AI SLIDE: A Breakthrough or a Shaped Insight
March 4, 2020
DarkCyber noted an interesting, although sketchy summary, of a CPU and hash table approach to machine learning. “Deep Learning Rethink Overcomes Major Obstacle in AI Industry” suggests that Amazon and Google are barking up the wrong artificial intelligence method.
The innovation is the use of hash tables for deep learning. The idea is that one looks up an item, perfect for Intel CPUs. The “old” way relies on matrix mathematics, perfect for nVidia graphics chips. In fact, the solution is a search problem, a point in the write up which may annoy the Googlers; to wit:
“You don’t need to train all the neurons on every case,” Medini [a Rice wizard] said. “We thought, ‘If we only want to pick the neurons that are relevant, then it’s a search problem.’ So, algorithmically, the idea was to use locality-sensitive hashing to get away from matrix multiplication.”
The reason this insight is important is that if it proves useful and can flip the opinions of those innovators with tens of thousands of GPUs generating heat is that machine learning becomes less expensive. (How much does it cost to cool lots of GPUs doing math? Answer: A lot.)
The approach is dubbed SLIDE. The acronym is about as slick as relying on an Intel processor: Sub-Linear Deep Learning Engine. Too bad AMD. You have Linus, the YouTube star, as your cheerleader.
Advantages include:
- Cheaper
- Faster
- More efficient training.
Disadvantages revealed include:
- Memory is needed, lots of memory
- Unexpected cache thrashing (data are here, oops, data are not hear, rinse and repeat)
- Access to Intel engineers reduced the inefficiency by 50 percent, but 50 percent of what? Misses, latency, halts, other?
The point of the announcement is to make clear that Amazon and Google are going about machine learning the wrong way. Does anyone at either firm care? Sure, and it will be fun for the researchers to check out their approach, look up what was investigated in the past, and figure out if it is better to switch than fight.
Net net: Seems interesting and definitely a rah rah for Intel. The write up makes no reference to IBM or other machine learning outfits. Marketing or shaped insight? It is too soon to answer this question definitively.
Stephen E Arnold, March 4, 2020
BA Insight: Interesting Spin for Enterprise Search
March 4, 2020
DarkCyber noted BA Insight’s blog post “Make Federation A Part Of Your Single Pane Of Glass.” What’s interesting in the write up are the assertions about enterprise search. Note that the BA Insight Web site include search along with a number of other terms, including “knowledge,” “seekers,” “connectors”, “smart hub”, and “auto classification.”
Let’s look at the assertions which attracted DarkCyber’s attention.
- “Many have considered enterprise search to be too complex.” Interesting but a number of companies have failed because what people want a search system to deliver is inherently tricky. The Google Search Appliance was “easier” to implement than a local install of Entopia, for example, but the GSA failed because meeting information needs is difficult in many cases.
- Users want a “single pane of glass.” Plus “This improved unified view will dramatically improve the search experience.” The problem remains is that information is not equal. Lawyers have to guard litigation information. Drug researchers have to keep pharma research under wraps. Human resources, what some millennials call “people” jobs have to guard employee health data, salary information, data related to hiring distributions. The “single pane of glass” is an interesting assertion, but federation is more difficult to achieve than some believe… until the services and consulting fees are tallied.
- “And, you go live quickly, instantly adding value (you don’t wait six months for crawling to complete).” The speed with which a customer can go live depends upon a number of factors; for example, dealing with security levels, processing content so that it is “findable” by a user, and latencies which creep into distributed systems. Instantly is an appealing term like new. But instantly?
Several observations:
- BA Insight is a vendor of search and retrieval services for organizations. The company has worked very hard to explain that search is more than search.
- The benefits of the BA Insight approach reads like a checklist of the types of problems which once plagued most enterprise search vendors from Autonomy and Verity. Unfortunately many of these challenges remain today.
- BA Insight has moved from its SharePoint centric approach to a wider range of platforms. T
The marketing is interesting. Data backing the assertions would be helpful.
Stephen E Arnold, March 4, 2020
Quantum Computing Dust Up: Is the Spirit of Jeffrey Influencing Some Academics?
