Amazonia: Chopping Digital Trees, November 12, 2018
November 12, 2018
After a few days wandering in the Peruvian mountains, I had a moment of either insight or oxygen deprivation. Amazon can yield Amazonia. No, not jungle insects. Digital information which provide some insight or shape shifting to the company which seems positioned to suck Google’s online revenue like a frisky mosquito.
Thus, we have the first installment of Amazonia:
Alexa Listens and Records
ITEM ONE: Everyone’s favorite surveillance device is in the news. According to a report from WMUR tv:
A judge has ordered Amazon to turn over recordings that might have been captured by an Echo smart speaker in the Farmington house where two women were stabbed to death in January 2017.
The write up points out:
“I think most people probably don’t even realize that Alexa is taking account of what’s going on in your house, in addition to responding to your demands and commands,” said Albert Scherr, a professor at the University of New Hampshire School of Law.
Don’t have an Alexa device? Keep in mind that Amazon’s Alexa virtual assistant now available on Windows 10 PCs as a standalone app. More info is here.
Alexa, are you connected to Sagemaker and DeepLens? Unfamiliar references, gentle reader? Worth tracking down our four part Amazon policeware series. Start here.
Oracle, What Database Will Amazon Use?
ITEM TWO: Amazon Eases Out the Troublesome Oracle
“Keep Talkin’ Larry: Amazon Is Close to Tossing Oracle Software” reveals that Amazon is about ready to undergo its final chemotherapy session. Most traces of the Oracle disease have now been eliminated. Sure, there are lingering side effects like Oracle PR creating inflammation in Amazon, but the end is in sight.
I learned from the real news, real accurate Bloomberg:
An executive with Amazon’s cloud-computing unit hit back at Oracle Executive Chairman Larry Ellison, who ridiculed the internet giant as recently as last month for relying on Oracle databases to track transactions and store information, even though Amazon sells competing software, including Redshift, Aurora and DynamoDB. Amazon’s effort to end its use of Oracle’s products has made new progress, Andy Jassy, the chief executive officer of Amazon Web Services, tweeted Friday. “In latest episode of ‘uh huh, keep talkin’ Larry,’ Amazon’s Consumer business turned off its Oracle data warehouse Nov. 1 and moved to Redshift,” Jassy wrote. By the end of 2018, Amazon will stop using 88 percent of its Oracle databases, including 97 percent of its mission-critical databases, he added.
Time’s are changing for the once dominant database giant.
Amazon: Free PR on a National Scale
ITEM 3: The location of a big Amazon office complex may be known. Surprise, Amazon’s giant PR play called HQ2, the erstwhile competition among cities for a second headquarters, may be over. Where is the online giant and policeware vendor heading? The Washington, DC, area. We learned in “Amazon and Microsoft Are Fighting for a $10 Billion Pentagon Contract — and HQ2 in Virginia Could Be Jeff Bezos’ Boss Move”:
“Let’s just put it this way. I don’t think the timing of Amazon moving its headquarters near D.C. is coincidental,” Daniel Ives, Managing Director of Equity Research at Wedbush Securities, told Business Insider.
Yep, coincidence. But can Amazon win JEDI? Microsoft is trying to prevent the juicy plum from ending up in a Whole Foods shopping basket. But Amazon does have that other government cloud contract, and it seems to deliver what In-Q-Tel could not. Plus, the bitter harvest of the Distributed Common Ground project still lingers in some mess halls.
Stephen E Arnold, November 12, 2018
The Journal Subscription Doomsday Is Upon Us
November 11, 2018
China might have the great firewall when it comes to blocking access, but another infamous digital wall is the great academic paywall a.k.a. subscription paywalls that block access to scientific and academic journals. That could all be changing. Researchers everywhere are shocking and gasping at the thought of have free access to expensive research materials. Could this be the sound of celebration? Nature fills us in on the details in the story, “Radical Open-Access Plan Could Spell End To Journal Subscriptions.”
The celebrating would only take place in Europe, however, as European research funders have banded together for a radical open access initiative that could revolutionize science publishing. The publishers are already angry over this. The European funders are eleven agencies that spend over $8.8 billion annually in research grants. From 2020 and onward, they want all research papers resulting from their grants to be available under a liberal publishing license. The idea is that science should not be kept behind paywalls. The endeavor is called Plan S.
