing”. The operative word struck me as “interactive”. Other study findings quantify what many of the Web 2.0 marketers have been touting—better marketing. The tools are those that have been available for many years; for example, videos, blogs, RSS, etc. The pivot point in the essay for me was McKinsey’s comments about the benefits of these now familiar technologies. Blue chip consulting firms go where the money is and that means firms with more than $1.0 billion in sales. McKinsey’s essay states:
Companies with revenues exceeding $1 billion—along with business-to-business organizations—are more likely to report benefits than are smaller companies or consumer companies. Among functions, respondents in information technology, business development, and sales and marketing are more likely to report seeing benefits at various levels than are those in finance or purchasing. IT executives, in general, are more focused on using Web tools to achieve internal improvements, while business development and sales functions often rely on the technologies to deliver better insights into markets or to interact with consumers.
I for one cannot disagree that firms with more resources are going to derive more benefits from new technology than smaller firms. I think the reason is a result of resource availability and the pay off from having more wheels and gears in motion. An investment in a new technology may flop in a resource rich organization and cause nary a ripple. In a smaller firm, a technology miss can impair the firm, often for a considerable period of time.
Another key segment of the article is the “Looking Ahead” section. McKinsey reports that “Web 2.0 use by companies seems to be developing hardy roots.” The statement will adds amps to the individuals and organizations working to infuse Web 2.0 technology into a business process or an entire organization.
In short, McKinsey has documented that certain new technology is making its way into the McKinsey survey sample.
My observations about this report are:
- I may have missed the definition, but I am not sure I know what Web 2.0 means. The article provides a laundry list of technologies which, I assumed, are taken as the Lego blocks of Web 2.0. The problem with this approach to definitions is that each reader forms his or her own definition. Without a clear understanding of what Web 2.0 is and is not, I think that the likelihood of confusion is high. Is, for example, a company with an RSS feed a Web 2.0 company? I think one can hire McKinsey to answer the question, and that’s good for McKinsey, but it does little to clarify Web 2.0 in my opinion.
- The list of benefits is interesting. My question is, “Where are the data that tie the investment to bottom-line pay off?” I did not see this type of hard data. McKinsey clients, like many organizations, worldwide are struggling to generate new revenue. I expected to see at least one concrete case that demonstrates a documented financial pay off. I would have settled for cost reduction (the principal method of hitting numbers in the last quarter or two for publicly traded companies) to new revenue (the top line growth that is sorely needed to start the remediation process for the financial collapse of many organizations). The omission is a serious one in my opinion. I don’t think there are hard data, but the survey offers some hints. But hints don’t create growth in my experience.
- The survey sample is described in a general way. “Executives” is a catch all term. With a sample of older executives (for example, more respondent above the age of 40), the data provide an indication that what has been fueling the growth of Facebook, Google, and Salesforce.com is not part of the daily equipment of doing business. In fact, the lack of uptake for mash ups (Exhibit 2) is downright scary in my opinion. “Mash up” technology – that is, knitting together disparate items of data to provide an answer – is one of the most significant needs in content processing. It ranks slightly above prediction markets as technologies in use. These two items’ low rank suggests to me that the survey sample is well behind the innovation curve.
I urge you to read and save the McKinsey report data. These data will be referenced for months, maybe years to come. McKinsey has done a good job of documenting why the Web 1.0 companies failed to deliver to their stakeholders. I don’t see much encouragement for a turnaround in the firm’s Web 2.0 data.
Stephen Arnold, September 3, 2009
Exclusive Interview with Mindbreeze CEO
September 3, 2009
Mindbreeze (based in Linz, Austria) has opened a US office near Boston. The company has experienced strong growth. The managing director, Daniel Fallmann, is bullish on the future of the company’s search and content processing system. In an exclusive interview for the Search Wizards Speak series, Mr. Fallmann said:
Mindbreeze Enterprise Search is able to handle information in context. Context can be user context, application context, information context, workflow context, or semantic context. This allows us to display a result and enable actions to be taken on that result depending on multiple context dimensions. For example, a Word document result returned from SharePoint can be checked out from a simple click from within the Mindbreeze client.
The company’s product, said Mr. Fallmann, “is the only solution priced either per named user or per concurrent user. This delivers incredible flexibility to our customers in terms of total cost of ownership.”
The system performs content processing functions that identify metadata for each information object processed. In addition, the Mindbreeze system supports retrieval of linked information objects. The example Mr. Fallmann demonstrated pulled patient data, physicians, and medications related to a particular medical issue. The Mindbreeze system ships with supported for advanced reasoning. Mr. Fallmann told Search Wizards Speak:
We can deal with use cases that require advanced reasoning. The use case is searching for an imap4 specification in a vertical solution of Mindbreeze enterprise search which combines it standards and products.
To read the full text of the interview, point your browser to Search Wizards Speak. For more information about Mindbreeze, visit the company’s Web site at http://www.mindbreeze.com. Search Wizards Speak is an exclusive collection of interviews with senior executives in the search and content processing sector. The collection of interviews now spans a wide range of firms and technologies. There are more than 36 interviews available in full text at http://www.arnoldit.com/search-wizards-speak.
Stephen Arnold, September 3, 2009
Open Source, Open Season for Confusion
September 3, 2009
I hear about open systems when I attend a meeting in Washington, DC. I hear about open source initiatives when I talk with Microsoft Certified Gold partners. I hear about open source when 20 something PR folks call me about their company’s new product.
I am confused because I get similar signals plotted on my radar as a way to sell engineering services. I like high end consulting and technical services. I think the notion of downloading software and paying for support is fine. What baffles me is the array of licenses in play for open source.
David Chisnall’s “The Failure of the GPL” provides a clear discussion of the GPL or General Public License. For me the most important passage in the write up was:
The GPL fails to achieve its original objectives, and it also discourages people who don’t want to employ a lawyer to read several pages of legalese and discover whether their intended use is likely to be allowed. It’s easy to infringe accidentally; but if you really want to violate the spirit of the license, then it’s still possible to comply with the letter of the license. With this situation in mind, it’s easy to see why permissively licensed alternatives to GPL’d programs have recently started to see a lot more commercially sponsored development.
The result is the use of the word “open” and a general Wild West business environment. Can open source vendors who follow rules, cultivate and grow the publicly available code, and operate with the interests of their customers firmly in mind thrive amidst the confusion? Some organizations may balk because of the risks only partially understood. I know I am confused when asked to explain these cloudy “risks”. Hopefully the open source community can ameliorate these poorly delineated risks and clear up some of the gnats Mr Chisnall identifies.
Stephen Arnold, September 3, 2009
-
- Subscribe to Beyond Search
Feature archive
News archive
-