March 2, 2020
If you are into quantum computing and the magic it will deliver… any minute now, you won’t bother reading the MIT Technology Review article “Inside the Race to Build the Best Quantum Computer on Earth.” Please, keep in mind that MIT allegedly accepted funds from the science loving Jeffrey Epstein and then seemed to forget about that money.
Here’s the key sentence in the write up:
None of these devices—or any other quantum computer in the world, except for Google’s Sycamore—has yet shown it can beat a classical machine at anything.
One minor point: MIT’s experts appear to have overlooked China, Israel, and Russia Is it really ignoring quantum computing?), to name three nation states with reasonably competent researchers.
The focus on IBM and Google is understandable. Did DarkCyber mention that IBM is contributing to MIT’s funding; for example, the IBM Watson Lab?
What’s the point of the MIT Magazine research? Let’s try to see if there are quantum-sized clues?
First, Google asserted in 2019 that the fun loving folks in Mountain View had achieved “quantum supremacy.” IBM responded, “Nope.” This write up expands on IBM’s viewpoint; specifically, Google’s quantum magic was meaningless. Okay, maybe from IBM’s point of view, but from Google’s, the announcement was super duper click bait.
Second, IBM is doing research and business development in parallel. Google sells ads; IBM sells … what? Consulting, mainframes, managed facilities, and Watson? Google sells ads. Ads generate money for Google moon shots and quantum PR. IBM spends its money on ads. Okay, that’s a heck of a point.
Third, IBM wants to build a quantum business that does business things. Google wants to build a cloud computer to [a] sell ads, [b] beat Amazon, IBM, and Microsoft in the cloud, [c] accomplish a goal like climbing a mountain, [d] it is just Googley, [e] two of the four choices.
Net net: The write up walks a fine line. On one side is IBM and its checkbook and on the other is the Google. Is the write up objective? From DarkCyber’s point of view, like artificial intelligence, quantum computing is just around the corner.
DarkCyber is checking to make sure that when NewEgg.com offers quantum components, the team can buy one. For now, we will stick with the Ryzen 3900x: It works, is stable, and does jobs without too much fiddling.
Quantum computers require a bit more work. But when deciding between funding and ads, maybe fancy dancing around quantum computing is the tune the MIT band is playing?
Stephen E Arnold, March 2, 2020
Facebook PR: Lean In, Reframe, and Output Word Salad
February 28, 2020
“Facebook’s Sheryl Sandberg Defends Her Company and Her Reputation in Wide-Ranging Interview” is an interesting example of corporate PR, “running for office” preparation, and really heartfelt, super sincere explanations.
A techno journalist of significant stature wrote a book about Facebook. Allegedly the effort involved hundreds of interviews with Facebookers past and present. DarkCyber has not read Facebook: The Inside Story. DarkCyber does not use Facebook to locate “friends”—although one of our deceased dogs has a Facebook page.
This interview with an NBC journalist appears to be a little bit about the book, a little bit about Facebook; for example:
“I wish so much that the world could see the Mark I know,” Sandberg said. “Mark is an enormously, enormously talented guy. He has a great product sense. … People think he doesn’t understand people. That’s just clearly wrong.
Right, clearly wrong. Facebook is suing NSO Group. Facebook is suing a partner OneAudience. Facebook is using legal weapons to demonstrate how Facebook “understands people.” Right. DarkCyber thinks it recalls a bit of an issue with Cambridge Analytica, a zippy researcher, and a whistle blower with a keen fashion sense. But maybe that was a hallucination. People do have them. People also reshape facts into confections of delight.
We noted this statement from the NBC story:
Sandberg also offered her most robust defense to date of Facebook’s business model and its vast collection of personal data, which she said was necessary to offer users a better content and advertising experience. “There is growing concern, which is based on a lack of understanding, that we are using people’s information in a bad way. We are selling it. We are giving it away. We are violating it. None of that’s true. We do not sell data,” she said. “Here’s what we do: We take your information and we show you personalized ads … [to give you] a much better experience.”
Yep, experience.
Lean in. Be sincere. Deliver factoids. Let the lawyers do their work.
Mr. Zuckerberg understands people. Right.
