Microsoft and the Next Fix Problem
July 11, 2022
I spotted a now routine story about a bug in Microsoft’s software. The story “Windows 11’s ‘Resolved’ Outlook Search Bug Resurfaces: When’s the Next Fix?” reveals a key insight into the software giant’s technical method.
I noted this statement in the article about an issue with search functionality in the Outlook email program, one of the original landscape apps which are pretty much orthogonal to the mobile phone’s display:
When doing a search in Outlook on Windows 11 PCs, the email program sometimes fails to provide results relevant to recent messages…
Yep, search. Microsoft. Not working.
But the important facet of the story appears in the story headline; specifically, “When’s the next fix?”
The Microsoft softies have experienced many issues with search and retrieval. Unlike Elizabeth Barrett Browning, I shall not count the ways. However, I will point out that there is now a fatalism about Microsoft. Stuff goes wrong. Microsoft attempts to fix the problem. Then the problem comes back
Whether it is the outstanding security systems or the brilliance of Word’s fascinating approach to automatic numbering, fixes beget more fixes.
So here we are: Unfixable code, persistent issues, and a giant theme park of opportunities for people to make bad decisions, waste time, and hunt for security flaws.
Yep, next fix. Working11ood. Which time is the charm? Third, fourth, nth? Is there a macro for excellence? Wait, let’s roll that macro thing back.
Stephen E Arnold, July 11, 2022
Intel: The Horse Collar Crowd Bucked Up and Stomped Down
June 28, 2022
Will Intel advertise in PCGamer? Probably not. Not even a former podcaster and the engineering team behind Horse Ridge II second-generation cryogenic control chip can make the allegedly accurate data in “Following Intel’s Graphics Card Misadventures Is Like Watching a Slow Motion Freeway Pile-Up.”
Intel also wants to build super advanced semiconductor plants in Arizona. Water challenged? You bet. With TSMC gunning for two nanometer chips, the Horse Ridge folks need to climb back on the tech pony and catch those stampeding cattle. Oops, I meant customers.
The write up says:
That would be an impressive result if it wasn’t for the fact that the actual game benchmarks shown by the Shenmedoungce on Bilibili(opens in new tab) (via Videocardz(opens in new tab)) didn’t have the Intel GPU behind all three of those rival cards in every test it ran. They’re not strange games either, with League of Legends, GTA V, PUBG, Shadow of the Tomb Raider, Forza Horizon 5, and Red Dead Redemption 2 all given a run out on the Arc GPU. I mean, it can run them all, which is grand. But when your new entry-level graphics card can’t compete with an entry-level GeForce GPU released over three years ago, well, we’ve got a bit of an issue.
A bit of an issue? My take is that them thar engineers inside have been bucked up and stomped down, Pilgrim, am I interpreting this PCGamer write up incorrectly? Oh, is that a hoof mark upside yer noggin’?
Stephen E Arnold, June 28, 2022
US Supercomputer Goes Super Faster
June 17, 2022
Computers used to occupy entire rooms and they could only process a few hundred megabytes of data. Today’s supercomputers still occupy entire rooms, but are more powerful than a few megabytes. PC Magazine has the details about the world’s fastest and most powerful supercomputer: “US Takes Supercomputer Top spot With First True Exascale Machine.” The world’s fastest supercomputer is located in the Department of Energy’s Oak Ridge National Laboratory (ORNL). It is the first exascale machine in history and has an HPL score of 1.102 exaflops/second.
The ORNL supercomputer is the Frontier and uses the Hewlett Packard Enterprise Cray EX platform. It used seventy-four cabinets, each containing an AMD EPYC 64C 2GHz processors and AMD Instinct 250X professional GPU. There are more than 37,000 GPUs and 9,400 CPUs used to power Frontier.
Frontier is a very “smart” machine:
“The huge amount of processing performance achieved equates to 52.23 gigaflops/watt and more than 1 quintillion calculations per second. That’s combined with 700 petabytes of storage and HPE Slingshot high-performance Ethernet for data transfers. In order to cool the system, HPE pumps 6,000 gallons of water through Frontier’s cabinets every minute using four 350-horsepower pumps.’
The previous number one supercomputer in the world was the Fugal system at Japan’s RIKEN Center for Computational Science.
ORNL Director Dr. Thomas Zacharia claimed Frontier will lead a new era of exascale computing and empower new scientific discoveries. ORNL had worked on Frontier for more than a decade with other laboratories, academic institutions, and private businesses. ORNL is in the process of testing and validating Frontier. ORNL plans to progress full science testing in 2023.
