Enterprise Search: Parkour for Venture Funded Enterprise Search Vendors
January 3, 2015
Parkour refers to the sport of jumping and climbing on man made constructions. Note that most of these “obstacles” have doors, staircases, and maybe elevators.
There are some terms that make this seemingly crazy activity sound really cool. For example, I learned whilst on vacation about the KONG. This is a suat de chat and involves “diving forward over an obstacle so that the body becomes horizontal, pushing off with the hands and tucking the legs such that the body is brought back to a vertical positio0n, ready to land.” See Parkour Terminology.
I also found this maneuver fascinating:
Kash vault This vault is a combination of two vaults; the cat pass and the dash vault. After pushing off with the hands in a cat pass, the body continues past vertical over the object until the feet are leading the body. The kash vault is then finished by pushing off the object at the end, as in a dash vault.
Here’s an image of a parkour expert doing parkour, of course:
Image source: http://parkourfreerunningblog.com/wp-content/uploads/2011/10/parkour.jpg
Now this looks like something a crazy person does: Jumping off a large concrete structure. Just my opinion, of course.
And, from my point of view, parkour is very similar to selling proprietary enterprise search and content processing solutions to commercial enterprises. The danger comes from having to pay stakeholders for the cash borrowed to keep the enterprise search company afloat. The thrill comes from the knife edge under feet: one error and some serious pain results. I suppose this focuses the mind.
As 2015 gets underway, enterprise search “experts” and vendors are gearing up to make sales. Some of the antics are beneficial to the mid tier consulting firms and publications that list the “visionaries,” the “companies that matter”, and the “leaders.” There are individual experts who conflate search with mastering Big Data or delivering the fuzzy wuzzy notion of information governance. Then there are the search vendors who wrap keyword search and classification in Dollar General wrapping paper. The idea is that keyword search is customer relationship management, analytics, and business intelligence.
For me, this is search vendor parkour, and it is okay for the tiny percentage of the population who want to jump off man-made structures. But for a person with a bit of information retrieval perspective, there are some other ways to get some exercise, remain whole, and not look absolutely crazy to an outside observer.
Here are some enterprise search realities to ponder this weekend:
First, if IBM and HP actually hit their magical billion collar goals for Watson and IDOL, how much money will be left for the hundreds and hundreds of smaller search system vendors. The answer is, “Generating billions from search is not possible, and the money available tends to be a tiny fraction of these behemoths’ projections.”
Second, why would a company pay for a commercial keyword search system when there are perfectly functional open source solutions like Elasticsearch, FLAX, and SphinxSearch?
Third, how can keyword search enriched with some clustering deliver actionable intelligence? There are companies specializing in delivering actionable intelligence. Such firms as BAE and Leidos have robust platforms that collect, analyze, and report automatically. Guessing which words unlock the treasures of an index seems somewhat old fashioned to me.
Fourth, how will the companies pouring millions upon millions into Attivio, BA Insight, Coveo, and a dozen other keyword search companies get their money back? I suppose there is the hope that Google, Microsoft, or Oracle will buy one of these firms. But that looks like a long shot. My view is that paying back the investors is going to be difficult, if not impossible.
Now these statements are sobering. One can immerse oneself in that baloney generated by the mid tier consultants (one of which Dave Schubmehls my research), the silliness generated by content management blogs about findability, and the wonkery of search engine optimization wizards.
The year 2015 will witness some significant shifts in the enterprise search landscape. In my forthcoming CyberOSINT: Next Generation Information Access, I explain the type of systems that are underpinning intelligence systems in the US and EC nations. I point out the specific functionalities of these next generation systems that make search a utility. Think of Mac OSX and its inclusion of Spotlight. Nice to have, for sure, but search is not OSX. My research team and I also identify some important lessons the NGIA vendors are teaching their customers. We also look ahead and identify some research areas that are likely to capture investors’ attention and yield measurable results.
Search is a utility. The fact that some brave people convert it to parkour does not change the fact that the activity itself is risky, entertaining, and useless. If I were an athlete, which I am not, I would focus on sports that generate the big bucks. Hoops. Football. Soccer. Parkour? That looks nuts from my vantage point in Harrod’s Creek.
