Open Source, Opens Up the Enterprise
October 17, 2010
I am seeing open source everywhere. At the gym in Louisville, Kentucky, a second tier wrestler with tattoos sported a T shirt that said, “Hadoop” and featured an elephant on the back. Go figure.
“Open Source Investment to Increase – Survey” provides more fuel for the community’s bonfire under proprietary software. The survey was cranked out by a blue chip consulting outfit (Accenture), so you know I know that those folks never make mistakes. Just like Arthur Andersen maybe?
Here is a passage I noted:
Exactly half of the respondents are fully committed to OSS in their business while just more than a quarter (28pc) said they are experimenting with it and keeping an open mind to using it. Some 65pc of those polled said they have a fully documented strategic approach for using OSS in their business, while another third (32pc) are developing a strategic plan. Of those organizations using open source, a massive 88pc said they will increase their investment in the software this year compared to 2009. The overall volume of open source software development is forecast to rise over the next three years to 27pc, up from 20pc in 2009.
Is it time for commercial software vendors to fill out applications at Wal-Mart and Costco? Not right away. If the data are overstating the update by a third, the implications of the study are tough to miss. Open source is free, and it is open. Quite a few outfits are taking a close look at open source.
I know that if I had a client trying to decide between Microsoft SharePoint and Alfresco, for example, I would probably point the outfit toward Alfresco. After all, some of the lessons of Documentum have influenced thinking at Alfresco I have heard.
Blue chip consultants can make a lot of money analyzing the pros and cons of options. This study may be the first shot in a broader push by a blue chip firm to surf the open source wave.
Stephen E Arnold, October 17, 2010
Freebie
Google: Snort of a Two Humped Camel
October 17, 2010
Years ago, I wrote a for fee column about one of those nightmares of high school math. Each y axis represents a different scale or a combined chart. The curves often illustrate interesting points about data. The x axis was a common base like the horizon viewed from a hiker plodding through the digital sand. The resulting curve had a fancy name, but the graphic representation looked like the humps on a two humped camel. I can almost hear that snort.
That camel is back.
There are other ways to represent data that show a lot of people generating lots of clicks and money and a few people generating an even larger amount of money. I like the camel metaphor because it is easy to visualize and camels can go a long way when their characteristics are understood.
I thought about a half dozen two humped camels trotting across the search desert when I read “Google Sees a Browser in Every TV.” The idea is that TV delivers eyeballs and advertisers will pay to reach these eyeballs. The key is the notion of “every”. Represent the revenue from “every” using just about any measurement one wants and you end up with revenue that just keeps on growing. “Every” says to me most TV viewers. Plot the clicks, mouse hovers, or whatever other data point one collects. Slice and dice the data. Match the interesting centroids to advertisers’ messages, and you get a big computational job. Plot the data and you may see a curve that suggests the outfit with the infrastructure is going to be in the catbird’s seat. No one else can duplicate the eyeballs, the investment in plumbing to do the matching, and so on. This two humped camel is likely to leave competitors with only one part of the camel. That part is not as nifty as the parts that yield the big revenue curves.
Here’s a passage from the write up I found interesting:
Asked about whether GoogleTV would be embraced by non-techies, those who might not be comfortable with a keyboard in the living room, Chandra [Google wizard] said, “Almost everyone is familiar with a keyboard by now, and uses it to watch videos on a computer. We’re just moving the keyboard to a different room and setting, one more conducive to watching TV.”When asked why WebTV and all of the previous attempts to bring the Web and TV together failed and why GoogleTV will be a success, Chandra says that, more than anything else, timing is key. “We are at a tipping point,” he said. Web video has finally matured and there is enough compelling content available on YouTube and other platforms that will drive usage.
Plot the timeline. Google wants lots of eyeballs and lots of clicks. One hump. Google wants the advertisers willing to pay to access the eyeballs chopped into demographically and pychographically tasty nuggets. That’s another hump. When you put the two humps together, Google may get the camel. The competitors get the camel exhaust.
Snort.
Stephen E Arnold, October 17, 2010
Connotate Rolls out Agent Community Version 3
October 16, 2010
Daily Finance from AOL reported that Connotate released the new version of the company’s Web data monitoring and collection system. The new story said:
Agent Community allows non-technical users to capture and analyze dynamic Web content and data via a patented visual abstraction technique and easy-to-use interface. Agent Community V3 includes new features such as increased automated Web data collection capabilities, improved text analytics, “slice-and-dice” Agent subscription and management views, and compatibility with the Mozilla Firefox Web browser.
The AOL news story summarizes the “track and collect” functions of the system. The new version includes tools to make it easier for users to edit and subscribe to Agents, essentially software robots that do the heavy lifting. Connotate’s CEO is reported to have asserted:
Agent Community automates previously manual, tedious processes, allowing users who don’t have programming skills to quickly gather insightful information from millions of websites and sources. Our customers have built tens of thousands of Agents in only a few weeks — and these Agents are resilient to the transformations a website goes through over time, including changes in URL or potential dead links. This enables Connotate to deliver the durability, scalability and reach organizations need to absorb and use information in today’s highly-competitive economic environment.