Stephen E Arnold, February 28, 2020
Publishing Desperation 101: Dark and Crafty Methods
February 27, 2020
DarkCyber noted a “dark” post on Ycombinator news. Here’s the zippy part: “I cannot cancel my NYTimes subscription online?” Options: Call or text. Yeah, those customer retention folks are available instantly… you wish.
No surprise. Dead tree publications want to attract and retain subscribers. Remember those magazines with the renewal notices which arrived a month after your first issue? Yeah, same thinking.
But the interesting part of the statement is that it is not as annoying as buying a print subscription and then having to call and obtain the online service as well. There is nothing like halving my support for the dead tree crowd and doubling the costs of “customer support.” Which wizardly outfit uses this method? Why the New York Times.
Navigate to the link above and read the comments. There are more examples of the dark arts of the dead tree brigade; for example:
This is a dark pattern that’s all too common on slightly sketchy small businesses. Slightly disappointed, but not terribly surprised, to see that it’s also common in “mainstream” newspapers.
or
If you actually want the subscription and are slightly stingy, this is also a good way to get a vastly reduced price. The people they make you phone to try get you to stay offer very different deals to what they display on pricing page. Definitely a really dark pattern though, and should be regulated against.
Does the dead tree world care? Not a whit. Desperation causes interesting behaviors.
Stephen E Arnold, February 27, 2020
More PR for Cognitive Search
February 20, 2020
With available data growing faster than traditional search technology seems able to handle, ToolBox predicts, “‘Cognitive Search’ May Be the Sector to Watch.” Writer Santiago Perez considers:
“On an individual level, we have all grappled with the frustrating experience of trying to enter just the right keyword or combination of letters and numbers to get to the exact bit of data we need. But as data multiplies continuously in libraries and archives, a new sort of search with the ability to cut through the chaff is coming into its own. It’s called ‘cognitive search.’ As the term suggests, the ‘thinking’ is deeper than that in a traditional keyword search. It’s leveraged by artificial intelligence and machine learning and gathers insights from signals and behavioral data. The insights can come from activities such as employee visits to web pages, their interactions with each other via chat media or the documents they produce and store.”
Perez cites research (PDF) that indicates between 60 percent and 73 percent of information corporations have gathered is currently unused. However, wonder whether the focus is in the right place here—what is the quality of such data? Where does it originate, how was it gathered, and has anyone verified it? For the vast majority, the answer is “of course not.”
Be that as it may, both Amazon and Microsoft are forging ahead with machine-learning based cognitive search solutions to more thoroughly analyze all that (suspect) data. AWS’s Kendra is currently only available in northern Virginia, Oregon, and Ireland, but they do have a preview available for AWS users. Microsoft is positioning its Project Cortex as the “fourth pillar” of Microsoft Office. See the write-up for more details on each of these products.
Cynthia Murrell, February 20, 2020
LucidWorks: Mom, Do My Three Cs Add Up to an A?
February 19, 2020
Search firm Lucidworks has put out a white paper explaining their new 3 C’s of enterprise search, we learn from the write-up, “Understanding Intention: Using Content, Context, and the Crowd to Build Better Search Applications” from InsideBigData. Registration is required to download and read the paper, but they have also put out a PDF called more simply, “Understanding Intention” that gives us their perspective.
In the 3 Cs section of that document, they note that enterprise search pretty much has content wrapped up. With tools like Hadoop, Solr, and NoSQL, we can now access unstructured as well as structured data. Context means, in part, understanding how different pieces of content relate to each other. It also means analyzing which pieces of information will be relevant to each searcher—and this is the exciting part for Lucidworks. The document explains:
“When a search app knows more about you, it can create a relevant search experience that helps you get personal, actionable search results on a consistent basis. Search apps have solved that problem with signal processing. A signal is any bit of information that tells the app more about who you are. Signals can include your job title, business unit, location, device, and search history, as well as past actions within the search app like clickstream, purchasing behavior, direct reports, upcoming meetings or events, and more.”
Interesting. As for the crowd portion, it has to do with matching searchers with content found by similar entities that have searched before. We’re told:
“When a search app uses the crowd, it goes beyond documents and data, past your specific user profile and relationship, and examines how other users are interacting with the data and information. A search app knows the behavioral information of thousands — sometimes millions — of other users. By keeping track of every user, search apps can bubble up what you will find important and relevant and what other users like you will want, too. The tech uses its knowledge of your office, role, and demographic to match to the same in other users and make intelligent judgments about what will help you the most.”