Why not sell something the customer can see solves a problem? Crazy jumps just call attention to the last gasps of a software sector that needs life support.
Stephen E Arnold, January 3, 2015
If an IDC Tweet Enters the Social Stream, Does Anyone Care? I Do Not.
December 31, 2014
This is a good question. The Twitter messages output by Beyond Search are automated. We know that most of these produce nothing substantive. But what about Tweets by an IDC search expert like Dave Schubmehl. You may recognize the name because he sold a report with my name on it for $3,500 on Amazon without my permission. Nifty. I don’t think of myself as a brand or fame surf board, but it appears that he does.
My Overflight system noted that since September 22, 2014, Mr. Schubmehl or an IDC software script generated 198 tweets if I counted correctly. There were quite a few tweets about BA Insight, a search vendor anchored in Microsoft SharePoint. I ask, “Is BA Insight paying for IDC to promote the brand?” I know that there may have been some brushes with IDC in the past. Whether for free or for fee, Mr. Schubmehl mentions BA Insight a half dozen times.
But Mr. Schubmehl is fascinated with IBM. He generated tweets about Watson, IBM “insights”, and IBM training 149 times. Perhaps IDC and Mr. Schubmehl should apply to be listed in the TopSEOs’ list?
Do McKinsey, Bain, and BCG consultants hammer out tweets about Watson? I suppose if the client pays. Is IDC and search expert Mr. Schubmehl in the pay-to-play business? If not, he has considerable affection for the IBM and its Watson system, which is supposed to be a $10 billion business in four or five years. I wonder how that will work out in a company that is playing poker with its financial guidance for the next fiscal year.
Stephen E Arnold, December 31, 2014
Beyond Search Content Flow
December 22, 2014
To my two or three readers:
We will be reducing the flow of stories from December 18, 2014, to January 1, 2015. Coverage in Beyond Search will be expanded to include the new Cyber OSINT data stream and including content about NGIA (next generation information access). I will be moving the IDC/Schubmehl content to the Xenky.com Web site to make on going references to the reputation surfing easier to reference.
Enjoy the holidays.
Stephen E Arnold, December 22, 2014
Google News in Spain: The Sound of Declining Traffic
December 17, 2014
Well, the Googley conquistadores seem to have caught the attention of the Spanish news sites. I read “External Traffic to Spanish News Sites Plummets after Google Move.” I find it remarkable that “real” journalism outfits fail to understand the power of the GOOG. Axil Springer pumped millions into Qwant. I bet you use that Pertimm-based service each and every day, right? A quick dust up with the Google, and the German publisher rolled over like my clueless boxer Tess. She is deaf, has three good legs, and one eye. But Tess figures stuff out without have to do much more than be aware of her environment. Perhaps there is a lesson there?
Is Tess the rescue boxer smarter than the average European publisher chock full of “real” journalistic wizardry? I can make a good case for Tess. She uses Google to help me research Cyber OSINT and NGIA.
The write up states:
Spanish publishers are now asking for help from the government because of the impact of the law, even though Google warned that it would have to remove their links if the law was passed (any links to Spanish sites are also removed from other content on non-Spanish versions of Google News, but they remain available through a regular Google search).
The reality is that the folks with the wonky logo and teenagers on the payroll are the gatekeepers. If you are not in Google, you do not exist. This applies to cold blooded northern Europeans and the more excitable southern Europeans. Thomas Mann explained this is his novels. Well, some “real” journalists may want to refresh their memories. Reality check: Google has traffic power. Sartre provided some insight in No Exit. I have an idea. Let’s run a modern European literature class for “real” journalists. Yes, students, you can use Google. I excuse from class the wizards at IDG/IDC who suggested that Google pull out of Europe. Europe may request that Google remain available. Look for a report from IDC expert Dave Schubmehl explaining why Google should put its tail between its legs and scurry back to Silicon Valley.