With competitors such as Fetch Technologies and JackBe, Connotate has some formidable competition. Other companies with findability and content acquisition technology are pushing into the “agent” sector. Should be interesting.
Stephen E Arnold, October 16, 2010
Freebie
Coveo Adds Connectors
October 16, 2010
Coveo has announced new information indexing connectors. Among the new connectors are those for Jive SBS Versions 3.0 to 4.5, support for Microsoft SharePoint 2010, and Microsoft Exchange 2010. Coveo updated its connector for Lotus Notes. In the news release, we learned that Coveo is working with Netezza. Earlier this year we heard that Netezza was hooked into Attivio. Netezza, as you may know, is now part of IBM, a company which has been on a mini-spending spree.
One of the interesting comments in the news story was:
Out of the box, Coveo Information Indexing Connectors seamlessly and securely index enterprise-wide systems and data repositories. Coveo-developed connectors offer superior functionality and integration, including with the native security model of each system. Coveo Connectors feature live monitoring and dynamically index new, deleted and modified documents, ensuring just-in-time access to the timeliest information.
Connectors continue to have a pipeline to our in box. The i2 – Palantir legal matter is about connectors. With the green light turned on for this dust up, connectors are edging from back stage to center stage.
More information about Coveo is available at www.coveo.com.
Stephen E Arnold, October 16, 2010
Freebie
Google and Its Revenue Diversity
October 16, 2010
Don’t get the goose wrong. Most companies would kill for Google’s torrent of cash. There is no bank in Harrod’s Creek large enough to hold a quarter’s profits. Nevertheless I was surprised at the information in “Google’s Long-Range Projects Appear to be Paying Off.” The story explained that Google’s revenue diversification was a result of three types of ad revenue. Er, advertising revenue is advertising. With 96 percent of Google’s $30 billion over the last 12 months coming from advertising, the goose is struggling with the meaning of diversification. To me, diversification means chunky revenues from enterprise licensing, OEM deals, an IBM like solutions business, or a line of toasters. The Google wants to grow. That’s good. After more than a decade, Google is selling ads. That seems to be one stream of revenue.
Stephen E Arnold, October 16, 2010
Freebie
Web Search Seems Interesting Again
October 15, 2010
From the goose pond, Web search has not been too interesting in the last few years. Others see an open market. I see a closed market. Others see many Web search options. I see limited options. The reason? Most people take one of a small number of options as naturally and unconsciously as a breath of air:
- Use a default search engine, which means either Bing or Google, enter a keyword or two, and pick one of the hits
- Navigate to Google, type a letter or two and click on a link
- Click something on almost any page that looks interesting.
Sounds simple right? I think it is simple and the usage data suggests that anything involving Google is good enough for two thirds of those who “search”. The other vendors are pretty much irrelevant in the US. In other countries a similar skewed usage pattern is evident. Some countries’ users rely on local systems, but for the purposes of this blog post, let’s focus on the US market.
In that market, there are, in my opinion, two players–Microsoft Bing and Google.
“Bing = Bing + Facebook” caused me to think that Web search is once again interesting. Here’s the point that hooked me:
Effective immediately, when you search on Bing you can see results based on what your friends and other people are recommending and talking about. So your search results will be different from mine because we know different people.
Eeek. Microsoft and Facebook appear to have — if the information in the write up is spot on — a search feature that Google does not. Google has stellar earnings, but Microsoft has a pipe into Facebook. What makes this more interesting is that Facebook has been hiring Googlers. The combination of chemicals could produce an interesting compound.
When I was in high school chemistry, I found that departing from the lab workbook produced some fascinating results. One example was a mixture of calcium carbide and concentrated hydrochloric acid. A click from the Bunsen burner ingniter thing was particularly memorable.
Search is interesting because the combination of Googlers migrating to Facebook, Microsoft’s Facebook data, and the poorly understood behavior of the Facebook addicts could produce a bang. Of course, the compound may be unpleasant. Google, on the other hand, is missing one of the ingredients: Facebook data. The void can be filled but at this moment, Bing is in a space and may expand.
The uncertainty adds interest to what is for most people a routine and method that gets little thought.