Plan S would change the current publishing model:
“As written, Plan S would bar researchers from publishing in 85% of journals, including influential titles such as Nature and Science. According to a December 2017 analysis, only around 15% of journals publish work immediately as open access (see ‘Publishing models’) — financed by charging per-article fees to authors or their funders, negotiating general open-publishing contracts with funders, or through other means. More than one-third of journals still publish papers behind a paywall, and typically permit online release of free-to-read versions only after a delay of at least six months — in compliance with the policies of influential funders such as the US National Institutes of Health (NIH).
We also noted:
And just less than half have adopted a ‘hybrid’ model of publishing, whereby they make papers immediately free to read for a fee if a scientist wishes, but keep most studies behind paywalls. Under Plan S, however, scientists wouldn’t be allowed to publish in these hybrid journals, except during a “transition period that should be as short as possible,” the preamble says.”
While eleven scientific organizations support Plan S, other European science organizations are still on the fence. These organizations are unsure of how the open access would alter their funding measures and affect their research. The publishers are even more concerned, because it disrupts their entire business model. While they support increasing access to journals, they do not want to get rid of hybrid journals. The publishers think it is better if they all act as one large conglomerate, instead of smaller groups so their goals align. Moving to entirely open access would diminish the quality, peer review, and research of papers.
Plan S would mean the end to subscription paywalls and allow more access to scientific research. The bigger question is who will pay the bill and will research suffer in quality if it becomes “free”?
Whitney Grace, November 11, 2018
Big Data NLP Search Engine
November 10, 2018
Adding natural language processing to big data search engines is not new, but new advances related to the technology are something to watch. Beta News reports that there are, “New Tools Bring Natural Language Search To Big Data.” The opener tells us something we have known for years: that organizations need quick, easy, and accurate search engines and if you do not have them it hinders business. The visual analytics company Arcadia Data has a new business information and analytics search tool in its enterprise suite Arcadia Enterprise.
Arcadia Data describes its new search tool akin to DuckDuckGo, Google, or Bing, except in an enterprise shell. All of the prior listed search engines use natural language processing in their search queries and return search results with quick and decent accuracy. The Arcadia Enterprise search tool responds to natural language questions and responds with visualizations based on size data sets. The Arcadia Enterprise search will also include:
“Features include AI-driven type-ahead and suggestion capabilities that recommend related questions users may be interested in. Arcadia Enterprise also scores questions against all datasets in the system. The best answer is displayed immediately, and a list of other possible answers with lower scores are shown as well. As users click on alternative answers, the system learns that those results are potentially more relevant to the typed question. Users can start with a simple search bar and then as they become more familiar with the system move into a detailed set of advanced BI interfaces to build and deploy data applications.”
Arcadia Data is offering a search tool that will be beneficial in a BI enterprise system and is necessary given the reliance on technology.
Whitney Grace, September 10, 2018
Amazon Rekognition: Great but…
November 9, 2018
I have been following the Amazon response to employee demands to cut off the US government. Put that facial recognition technology on “ice.” The issue is an intriguing one; for example, Rekognition plugs into DeepLens. DeepLens connects with Sagemaker. The construct allows some interesting policeware functions. Ah, you didn’t know that? Some info is available if you view the October 30 and November 6, 2018, DarkCyber. Want more info? Write benkent2020 at yahoo dot com.
How realistic is 99 percent accuracy? Pretty realistic when one has one image and a bounded data set against which to compare a single image of of adequate resolution and sharpness.
What caught my attention was the “real” news in “Amazon Told Employees It Would Continue to Sell Facial Recognition Software to Law Enforcement.” I am less concerned about the sales to the US government. I was drawn to these verbal perception shifters:
- under fire. [Amazon is taking flak from its employees who don’t want Amazon technology used by LE and similar services.]
- track human beings [The assumption is tracking is bad until the bad actor tracked is trying to kidnap your child, then tracking is wonderful. This is the worse type of situational reasoning.]
- send them back into potentially dangerous environments overseas. [Are Central and South America overseas, gentle reader?]
These are hot buttons.
But I circled in pink this phrase:
Rekognition is research proving the system is deeply flawed, both in terms of accuracy and regarding inherent racial bias.
Well, what does one make of the statement that Rekognition is powerful but has fatal flaws?