Stephen E Arnold, December 17, 2014
A Fine Idea from IDG: Google, Bail Out of Europe
December 15, 2014
I found this write up presumptuous. Imagine a traditional publishing company and owner of the mid-tier consulting company IDC providing unsolicited suggestions to the Google. (IDC as you may recall employs the “expert” Dave Schubmehl, who has be known to sell work with my name on it via Amazon. Information is here.)
Navigate to this write up: “Why Google Should Leave Europe.” The story suggests that Google should turn its back on Europe. Okay, great idea. The wizards at IDG/IDC do not have a suggestion for replacing Google’s European revenue. The analysis appears to overlook the Android/Chrome business. The stroke of genius ignores the government contracts on which a few, industrious Google labors. Nope. Bail out. Abandon ship. Leave the market to the stellar alternatives like Qwanta, Exalead Search, and my favorite Yandex.
Here’s a passage I noted when I stopping laughing:
Google is more popular among the European public than any other region in the world. The company has higher than 90% market share in Europe simply because users there prefer it over alternatives. (The company has less than 68% market share in the U.S.) So European corporations and the politicians they lobby are out to destroy Google even as the European public loves Google. To summarize, you have government obsessively and shamelessly pushing unfair protectionism under the guise of various righteous bureaucratic causes and hammering away with censorship, fines, threats, bans and constant harassment. Sound familiar? It should. This is the situation found itself in China five years ago.
Now if I am not mistaken, Google’s issues with China are country specific, not a couple of dozen countries with only a subset of “Europe” united, however loosely, by the European Community thing.
The logic of the mid-tier thinkers is that Google should bail out.
I don’t want to spend any time pointing out that the idea has some hurdles to overcome. I would remind you, gentle reader, that Google has stakeholders. Some of these folks are “European.” Nuking the value of the company with the IDG/IDC approach would create what I would describe as pushback.
You can identify two or three other reasons why cutting ties with Europe might not be a great maneuver at this time. Are you familiar with Google’s employees in Europe? What about Google’s operation in Switzerland? Oh, well, details do not trouble an outfit that sells my content without my permission on a digital Wal-Mart.
If Google does follow this advice, I would be mildly surprised. Perhaps IDG and IDG should turn its attention on remediating its contract processes, its reselling of my content without my permission, and coming up with forecasts that are able to put McKinsey, Bain, and Boston Consulting Group to shame.
That, however, is probably of less interest than offering Google unsolicited advice that sails into the digital aether to disappear. Quickly I assume.
Stephen E Arnold, December 15, 2015
Sentient Technologies: AI via Barbados and Hong Kong
December 14, 2014
I noted a blog post in the Wall Street Journal called “Artificial Intelligence Company Sentient Emerges from Stealth.” The company then had an enthusiastic PR person named Peter Lo contact me. I asked for a list of the company’s patents. These are public documents and the law librarian and paralegal who work with me on my research for Cyber OSINT are ever at the ready to provide a list of patent information to me.
Not Sentient, the “just emerging from stealth company.” Here’s what I received in response to a polite, legitimate request for patent numbers:
Hi Stephen,
We’re checking on the patent numbers to see if we can share these publicly and will disclose if we can do so.
Best,
Peter [on behalf of Sentient Technologies, Zenogroup]
I pointed out that patents were in the US as far as I knew information available at USPTO, via Google, and free services that seem to be as plentiful as Microsoft Surface ads on televised basketball games.
Mr. Lo from Zenogroup replied:
Hey Stephen,
Completely understood that patents are public. We’re checking if we can give the exact numbers to share those details specifically. While these patents can be found publicly, the company overall has been careful about revealing the precise patents so as not to tip off other competitors.
Will let you know what we have!
Best,
Peter
Well, Peter replied a couple of days later with this information:
About two days later I received this information from Mr. Lo, who at this point, had become a center of interest for my research team:
Hi Stephen,
I certainly didn’t mean to come off as trying to withhold any information. I’m sorry if I gave that impression. Here’s the patents numbers for the Evolutionary Algorithm (EA) patents that we’ve been granted thus far. You’re also welcome to Google other patents by Babak Hodjat, Sentient Founder and Chief Scientist, to see where we’re building our expertise. Please feel free to also Google the patents by our other founders for more of the team’s collective background.
