Fascinating. I assume that’s lemonade in the IDC mug.
Satellites or Balloons?
January 19, 2015
I enjoy conversations about how innovative Google is. Prior to 2006, I was on the Google is the leader bandwagon. Now it may be time to shift from the GOOG to the Musk.
I read “Elon Musk touts launch of ‘SpaceX Seattle’.” Tucked in the encomium to Mr. Musk was this passage:
As guests drank beer and wine and sipped Champagne from glasses etched with the SpaceX logo, Musk outlined an audacious plan to build a constellation of some 4,000 geosynchronous satellites, a network in space that could deliver high-speed Internet access anywhere on Earth. Those satellites are to be designed by software and aerospace engineers in SpaceX’s new engineering office in Redmond.
Now satellite communications is not new. Anyone remember Equatorial Communications? History lesson aside, Google wants to deliver the Internet via balloons which rhymes with loons.
According to eBalloon.org, the first hot air balloon was launched in 1783. Satellites came along when I was in grade school.
It seems that Google is not thinking on the Musk scale. Wasn’t X Labs supposed to do moon shots. Mr. Musk now does transportation and digital thingies.
In terms of search, I question whether Google’s innovations in search as described by “How Google Tries To Figure Out Our Hidden Needs” improves relevance or tweaks the PT Barnum formula for revenue.
Stephen E Arnold, January 19, 2015
HP: Serious Fraud Office Served Up Cold Bangers and Beans
January 19, 2015
I assume that the New York Times has this story straight: “British Fraud Office Ends HP-Autonomy Inquiry.” If you have to pay to access the source or it disappears, don’t hassle me.
Here’s the paragraph about Hewlett Packard’s decision to pay big buck billions to purchase Autonomy:
“In respect of some aspects of the allegations, the S.F.O. has concluded that, on the information available to it, there is insufficient evidence for a realistic prospect of conviction,” the British authorities said in a statement on Monday.
HP will pump up in the Palo Alto gym and hit the US courts. But the decision is probably about as appetizing as an expense plate of bangers and beans at a Millwall football match.
HP hopes for a billion dollar burger from the US courts. The only question in my mind is, “Didn’t HP make the decision to buy Autonomy.”
Sir Mike Lynch is a compelling individual, but not even his skills can convince a printer ink company to spend $11 billion for mid 1990s’ technology. Perhaps HP would like to purchase DarkTrace which seems to be shaping up into another winner.
Stephen E Arnold, January 19, 2015
IBM on Skin Care
January 19, 2015
Watson has been going to town in different industries, putting to use its massive artificial brain. It has been working in the medical field interpreting electronic medical record data. According to Open Health News, IBM has used its technology in other medical ways: “IBM Research Scientists Investigate Use Of Cognitive Computing-Based Visual Analytics For Skin Cancer Image Analysis.”
IBM partnered with Memorial Sloan Kettering to use cognitive computing to analyze dermatological images to help doctors identify cancerous states. The goal is to help doctors detect cancer earlier. Skin cancer is the most common type of cancer in the United States, but diagnostics expertise varies. It takes experience to be able to detect cancer, but cognitive computing might take out some of the guess work.
Using cognitive visual capabilities being developed at IBM, computers can be trained to identify specific patterns in images by gaining experience and knowledge through analysis of large collections of educational research data, and performing finely detailed measurements that would otherwise be too large and laborious for a doctor to perform. Such examples of finely detailed measurements include the objective quantification of visual features, such as color distributions, texture patterns, shape, and edge information.”
IBM is already a leader in visual analytics and the new skin cancer project has a 97% sensitivity and 95% specificity rate in preliminary tests. It translates to cognitive computing being accurate.
Could the cognitive computing be applied to identifying other cancer types?
Whitney Grace, January 19, 2015
Sponsored by ArnoldIT.com, developer of Augmentext
Fujitsu Creates its Own Hadoop Tool
January 19, 2015
Fujitsu has joined many other companies by taking Hadoop and creating its own software from it to leverage big data. IT Web Open Source’s article, “Fujitsu Makes It Easy For Customers To Reap The Benefits Of Big Data With PRIMEFLEX For Hadoop” divulges the details about the new software.
The new Hadoop application is part of Fijitsu’s PRIMEFLEX software line of workload specific integrated systems. Its purpose is similar to many other big data software on the market: harness big data and make use of actionable analytics. Fujitsu describes it as a wonder software:
“Fujitsu has developed PRIMEFLEX for Hadoop to simplify and tame big data. The powerful, dedicated all-in-one hardware cluster is designed to integrate with existing hardware infrastructures, introducing distributed parallel processing based on Cloudera Enterprise Hadoop. This is an open-source software framework which gathers, processes and analyses data from various sources, then puts together and presents the big picture on how to act on the information gathered.”