Want proof that Rekognition is something more closely associated with Big Lots than Amazon Prime? The write up states:
The American Civil Liberties Union tested Rekognition over the summer and found that the system falsely identified 28 members of Congress from a database of 25,000 mug shots. (Amazon pushed back against the ACLU’s findings in its study, with Matt Wood, its general manager of deep learning and AI, saying in a blog post back in July that the data from its test with the Rekognition API was generated with an 80 percent confidence rate, far below the 99 percent confidence rate it recommends for law enforcement matches.)
Yeah, 99 percent confidence. Think about that. Pretty reasonable, right? Unfortunately 99 percent is like believing in the tooth fairy, just in terms of a US government spec or Statement of Work. Reality for the vast majority of policeware systems is in the 75 to 85 percent range. Pretty good in my book because these are achievable accuracy percentages. The 99 percent stuff is window dressing and will be for years to come.
Also, Amazon, the Verge points out, is not going to let folks tinker with the Rekognition system to determine how accurate it really is. I learned:
The company has also declined to participate in a comprehensive study of algorithmic bias run by the National Institute of Standards and Technology that seeks to identify when racial and gender bias may be influencing a facial recognition algorithm’s error rate.
Yep, how about those TREC accuracy reports?
My take on this write up is that Amazon is now in the sites of the “real” journalists.
Perhaps the Verge would like Amazon to pull out of the JEDI procurement?
Great idea for some folks.
Perhaps the Verge will dig into the other components of Rekognition and then plot the improvements in accuracy when certain types of data sets are used in the analysis.
Facial recognition is not the whole cloth. Rekognition is one technology thread which needs a context that moves beyond charged language and accuracy rates which are in line with those of other advanced systems.
Amazon’s strength is not facial recognition. The company has assembled a policeware construct. That’s news.
Stephen E Arnold, November 9, 2018
High School Science Club: Making Traffic Better
November 9, 2018
I read “BlackFly the Autonomous VTOL Backed by Larry Page.” I immediately thought of my high school science club. I also thought about a fellow whom I know. This person built an airplane in his garage. No one will go for a ride in the thing. Experimental is not a magnet for some in Harrod’s Creek. However, one will allegedly be able to beat traffic by buying a BlackFly. Imagine. Those wonderfully courteous drivers piloting their own flying car. What could possibly go wrong? Here’s what the well healed high school science club member will be “flying” to the prom with a beauty content winner from another school or a source provided by doting parents.
Traffic problems in the air? Never. Science club person traveling alone. Yep.
Stephen E Arnold, November 9, 2018
Watson: A Barbie?
November 9, 2018
IBM’s Watson is starting to remind me of Barbie. Why is Watson reminding me of the popular doll?
Watson, like Barbie, has had many interests and careers. She embraces fashion styles and, apparently, so does, Watson.
Watson moonlights as a chef, race car driver, medical worker, sports commentator, fan guide, hotel concierge, professor, and many more. Yes, we know that Watson is a sophisticated machine learning AI, while Barbie is a toy.
However, the similarities are uncanny. Biz Journals shares how Watson is now being used as a tennis coach: “IBM’s Watson, AI Now Involved In Coaching Top Players In Tennis.”
Watson has been deployed in many sports to create the ultimate fan experience, but now players are using them to improve their game. The US Tennis Association (USTA) is using Watson’s advanced AI to watch thousands of tennis videos to create customized reports for pro and junior players. AI is very powerful because it can track metadata, form patterns, and objectively analyze it. IBM and the USTA are both happy with Watson:
“ ‘We are treating video really like a rich data source,’ said Elizabeth O’Brien, program director, IBM Sports & Entertainment Partnerships. ‘How can we actually see all the things that are hidden in video and turn that into data.’
Martin Blackman, the USTA’s general manager of player development, said the data has helped create a new statistic that measures acceleration and deceleration of players. ‘We are able with Watson to look at a player’s acceleration, movement and speed over the course of the match and show them how their court position improves when they are moving at an optimal level,’ he said.”
Before Watson, human taggers used to manually watch footage and tag important moves. The entire process took two hours, while Watson can digest the footage in two minutes. The USTA can deliver instant feedback to players.
Watson’s uses appear to be endless and can save hundreds of human work hours. Now that time can be used on something else, such as improving your backhand.