· 8825560 – Distributed evolutionary algorithm for asset management and trading
· 8527433 – Distributed evolutionary algorithm for asset management and trading
· 8768811 – Class-based distributed evolutionary algorithm for asset management and trading
Please don’t hesitate to ask any questions, and I’m sorry regarding the impression I gave you earlier. Happy to help if you need anything else.
Best,
Peter
So, case closed, right? No.
I asked one of the ArnoldIT goslings to check the the company a bit more closely, particularly the claim made about patents in the PR spam email I received on December 9, 2014:
Sentient now has 15 U.S. patents – 6 issued and 9 pending.
My researcher pointed out that the email from Mr. Lo contained references to three patent documents. The clever ArnoldIT professional told me that we should have received 12 patent application and patent numbers.
Math is a strong suit of this particular researcher.
In poking around, we found that the folks involved with Sentient have or had some connection with Sentient Technologies Holdings Ltd, which in turn, is hooked with Genetic Finance (Barbados) Limited. A fair number of the inventions are related to finance; for example:
Distributed evolutionary algorithm for asset management and trading United States Patent 8825560 B2 · Filed: 05/15/2013 · Published: 09/02/2014
Abstract: The cost of performing sophisticated software-based financial trend and pattern analysis is significantly reduced by distributing the processing power required to carry out the analysis and computational task across a large number of networked individual or cluster of computing nodes. To achieve this, the computational task is divided into a number of sub tasks. Each sub task is then executed on one of a number of processing devices to generate a multitude of solutions. The solutions are subsequently combined to generate a result for the computational task. The individuals controlling the processing devices are compensated for use of their associated processing devices. The algorithms are optionally enabled to evolve over time. Thereafter, one or more of the evolved algorithms is selected in accordance with a predefined condition.
Assignee:
Genetic Finance (Barbados) Limited (Belleville, BB)
Inventors:
Hodjat, Babak (Dublin, CA, US)
Shahrzad, Hormoz (Dublin, CA, US)
Blondeau, Antoine (Hong Kong, CN)
Cheyer, Adam (Oakland, CA, US)
Harrigan, Peter (San Francisco, CA, US)
Others relate to executing algorithms via a network; for example:
DISTRIBUTED NETWORK FOR PERFORMING COMPLEX ALGORITHMS United States Patent Application 20140006316 A1 · Filed: 08/29/2013 · Published: 01/02/2014
Abstract: A server computer and a multitude of client computers form a network computing system that is scalable and adapted to continue to evaluate the performance characteristics of a number of genes generated using a software application running on the client computers. Each client computer continues to periodically receive data associated with the genes stored in its memory. Using this data, the client computers evaluate the performance characteristic of their genes by comparing a solution provided by the gene with the periodically received data associated with that gene. Accordingly, the performance characteristic of each gene may be updated and varied with each periodically received data. The performance characteristic of a gene defines its fitness. The genes may be virtual asset traders that recommend trading options, and the data associated with the genes may be historical trading data.
Assignee:
GENETIC FINANCE (BARBADOS) LIMITED (Belleville, BB)
Inventors:
Hodjat, Babak (Dublin, CA, US)
Shahrzad, Hormoz (Dublin, CA, US)
Blondeau, Antoine (Hong Kong, CN)
Cheyer, Adam (Oakland, CA, US)
Harrigan, Peter (San Francisco, CA, US)
The company has a Web site, it seems, for its financial applications. It looks like this:
The company also has an artificial intelligence centric Web site. It looks like this:
The company, according to Crunchbase here, has raised more than $140 million. Founded in 2007, Crunchbase describes the company this way:
Using advanced Artificial Intelligence technology, massively distributed computing, and a scientific approach to the verification of newly discovered strategies, Genetic Finance delivers novel solutions to complex problems in a wide variety of fields.