Fijitsu is a recognized and respected brand, but the big data market is saturated with other companies that offer comparable software. Other companies also started with a Hadoop based application as part of their software line-up. Fujitsu is entering the Hadoop analytics a little late.
Whitney Grace, January 19, 2015
Sponsored by ArnoldIT.com, developer of Augmentext
IDC PCWorld Sees Lemonade in Google Glass
January 18, 2015
I am not sure what the folks in Massachusetts are thinking. I read “Google Glass: Down but Not Out.” Reading the story was an interesting exercise in filling a small tumbler with not-so-hot lemonade.
Here’s the passage I noted:
True, Glass has struggled to find its place in the mainstream.
Now that’s a statement that puts Brin and X Labs in some context.
“Struggled.”
Then I highlighted in yellow:
But make no mistake: Glass isn’t going away—not without more of a fight. While it’s struggled to find support among consumers, some businesses have been highly receptive to the electronic eyewear, and the next iterations of Glass might suit them even better.
Chuck full of quotes, the write up points out some of the issues; for example, backlash, the economics, and wonderful phrase “glasshole.”
But what was not said may be more important. The big thinker on this project was the multi-named Babak Amir Parviz. Then there was the alleged interaction between one of Google’s founders and an employee. There was some discord, which is a pretty nice way of summing up some fast dancing in the interpersonal relationships department. Finally there has been the two step in reorganizing the Glass house.
What we have is a write up that ignores the impact of business and personal decisions on a product that warranted a business school type of analysis. That means looking at the company, the professionals involved, the market reaction, and the social implications of hard-to-detect Glass functions.
Like much of IDC’s work, including Dave Schubmehl’s sale of my research under his name on Amazon, the issue of covering a topic is almost more interesting than the high school lab experiments themselves. But who cares? In an age of content marketing and zoom-zoom analysis, there just isn’t time between Oscar antics, analysis, and making sales, is there?
A photo of a Glass fashion show featuring Mr. Brin and Ms. Rosenberg would have helped put the write up in a fashion context. Remarkable that Vanity Fair would have more substance than an IDC publication. Fascinating.
Stephen E Arnold, January 18, 2015
NGIA Palantir Worth Almost As Much As Uber and Xiaomi
January 18, 2015
Short honk: Value is in the eye of the beholder. I am reminded of this each time I see an odd ball automobile sell for six figures on the Barrett Jackson auction.
Navigate to “Palantir Raising More Money After Tagged With $15 Billion Valuation.” Keep in mind that you may have to pay to view the article, or you can check out the apparently free link to source data at http://bit.ly/KKOAw1.
The key point is that Palantir is an NGIA system. Obviously it appears on the surface to have more “value” than Hewlett Packard’s Autonomy or the other content processing companies in the hunt for staggering revenues.
Stephen E Arnold, January 18, 2015
Basho: A Comeback?
January 18, 2015
I read “NoSQL Pioneer Basho Scores $25M to Attempt a Comeback.” In 2012, Basho looked like a player. Then the company lost traction. The all-too-familiar “staff changes” kicked in. Now the company has gobbled another $25 million to the $32 million previously raised. My thought is that generating this much cash from a NoSQL system is going to be a task I would not undertake. I do have a profile of Basho when it was looking like a contender. I will hunt it down and post a version on the Xenky Vendor Profiles page. I will put an item in Beyond Search and provide the link to a free profile of the company in the next few days. Availability of the free report will be in Beyond Search.
Stephen E Arnold, January 18, 2015
Enterprise Search: Is Search Big Data Ready?
January 17, 2015
At lunch on Thursday, January 15, 2015, one of my colleagues called my attention to “10 Hot Big Data Startups to Watch in 2015 from A to Z.” The story is by a professional at a company named Zementis. The story appears in or on a LinkedIn page, and I believe this may be from a person which LinkedIn considers a thought leader.
The reason I perked up when my colleague read the list of 10 companies was two fold. First, the author put his company Zementis on the list. Second, the consulting services firm LucidWorks—which I write in this way LucidWorks (Really?)—turned up.
Straight away, here’s the list of the “hot start ups” I am enjoined to “watch” in 2015. I assume that start up means “a newly established business,” according to Google’s nifty, attribution free definition service. “New” means “not existing before; made, introduced, or discovered recently or now for the first time.” Okay, with the housekeeping out of the way, on to the list:
- Alpine Data Labs, founded in 2010
- Confluent, founded in 2014 by LinkedIn engineers
- Databricks, founded in 2013
- Datameer, founded in 2009
- Hadoop, now 10 years old and originally an open source project and not a company but figure 2004
- Interana, founded in 2014 by former Facebook engineers
- LucidWorks (Really?), né Lucid Imagination, founded in 2007
- Paxata, founded in 2012
- Trifacta, founded in 2012
- Zementis, founded in 2004
Of these 10 companies, the firms that is not a commercial enterprise is Hadoop. Wikipedia suggests that Hadoop is a set of algorithms based on Google’s MapReduce open source version of code the search giant developed prior to 2004.