But how good is the tech, really, at identifying what information one truly needs, and how would we know? Do three Cs add up to an A in search? Not yet, Willy.
Cynthia Murrell, February 19, 2020s
A $600 Desktop Quantum Computer That Breaks Encryption. Wow or Woof?
February 17, 2020
DarkCyber spotted a remarkable claim. A fellow named Dan Gleason, created a portable quantum computer. The idea is that this computing system can hack passwords and maybe cyber security protocols.
The Assertion
The information appeared in an article in BetaNews. “The $600 Quantum Computer That Could Spell the End for Conventional Encryption” reports as actual factual:
Using easily available parts costing just $600…, QUBY runs recently open-sourced quantum algorithms capable of executing within a quantum emulator that can perform cryptographic cracking algorithms. Calculations that would have otherwise taken years on conventional computers are now performed in seconds on QUBY.
Sounds good, almost like a folding mobile phone from Motorola or Samsung, the marketing collateral from an enterprise search vendor like Coveo or LucidWorks, or the breathless assurances of Weaviate. (Dare I say Google or Watson?)
The Team
Greg Morrell, Founder and President, Active Cypher. Formerly president of Amtec Technologies, a management and capital placement limited liability company, and before that a vice president of development at LNR Property Corp. More information about the company appears in an ETS article.
“Dan Gleason is the chief architect and product developer of Active Cyper’s file level security solution. His special skills are in bring elegant solutions to complex problems.” Source: Active Cypher DarkCyber believes that a $600 portable quantum computer is a complex challenge but with many, many problems to solve. Mr. Gleason, according to Active Cypher’s Web site, possesses “special expertise.” This is “in all Microsoft products and programs.” The “all” is interesting.
Caspian Tavallali is the chief operating officer for Active Cypher. He worked in the office of the chairman at the Parman Capital Group. Previously he worked on an MBA at IE Business School in Madrid.
Mike Quinn, Chief Strategy Officer, Active Cypher. Mr. Quinn worked at Citadel Consulting and previously at Microsoft as “Partner” and General Manager of the Enterprise Cyber Security Group. He also worked at Cisco Systems in “services”.
The teams does not appear deeply steeped in the technology of quantum computing in use at Google, IBM, and other firms able to afford the research, demonstrations, and systems.
What’s the business model for the open source infused portable quantum computer? Here’s the answer according to Mr. Gleason:
In response to the threat, Active Cypher has developed advanced dynamic cyphering encryption that is built to be quantum resilient. Gleason explains that, “Our encryption is not based on solving a mathematical problem. It’s based on a very large, random key which is used in creating the obfuscated cyphertext, without any key information within the cyphertext, and is thus impossible to be derived through prime factorization — traditional brute force attempts which use the cyphertext to extract key information from patterns derived from the key material.”
Ah, ha. License the company’s dynamic ciphering encryption!
Additional Information
More detail about the company’s encryption innovations appears in “Maintaining a Zero-Trust Security Model.” That document references quantum in the context of “quantum resilient.” The idea is that the firm’s approach will not be breakable by quantum computer technology directed at decryption or similar functions. There’s no reference to a portable $600 quantum computer. DarkCyber finds this interesting since the white paper was updated in February 2020. (Amazon has a number of patents related to its zero trust systems and methods. Some of these are reviewed in our Amazon Blockchain white paper. You can request a free summary at this link.)
Who is buying into this concept? The write up suggests that Microsoft is curious and attendees at the RSA Conference (if it is held) will be able to check out the device. The algorithms will take more time to analyze unless one has access to Google’s or IBM’s quantum systems.
Observations
A few observations seem to be in order:
- What comprises a quantum computer? Hand crafted hardware from IBM or systems from DWave?
- Are there programming languages for the portable quantum computer?
- How are the “instabilities” associated with quantum demonstrations resolved?
- How was Mr. Gleason able to create a “$600” quantum computer when the cost of Google’s DWave gizmos such down money in seven figure gulps.