Sentient is described this way in an IDC article (without the ministrations of Dave Schubmehl it seems):
Sentient works on scaling artificial intelligence to help companies solve problems: “We take machine learning AI and algorithms – and scale them dramatically so by that we tailor and distribute them around thousands of sites and millions of processors.” Sentient has been around for about six years and has managed to raise a total of $143 million in funding. The founding team actually worked on the technology that became Siri, Apple’s famous voice-controlled virtual assistant. I ask Blondeau if his work on Siri influenced what Sentient are working on now.
Fresh from the research for my forthcoming monograph CyberOSINT, several thoughts wafted through my mind as my researchers bored me with this information:
First, it seems that the “play” for sentient is to repurpose “smart” trading methods for more general purpose applications; hence, the stealth and the second Web site.
Second, the company does not equip Zenogroup to email a person receiving PR spam a list of public documents. Instead the company operates as if it were one of the firms providing specialized services to the law enforcement and intelligence communities. I did a quick check on the vendors known to me to be involved in law enforcement and intelligence, and Sentient rated a hit on the WSJ blog I mentioned but nothing else. Perhaps my files are incomplete? My thought is that pretending to be secret and being secret are two different things. Maybe not?
Third, the Barbados connection fascinates me. Years ago I encountered a financial professional working on a business matter with me. In our conversations, he identified what he called flashing yellow lights. Among those were senior managers who operated away from the primary place of business. Another was having legal incorporation at some interesting places.
Net net: Once again PR backfires for companies trying to cash in with their artificial intelligence technology. I think this company will be interesting to monitor. At the February 2015 CyberOSINT Conference in Washington, DC, I will ask around about Sentient’s technology.
There is a great deal of talk about artificial intelligence. However, human intelligence may be needed when trying to whip up buzz. I suppose the approach works in Barbados, but it does not work in rural Kentucky.
We have added this company to our forthcoming and very public Overflight for CyberOSINT. News about this free service will be available in early 2015 and no PR professional will flog you with stealth baloney.
Stephen E Arnold, December 14, 2014
Machine Intelligence on One Big Poster
December 12, 2014
I need this in my office. I will dump my early 1940s French posters and go for logos.
Navigate to this link: http://bit.ly/1sdmBL0. You will be able to download a copy of an infographic (poster) that summarizes “The Current State of Machine Intelligence.” There are some interesting editorial decisions; for example, the cheery Google logo turns up in deep learning, predictive APIs, automotive, and personal assistant. I quite liked the inclusion of IBM Watson in artificial intelligence—recipes with tamarind and post-video editing game show champion. I found the listing of Palantir as one of the “intelligence tools” outfits. Three observations:
- I am not sure if the landscape captures what machine intelligence is
- The categories, while brightly colored, do not make clear how a core technology can be speech recognition but not part of the “rethinking industries” category
- Shouldn’t Google be in every category?
I am confident that mid tier consultants and reputation surfers like Dave Schubmehl will find the chart a source of inspiration. Does Digital Reasoning actually have a product? The company did not make the cut for the top 60 companies in NGIA systems. Hmmm. Live and learn.
Stephen E Arnold, December 12, 2014
More IDC Cleverness: Innovation Is Alive and Well
December 10, 2014
Short honk: Navigate to this link on the Attivio Web site. I verified this on December 9, 2014. Here’s the link:
http://go.attivio.com/l/5752/2012-12-13/l9m2w/5752/96006/idc_attivio_uia_profile_2012_1_.pdf
What do you see? I see this:
IDC published this information based loosely on my team’s research. There was no written permission take this action. My attorney requested that IDC pay for the rights to use my information, including its resale on Amazon without my permission. As I understand my legal eagle, IDC was to stop selling documents with my name and the name of an IDC expert: Dave Schubmehl.
Well, here we go. After months of fiddling, a report with my name is attached to Attivio.
The only hitch in the git along is that the Attivio described in the IDC report does not match up with the Attivio with which I described in my research reports.
Attivio, instead of struggling to generate sufficient revenue to repay its stakeholders, morphs into a different company.
I care because misrepresenting who wrote what, using another’s work for personal aggrandizement and economic benefit, and trampling over the professionalism of a 70 year old strikes me as uncomfortable.