Okay, now we have nine hot data startups.
I am okay with Confluent and Interana being considered as new. Now we have seven companies that do not strike me as either “hot” or “new”. These non-hot and non-new outfits are Databricks (two years old), Datameer (four years old), LucidWorks Really? (eight years old), Paxata (three years old), and Zementis (11 years old).
I guess I can see that one could describe five of these companies as startups, but I cannot accept the “new” or “hot” moniker without some client names, revenue data, or some sort of factual substantiation.,
Now we have two companies to consider: LucidWorks Really? and Zementis.
LucidWorks Really? is a value added services firm based on Lucene/Solr. The company charges for its home-brew software and consulting and engineering services. According to Wikipedia, Lucene is:
Apache Lucene is a free open source information retrieval software library, originally written in Java by Doug Cutting. It is supported by the Apache Software Foundation and is released under the Apache Software License.
Apache offers this about Solr:
Solr is the popular, blazing-fast, open source enterprise search platform built on Apache Lucene. [Lucene is a trademark of Apache it seems]
As Elasticsearch’s success in combining several open source products as a mechanism for accessing large datasets shows, it is possible to use Lucene as a query tool for information. But, and this is a large but, both the thriving Elasticsearch and LucidWorks Really? are search and retrieval systems. Yep, good old keyword search with some frosting tossed in by various community members and companies repackaging and marketing special builds of what is free software. LucidWorks has been around for eight years. I have trouble perceiving this company and its repositionings as “new”. The Big Data label seems little more than a marketing move as the company struggles to generate revenues.
Now Zementis. Like Recorded Future (funded by the GOOG and In-Q-Tel), Zementis is in the predictive analytics game. The company focuses on “holistic and actionable customer insight across all channels.” I did not include this company in my CyberOSINT study because the company seems to focus on commercial clients like retail stores and financial services. CyberOSINT is an analysis of next generation information access companies primarily serving law enforcement and intelligence entities.
But the deal breaker for me is not the company’s technology. I find it difficult to accept that a company founded 11 years ago is new. Like LucidWorks Really?, the label start up has more to do with the need to find a positioning that allows the company to generate sales and sustainable revenue.
These are essential imperatives. I do not accept the assertions about new, startup, and, to some degree, Big Data.
Furthermore, the inclusion of a project as a startup just adds evidence to support this hypothesis:
The write up is a listicle with little knowledge value. See http://amzn.to/1rUoQyn.
Why am I summarizing this information? The volume of disinformation about companies engaged in next generation information access are making the same marketing mistakes that pushed Delphes, Fast Search & Transfer, Entopia, Fulcrum Technology, iPhrase, and other hype oriented vendors into a corner.
Why not explain what a product does to solve a problem, offer specific case examples, and deal in concrete facts?
I assume that is just too much for the enterprise search and content processing “experts” to achieve in today’s business climate. Wow, what a confused listicle.
Stephen E Arnold, January 17, 2015
Enterprise Search: Evidence It Is a Commodity
January 17, 2015
I was browsing through some information gathered by Overflight last week. I cam across an interesting page showing Libraries Australia Architecture Overview. Here’s a miniature of the diagram. The link provides a larger version. Where is search? Well, it is in the middle, represented by a purple storage icon.
The search system is Solr. I find this interesting for several reasons:
First, Solr replaced the Australian-developed TeraText search system, which I think is pretty good. TeraText was a commercial product, and Solr is an open source system.
Second, Solr is a component in a far larger system. No surprise here, but the diagram makes clear that search is a utility supporting many other library functions. For vendors who make search the fabric for a large-scale application, the Libraries Australia team may want you to give them a lecture about ways to improve their system.
Third, Libraries Australia has a number of systems, each of which presumably has its native search tools. The implication is that Solr provides one screen access to these diverse resources. I wonder if the Oracle DBA uses Solr instead of the native Oracle tools. My thought is that the Solr champions see no reason to fool with Oracle command lines. The DBA, on the other hand, may see information access from a different point of view.
Net net: A commercial account closes, and an open source account begins. Does this fact suggest that closing deals for proprietary search systems might be more difficult in 2015?
Stephen E Arnold, January 17, 2015
Deep Learning Duel: Facebook or Google?
January 16, 2015
Ads mean money.I want to make sure the motivation behind smart software is on the table. I read two PRish items today: “Facebook Open Sources Tools for Bigger, Faster Deep Learning Modules” and (I think) “Google Search Will Be Your Next Brain.”
Both companies are shaking and baking in the “smart software” aisle of the local online store. The Facebook play is “we’re good guys. We do opens source.” The Google angle is “We’re pretty far ahead.”
Net net: This is less about smart software and more about money via advertising from the customer (you and me). Ah, science.
Stephen E Arnold, January 16, 2015
-
- Subscribe to Beyond Search
Feature archive
News archive
-