Stephen E Arnold, October 15, 2010
Freebie
x
Search Expert Predicts Search Market Shrinkage
October 15, 2010
Years ago, I pointed out that enterprise search was a gone goose. (No pun intended.) There were many structural reasons, which I pointed out in various monographs, talks, and “real media” articles. The highlights of the obituary includes:
- No big oomph in precision and recall on such tests as the NIST TREC annual dive into the frosty waters of a controlled tests of search systems
- Lots of dissatisfaction among users who bleat about their inability to find a document “that I know is in the system” to having to use three, four or more systems and then crack a search soduku to find a document
- Big, big costs that no matter how many IT wizards and junior wizards the organization hired just kept on doing the French “gonfler” thing
- Stuff that never really worked
- System professionals who got a system up and running and then turned over management, administration, and hubcap washing to consultants
“A Search Market Craters – Welcome to the App Diaspora” saddles up the Pony Express and delivers the news, saying:
Hadley Reynolds, a senior analyst at IDC, said in a presentation last week at Lucene Revolution that the enterprise “OEM” search market is expected to drop from more than $100 million in 2007 to $50 million by 2011. The market has traditionally included companies such as Autonomy and FAST, which was sold to Microsoft and is now integrated with SharePoint. “We see the OEM search market as essentially cratering,” Reynolds said.
Eeek. I track a couple of hundred search and content processing outfits. Many of these are funded by IPO and BMW craving investment types. My hunch is that some of these search and content processing outfits may be shivering in their cubicles.
Is open source the Grim Reaper? In my limited view, nah. Search and content processing are chugging along, with a number of companies doing quite well. Search is being embedded in other enterprise applications. Vertical applications like those from Exalead are selling in the US, Europe, and elsewhere. Big dogs like Autonomy don’t have their ribs and haunches poking from an emaciated framework. In short, open source search is growing, and my hunch is that the total market for “findability” is getting bigger not smaller. Search is not longer called search. Synonyms range from data fusion to discovery to facets and odd coinages that baffle me like information optimization. Still search. Still growing. And those flaws? Those five issues will be around for a long, long time. Just my opinion from the goose pond.
Stephen E Arnold, October 15, 2010
Freebie
Google, China: Uncertain
October 15, 2010
Short honk: What happens when a company takes on a country? The country usually wins. Countries have laws, a police force, an army, and elected officials who pay attention to the country’s sovereign rights. Why bring this up? Google is learning about China via the University of Hard Knocks. No football team, but pretty good political science program.
Navigate to “Google Says Company’s Future in China ‘Very Uncertain’ after Search Engine Shut.” Even if the story is a hoax my hunch is that China will come out on top. Google may have to explain to some stakeholders how the world’s largest market slipped away.
I have nothing to add. Companies and countries can be a volatile combination. The UHK foosball team is pretty tough. I bet China has some killer foosers, however. Okay, refoos!
Stephen E Arnold, October 15, 2010
Freebie
Cognition Grabs Grabbit
October 15, 2010
We learned that Cognition Technologies has landed a deal with Grabbit, the developer of cloud hosted services for social media.
Cognition’s semantic tools will add zing to Grabbit’s new Web services. The scope of Cognition’s Semantic Map is more than double the size of any other computational linguistic dictionary for English, and includes more than ten million semantic connections that are comprised of semantic contexts, meaning representations, taxonomy and word meaning distinctions. The Map encompasses over 540,000 word senses (word and phrase meanings); 75,000 concept classes (or synonym classes of word meanings); 8,000 nodes in the technology’s ontology or classification scheme; and 510,000 word stems (roots of words) for the English language. Cognition’s lexical resources encode a wealth of semantic, morphological and syntactic information about the words contained within documents and their relationships to each other.
According to “Cognition to Power Grabbit’s Online Recommendation Engine”:
By using Cognition’s state-of-the-art semantic technology to analyze Tweets, Facebook updates, blog posts, friends, locations, product purchases, brand preferences, and other data, Grabbit will be able to intelligently recommend products, content, and people that a Grabbit user might find interesting.
For more information about Grabbit, Inc., navigate to Grabbit.net. For more information about Cognition Technologies, click to www.cognition.com.
Stephen E Arnold, October 15, 2010
Freebie
New Virtual Shopping Assistant Opens Doors
October 15, 2010
More about the search-without-search boomlet. With the large number of products available on the market, picking out a simple item can seem as complicated as buying a new car. When it comes to technology such as computers, the fact that they change so often can make it difficult to know exactly what to buy. The company SalesClark wants to help shoppers make those hard decisions with the release of their automated virtual shopping assistant. PR Web article “SalesClark Brings Artificial Intelligence to Online Shopping” describes the program which is designed to make shopping easier and more efficient. The CEO of SalesClark describes the program as “An intelligent shopping assistant saves time and money for busy customers.” A brief customer consultation allows the program to connect shoppers with the products that best meet their needs. Users can enjoy the convenience of finding the products they want quicker and easier. The SalesClark beta version is being released to the public and is geared towards helping shoppers find a new computer. Users can choose the exact features or options they want and get quick and accurate results. Whether you are a beginner making your first computer purchase or a skilled novice this virtual shopping assistant does all the work for you.
April Holmes, October 15, 2010
Freebie
-
- Subscribe to Beyond Search
Feature archive
News archive
-