Whitney Grace, June 17, 2022
Quantum Baloney Gives Money People Indigestion
June 9, 2022
I won’t mention quantum supremacy. Okay, I did mention quantum supremacy. No, I won’t explain why trivial issues like chaos make assertions about quantum computing less than a slam dunk. I will mention a report with the snappy title “The “World’s Most Powerful Quantum Computer” Is A Hoax With Staged Nikola-Style Photos – An Absurd VC Pump With A Recent Lock-Up Expiration Takes SPAC Abuses To New Extremes.” The document consumes more than 180 pages. The author or authors obviously wanted to explain that there’s a burr under the Wild Rest pony herders’ saddle.
The main idea is that a couple of academics used jargon, nice personalities, and the pixie dust of quantum computing to suck in some investment and deliver digital digital horse manure. Now is the criticism justified? I mean more than 180 pages to make clear that talking about quantum computing is really easy. Demonstrations are only a bit more difficult unless one is an expert in 18th century American buttons. (No, that’s a real thing.)
My reaction to the write up in particular and the quantum computing baloney in general is that some folks have engaged in disinformation.
From the point of view of the authors of the 180 page document, the information seems clear, reasonably well documented, and focused on making life difficult for those who cooked up the “hoax.”
From the point of view of quantum researchers, there may be a different view. What self respective quantum wizards wants to dump on a colleague unless there is a specific payoff in the criticism.
Now here’s the problem: Disinformation.
The quantum computing “discipline” is chock full of claims, reports of breakthroughs, and marketing opportunities. A good example is that one vendor has developed a quantum resistant cryptographic system using plain Jane computers using traditional methods which would be familiar to Grace Hopper.
I can envision a scenario in which the founders of the company drawn and quartered in the cited document can explain what has been accomplished. If a really tough question comes up, the Silicon Valley ploy of apologizing and sending more information may work. Competitors will be able to explain why their approach is a home run. Commercialization is just around the corner. Lawyers will be compensated to try and figure out who is on first and why is I don’t know such a popular reference.
What’s accurate? What’s not accurate?
Welcome to the remarkable world of disinformation with a touch of information weaponization.
Stephen E Arnold, June 9, 2022
Quantum Computer: A Home Version?
June 8, 2022
I read “This Is The World First Operational Room Temperature Quantum Computer.”
Source: Quantum Brilliance, Pawsey, and Wonderful Engineering who presumably have rights to the image of a black box with silver handles and the nifty diamond logo thing.
The idea is that this device uses “implanted nitrogen-vacancy centers in synthetic diamonds (where a nitrogen atom is used in place of a carbon one).”
If you understand that phrase, you are ready to slap this puppy under your desk. Imagine. You could crack any known cryptographic scheme. Keep in mind that some quantum supremacy outfits defeat any attempt to crack quantum cryptographic methods even if those don’t yet exist.
The write up points out:
“We [CEO and engineers of Quantum Brilliance] the look forward to seeing enterprises and researchers utilizing HPC as a hub to explore novel classical-quantum codes using Setonix and the quantum accelerator as a step towards the hybrid computing future…”
Heat, programming tools, apps — coming any day. Next up, a mobile phone form factor. And power draw? What?
Stephen E Arnold, June 8, 2022
Systems and Software: Make Them Really Easy to Use
May 18, 2022
Isn’t software supposed to make work easier and more efficient? Even as reliance on technology in the workplace has increased, it appears to have become just the opposite. TechRadar Pro reports, “Software Frustration Is Costing Workers Millions of Hours Every Week.” Writer Will McCurdy cites a recent survey from Userlane, a company that makes and sells a digital adoption platform. He tells us:
“The company found a third – 35% – of UK employees waste at least one hour per week tackling software-related issues, while 61% spend at least 30 minutes per week on these challenges. The majority – 70% – of employers state that their overall use of technology at work has increased over the past two years according to Userlane’s data, as the demand for online collaboration in particular has skyrocketed with the move to hybrid working. What’s frustrating workers? The fact that software can be time-consuming to use was the most common complaint among those surveyed and was cited by 44% of the survey’s respondents. The IT department not responding to queries or issues quickly enough was another common complaint, cited by 39% of respondents. Software that involves too many complex processes was another common issue, cited by 23% of users. Userlane’s survey also suggests that software challenges are impacting how users approach their jobs.”
For example, nearly half the respondents have put off important tasks because of this frustration, almost 20% have dropped back to manual methods, and 8% have considered quitting over software woes. Apparently, the most common way for companies to battle complaints is to explain the technology’s benefits to workers—an approach we expect some may find patronizing. Other, perhaps wiser, methods include expanding IT support capacity and supplying workers with classroom and/or written training. Userlane also found nearly a third of companies are using a digital adoption platform, like the one it happens to sell. Whether such a guidance platform helps, though, will vary greatly by employee. It is, after all, another layer of software.