Net Net
If true the $600 quantum computer is “real,” Mr. Gleason will be the Marc Zukerberg – Sergey Brin – Steve Jobs of quantum computing. If not true, Mr. Gleason will be well positioned to work as a social media PR expert.
For now, DarkCyber will sit on the quantum fence. Why? The DWave quantum computer costs about $15 million. DarkCyber is not sure if this includes the cost of staff, refrigeration equipment, and maintenance.
But $600. Almost sci-fi made real in the actual factual world.
Stephen E Arnold, February 17, 2020
Why Techno-Babble and Crazy Promises Are Necessary
February 3, 2020
Do you believe the assertions about artificial intelligence, natural language processing, and quantum computing? The question is important because, according to the Nieman Lab, “Humans are hardwire to dismiss facts that don’t fit their worldview.” For those who believe in unicorns and fantasize about unicornification, the wilder and crazier the explanations about technology, the more coherent they sound. But try to provide facts, and the human brain is just not that interested if the research is accurate.
The write up asserts:
In theory, resolving factual disputes should be relatively easy: Just present the evidence of a strong expert consensus. This approach succeeds most of the time when the issue is, say, the atomic weight of hydrogen. But things don’t work that way when the scientific consensus presents a picture that threatens someone’s ideological worldview. In practice, it turns out that one’s political, religious, or ethnic identity quite effectively predicts one’s willingness to accept expertise on any given politicized issue.
What do these references to politicization have to do with technology sales and marketing?
DarkCyber believes that when one points out that an error rate of 85 percent means that there are 15 mistakes per 100 items. People think that error rate is okay, acceptable, maybe great. Apply the error rate to identifying potential bad actors, and someone has to figure out how to explain what happened to the 15 actors put in the bad egg bin.
Present this type of “fact” to a group, and most of the people exposed to the fact will ignore it.
But— and here’s the important point — evoke Star Trek, some magical numerical recipe, or just plain old hocus pocus like Google’s endless yammering about search quality, and people believe this stuff.
Years ago, enterprise search pitch men and pitch women discovered that promising to index “all of an organization’s information” and “eliminating time wasted looking for information” was the key to sales. Explaining that enterprise search was more like crafting a specific search system for a particular and quite specific problem was the more rational approach.
Sales were made, but the users were unhappy. The consequences were dire. Companies failed. Investors lost their money. One search executive was convicted of a criminal offense.
Flash forward to today. Predictive analytics, algorithms, and smart software will improve efficiency, reduce costs, unleash innovation, extract value from dark data, and generate new revenue.
Facts are one thing. Marketing hype another. Guess which takes precedence in search, analytics, artificial intelligence, and quantum computing?
If you said facts, you are in the minority if the Neiman Lab write up is correct.
Stephen E Arnold, February 3, 2020
Google Allegedly Ostracized
January 18, 2020
I worked in the San Francisco area once affectionately known as Plastic Fantastic. My recollection is that most of the people with whom I worked and socialized were flexible. There was the occassional throwback who longed for the rigidity of the Midwestern farm life. But overall, chill was the word. The outfit who paid me to do whatever it was they thought I was my skill was an easy going money machine. Most of the high technology outfits were just starting to get a sense of the power and impact afforded those who were comfortable with online technologies, nifty must have gadgets, and a realization that members of the high school science club could call the shots.
Imagine my surprise when I read the allegedly accurate “San Francisco Pride Members Pass Resolution to Ban Google, YouTube from Future Parades.” The write up states:
Members of the LGBTQ+ organization say they passed an amendment to ban Google, YouTube and Alphabet, as well as the Alameda County Sheriff’s Office, from future celebrations after a vote at their monthly membership meeting Wednesday night. In a statement released to SFGATE on Thursday, SF Pride members and former Google engineers Laurence Berland and Tyler Breisacher said they are now urging the board of directors to formally approve the motion at their upcoming meeting on Feb. 5.
Remarkable if true. The Google HR and marketing departments will have to step up their efforts. Recruitment may become more difficult. The PR vibes are doing the Hopf fibration thing. (This is a nice way of saying, “Difficult to understand.”)
Stephen E Arnold, January 18, 2020
-
- Subscribe to Beyond Search
Feature archive
News archive
-