I had a Barbie when I was much younger. That Barbie wore a tennis outfit and had a tiny racket. Racket?
Whitney Grace, November 9, 2018
Google AI: One Model to Learn Everything. Yep, Everything
November 8, 2018
What’s an online advertising company with interesting ethical norms doing with artificial intelligence? Part of the answer appears in the ad stuffed The Stanford Daily’s story “Jeff Dean Discusses Google’s Current Efforts around AI.”
I noted this interesting point:
The future, he said, lies not in creating lots of models and algorithms for distinct tasks but rather in having one model that can learn everything.
Makes sense. A monopoly on information. Now about the ethics part from a company with a founder who dallied with a Glass wearer, a lawyer who sired a Googler at Google, and a Google VP who ended up dead on a yacht from a needle mishap involving a contract worker and a controlled substance.
Will the magic algorithm operate without bias or unusual tendencies?
Yeah, one model to learn everything. Universal. Perfect right?
Sounds good. But Google is not your mother’s AltaVista, is it?
Stephen E Arnold, November 8, 2018
References for those who do not link these ideas:
The Google Glass affair: https://www.bizjournals.com/sanfrancisco/blog/2013/08/sergey-brin-and-susan-wojcicki-split.html
The productive Google lawyer reference: https://www.theatlantic.com/technology/archive/2018/10/your-move-google-board/574036/
The dead Googler and alleged drug explorer: https://www.forbes.com/sites/ellenhuet/2014/07/09/google-executive-yacht-overdose/#ae67e3e9e255
MIT: IBM a Go To Player in AI School
November 8, 2018
I found this item from AI Dreams quite interesting. I learned:
“As MIT’s partner in shaping the future of AI, IBM is excited by this new initiative,” says Ginni Rometty IBM chairman, president, and CEO. “The establishment of the MIT Schwarzman College of Computing is an unprecedented investment in the promise of this technology. It will build powerfully on the pioneering research taking place through the MIT-IBM Watson AI Lab. Together, we will continue to unlock the massive potential of AI and explore its ethical and economic impacts on society.”
MIT = Big Blue?
Stephen E Arnold, November 8, 2018
Google: The Right Stuff?
November 8, 2018
Much has been written about Google’s 20th birthday, recently. While a lot of fun retrospectives have taken us down memory lane, it’s worth thinking for a while about what it aims to do in the future and whether it can survive. One recent story gave us much food for thought, according to The Conversation, “Google Hits 20, But It Will Struggle To Become a Billion Dollar Company Like Apple.”
According to the story:
“Amid the success of Gmail, Google Maps, Google Docs, Google Chrome, Google Drive, Google Translate, and Google Analytics, one thing has never changed. Since its founding, Google’s mainstay has always been advertising revenue. Of Google’s US$60 billion revenue in 2015, only US$8 billion came from non-advertising activities.”
This is not to mention all of the famously failed products, like Google Glass, that have diverted the company’s time and money. But it’s those beautiful losers that kind of make us root for Google. We love their interest in trying new things, even when we laugh at it. However, some enemies of the Search King are actually claiming they are too powerful. With looming lawsuits at home and abroad, it might be time for Google Lawyer to become a thing if they want to survive another two decades.
Patrick Roland, November 8, 2018
Does Search Mean Bias?
November 7, 2018
As CEOs from Facebook, Twitter, and the like get paraded before Washington, one company has been suspiciously absent: Google. The search giant is in a tough spot and much of it stems from how uneven trust is in government for its product. We learned more in a recent Axios story, “Exclusive Poll: Big GOP Majority Fears Bias in Search Engines.”
According to the story:
“The survey shows that tech companies will have a hard time convincing the public that their algorithms aren’t built to favor any point of view, regardless of the reality. The distrust is driven largely by the right, but a significant minority of independents believe the results are biased toward the left, too.”
This lack of clarity and trust in Washington might seem laughable on the surface. But this epidemic of “fake news” could have real business implications for search and social tech companies. For example, the FCC is now looking into greater oversight into all three. If Republican lawmakers had trust in these institutions, you can bet they wouldn’t be cracking down as hard. Clearly, these companies have some PR work to do, and fast. Otherwise, they might be drowning in new regulations.
Patrick Roland, November 7, 2018
-
- Subscribe to Beyond Search
Feature archive
News archive
-