Cynthia Murrell, May 18, 2022
Big Tech, Big Winners: Good or Bad
May 17, 2022
Science-fiction and many different types of smart people have informed us that technology and related information is dangerous if unregulated and left in the hands of a few individuals. Engadget focuses on the current reasons why big tech companies are dangerous in the article, “Hitting the Books: US Regulators Are Losing The Fight Against Big Tech.” Meta (formerly Zuckbook), Amazon, Google, and Apple control the technology space and consume…er…purchase startups before they can become a competitor. The government used to regulate the technology marketplace and, according to some written laws, they still do. The current advancement in technology has overwhelmed the government’s capacity to govern it.
Oxford professor Viktor Mayer-Schönberger and author Thomas Range wrote Access Rules: Freeing Data From Big Tech For a Better Future agree that Big Tech companies are hoarding information and there needs to be a more equitable way of accessing it. Biden’s administration has attempted to address Big Tech’s monopolies, but their efforts aren’t effective.
Biden appointed Tim to the National Economic Council as a special assistant to the president for technology and competition policy. Wu favors breaking up Big Tech companies and it was a sign that Biden leaned this way. Another signal of Biden’s leanings was Lina Khan as the Federal Trade Commission chair. Khan favors regulating Big Tech like utilities similar to electricity and AT&T before telecom deregulation. The Big Tech monopolies are not good, because it is preventing future innovation, but politicians are arguing over how to solve a convoluted issue. There are antitrust laws but are they enforceable? The complicated issue is:
“And yet it’s questionable that well-intentioned activist regulators bolstered by broad public support will succeed. The challenge is a combination of the structural and the political. As Lina Khan herself argued, existing antitrust laws are less than useful. Big Tech may not have violated them sufficiently to warrant breaking them up. And other powerful measures, such as declaring them utilities, require legislative action. Given the delicate power balance in Congress and hyper-partisan politics, it’s likely that such bold legislative proposals would not get enough votes to become enacted. The political factions may agree on the problem, but they are far apart on the solution. The left wants an effective remedy, while the right insists on the importance of market forces and worries about antitrust action micromanaging economic activity. That leaves a fairly narrow corridor of acceptable incremental legislative steps, such as “post-acquisition lockups.” This may be politically palatable, but insufficient to achieve real and sustained success.”
The Big Tech people, politicians, and other involved parties are concerned with short-term gains. The long game is being ignored in favor of the present benefits, while the future is left to deteriorate. Europe has better antitrust laws in actions against Big Tech companies. To plan for a better future, the US should copy Europe.
Whitney Grace, May 17, 2022
Why Stuff Is Stupid: Yep, Online Is One Factor
May 16, 2022
I read “IQ Scores Are Falling and Have Been for Decades, New Study Finds.” Once again academic research has verified what anyone asking a young person to make change at a fast food restaurant knows: Ain’t happening.
The article reports:
IQ scores have been steadily falling for the past few decades, and environmental factors are to blame, a new study says. The research suggests that genes aren’t what’s driving the decline in IQ scores…
What, pray tell and back up with allegedly accurate data from numerous sources? I learned:
“The causes in IQ increases over time and now the decline is due to environmental factors,” said Rogeburg [Ole Rogeberg, a senior research fellow at the Ragnar Frisch Center for Economic Research in Norway], who believes the change is not due to genetics. “It’s not that dumb people are having more kids than smart people, to put it crudely. It’s something to do with the environment, because we’re seeing the same differences within families,” he said. These environmental factors could include changes in the education system and media environment, nutrition, reading less and being online more, Rogeberg said.
Ah, ha. Media and online.
Were not these innovations going to super charge learning?
I know how it is working out when I watch a teen struggling to calculate that 57 cents from $1.00 is $5.00 and 43 cents. Yes!
I am not sure to what to make of another research study. “Why Do Those with Higher IQs Live Longer? A New Study Points to Answers” reveals:
“The slight benefit to longevity from higher intelligence seems to increase all the way up the intelligence scale, so that very smart people live longer than smart people, who live longer than averagely intelligent people, and so on.” The researchers … found an association between childhood intelligence and a reduced risk of death from dementia and, on a smaller scale, suicide. Similar results were seen among men and women, except for lower rates of suicide, which had a correlation to higher childhood intelligence among men but not women.
My rule of thumb is not to stand in front of a smart self-driving automobile.
Stephen E Arnold, May 16, 2022
Quantum Computing: Who Sent the Memo? Who Read It?
May 12, 2022
I read “America Is Losing the Quantum Race with China.” Interesting because Google claimed “quantum supremacy” in 2019. NASA helped out the Google, and I wonder if the author of the Newsweek article got the memo. Then, as I recall, an IBM blog took a positive view of Google’s PR play, but in 2021 fired up its marketing system and announced it had achieved quantum supremacy (whatever this term means).
Okay, what’s the story? Is it IBM, Google, or the mysterious and semi-questionable Chinese?