My suggestion? Think about the source of the information. Figure out who is the expert. Ask yourself, “Do I want to be treated in the IDC manner?”
My answer is, “I want experts to be experts. I want high value information to be fairly presented, not massaged. I want basic business practices observed.”
What’s your answer? We know Mr. Schubmehl’s and IDC’s answer.
Stephen E Arnold, December 10, 2014
Storage, Like Search, a Commodity: Key Vendors Omitted from the Report
December 9, 2014
Nothing spotlights the hungry like a price war. Low prices win. Now what happens to the expensive option? If you are Ferrari and enjoy a cult following of big money car people, you are sort of in business. If you offer expensive online storage, well, that is a good question.
Navigate to “IBM, NetApp Suffer As Storage Buyers Shun Mainstream Suppliers.” The write up points out:
A year ago IBM would have been the third-ranked vendor, but its revenues fell 7.2 per cent to $866m, giving it fourth place. Not so Big Blue is failing to keep up with modern storage technologies and its ageing product set has less and less appeal to customers.
The report from the ever resourceful, expert packed IDC (yep the outfit that sold my information on Amazon without my permission) looks at the world through glasses that give me a headache.
In terms of search, I recall that Coveo was at one time the supplier of search to this outfit. Since IBM bought NetApp, I am not sure what happened to the deal.
Vendors of search hoping for a home run by tagging on to a storage vendors’ wagon train may also be disappointed at the outlook for Big Blue.
Omitted from the mid tier consulting firm’s study were the many low cost storage options that are “good enough.” I, for example, use low cost online storage services and just set up the system to allow each system to copy data from my happy little Drobo. Cheap, multiple copies onsite and off site and none of the crazy pricing that accompanies the folks IDC studies like a Ouija board. I suppose IDC could consult its very own oracle, Dave (surf on Arnold) Schubmehl. Why not?
Stephen E Arnold, December 9, 2014
Attivio Is Now an Oracle Competitor, Not a Search Vendor
December 7, 2014
I read “Oracle Competitor Attivio Promotes Stephen Baker to CEO.” Quite a surprise because Attivio is a search-and-retrieval company with a layer of analytics wrappers. Founded by former Fast Search & Transfer executives, the company ingested more than $30 million in venture funding and now has to generate a return for the stakeholders.
I am not sure if Oracle perceives Attivio as a competitor. MarkLogic, an XML data management vendor, also positioned itself as an Oracle competitor. After hitting a wall at about $60 million and grinding through some new presidents, MarkLogic is keeping a low profile in the markets I track.
Now Attivio may be following this MarkLogic path. Two of the founders of Attivio are moving up. Below Ali Riaz and Sid Probstein is Stephen Baker. Mr. Baker also was a Fast Search & Transfer professional. He worked at RAMP Holdings afar a stint at Reed Elsevier where he was responsible for—wait for it—search.
Attivio co-founder Will Johnson is now the chief technology officer. Mr Johnson is another Fast Search alum. He has worked at GetConnected as—wait for it—a search architect.
My thought is that saying Attivio is a competitor to Oracle is one way to connect semantically with “Oracle.”
But as MarkLogic’s trajectory has demonstrated, there is more to saying a company is “like” Oracle than generating revenue on the scale of Oracle.
Both Attivio and MarkLogic are information access companies. Both want to generate more revenue for their stakeholders. Perhaps a management shift will do the trick.
My view is that if Oracle thought either Attivio or MarkLogic offered a unique, high value service, Oracle would have acquired these companies. Oracle may buy Attivio and MarkLogic. I think the catalyst would be generating and demonstrating rapid revenue growth, expanding margins, and a track record of sustainable revenues. i look forward to a glowing analysis of each firm by IDC’s “expert” Dave Schubmehl in the next month or so. Maybe saying something does make reality change?
Stakeholders want a payback. Management change is a precursor to even more significant activity to benefit those who pumped tens of millions into what may be an old-school approach to information access.
Stephen E Arnold, December 7, 2014
-
- Subscribe to Beyond Search
Feature archive
News archive
-