The Newsweek story designed to strike fear into the hearts of those who care about keeping encrypted messaging encrypted learned:
Quantum computing, a form of high-speed calculation at the subatomic level conducted at extraordinarily cold temperatures, will bring computers to speeds barely imaginable today. Atoms, photons and electrons that operate beyond the classical laws of physics and in the realm of “quantum” can be harnessed for extraordinary computing power. Complex problems that once took years to solve could take seconds. And that means everything we know about cybersecurity—every lock secured by current encryption methods—could get blown wide open.
Yikes.
What’s the solution? Teaching American students to make change, avoid the plague of innumeracy as one expert called being sort of stupid, and reading books, not TikTok hashtags?
Here’s the fix:
President Biden’s recent moves will better coordinate our government’s efforts to prevent this nightmare scenario, by bringing federal agencies and critical infrastructure companies together to address quantum threats. It also brings the National Quantum Initiative Advisory Committee under White House control.
Sounds like a plan. However, with US firms already quantumly supreme is a committee necessary? And if the US methods fall short of the mark, there’s always PR. TikTok videos are much more important than wrestling with a differential equation.
Stephen E Arnold, May 12, 2022
Kyndi: Advanced Search Technology with Quanton Methods. Yes, Quonton
April 29, 2022
One of my newsfeeds spit out this story: “Kyndi Unveils the Kyndi Natural Language Search Solution – Enables Enterprises to Discover and Deliver the Most Relevant and Precise Contextual Business Information at Unprecedented Speed.” The Kyndi founders appear to be business oriented, not engineering focused. The use of jargon like natural language understanding, contextual information, artificial intelligence, software robots, explainable artificial intelligence, and others is now almost automatic as if generated by smart software, not people who have struggled to make content processing and information retrieval work for users.
The firm’s Web site does not provide much detail about the technical pl8umbing for the company’s search and retrieval system. I took a quick look at the firm’s patents and noted these. I have added bold face to highlight some of the interesting words in these documents.
- A method using Birkhoff polytopes and Landau numbers. See US11205135 “Quanton [sic] Representation for Emulating Quantum-lie Computation on Classical Processors,” granted December 21, 2021. Inventor: Arun Majumdar, possibly in Alexandria, Virginia.
- A method employing combinatorial hyper maps. See US10985775 “System and Method of Combinatorial Hypermap Based Data Representations and Operations,” Granted April 20, 2021. Inventor: Arun Majumdar, possibly in Alexandria, Virginia. (As a point of interest the document Includes the word bijectively.)
- A method making use of Q-Medoids and Q-Hashing. See US10747740 “Cognitive Memory Graph Indexing, Storage and Retrieval,” granted August 18, 2020. Inventor: Arun Majumdar, possibly in San Mateo, California.
- A method using Semantic Boundary Indices and a variant of the VivoMind* Analogy Engine. See US10387784 “Technical and Semantic Signal Processing in Large, Unstructured Data Fields,” granted August 20, 2019. Inventor: Arun Majumdar, possibly in Alexandria, Virginia. *VivoMind was a company started my Arun Majumdar prior to his relationship with Kyndi.
- A method using rvachev functions and transfinite interpolations. See US10372724 “Relativistic Concept Measuring System for Data Clustering,” granted August 6, 2019. Inventor: Arun Majumdar, possibly in Alexandria, Virginia.
- A method using Clifford algebra. See US10120933 “Weighted Subsymbolic Data Encoding,” granted November 6, 2018. Inventor: Arun Majumdar, possibly in Alexandria, Virginia.
The inventor is not listed on the firm’s Web site. Mr. Majumdar’s contributions are significant. The chief technology officer is Dan Gartung, who is a programmer and entrepreneur. However, there does not seem to be an observable link among the founders, the current CTO, and Mr. Majumdar.
The company will have to work hard to capture mindshare from companies like Algolia (now working to reinvent enterprise search), Mindbreeze, Yext, and X1 (morphing into an eDiscovery system it seems), among others. Kyndi has absorbed more than $20 million plus in venture funding, but a competitor like Lucidworks has captured in the neighborhood of $200 million.
It is worth noting that one facet of the firm’s marketing is to hire the whiz kids from a couple of mid tier consulting firms to explain the firm’s approach to search. It might be a good idea for the analysts from these firms to read the Kyndi patents and determine how the Vivomind methods have been updated and applied to the Kyndi product. A bit of benchmarking might be helpful. For example, my team uses a collection of Google patents and indexes them, runs tests queries, and analyzes the result sets. Almost incomprehensible specialist terminology is one thing, but solid, methodical analysis of a system’s real life performance is another. Precision and recall scores remain helpful, particularly for certain content; for example, pharma research, engineered materials, and nuclear physics.
Stephen E Arnold, April 29, 2022
-
- Subscribe to Beyond Search
Feature archive
News archive
-
