Serendipity35 - Research

Why Space-Based Solar Power Sounds Like Science Fiction

ronkowitz@gmail.com (Kenneth Ronkowitz) — Mon, 25 May 2026 09:55:00 +0000

I wrote last week about plans to harvest solar power from space for places like data centers. If it sounds like science fiction, that might be because it was first imagined in a 1941 short story, "Reason," by Isaac Asimov. (see below)

It was formally proposed by engineer Peter Glasser in 1968, a space pioneer who introduced the idea of using satellites to beam solar energy from space down to Earth. Over the decades, what Glaser envisioned has been known by many names — space-based solar power (SBSP), solar-power satellites or satellite power system (SPS), as well as satellite solar-power system (SSPS). Glaser's contributions to space science and technology were not limited to the solar-power satellite concept. He also worked on NASA's Apollo moon missions and headed an experiment that flew aboard the space shuttle Columbia in 1986.

But solar power from space beamed to Earth has remained mostly theoretical due to cost and complexity.

Moving Closer to Superintelligence

ronkowitz@gmail.com (Kenneth Ronkowitz) — Tue, 05 May 2026 10:31:00 +0000

It is difficult to keep up with AI advances and new tools. Recently, I have seen the term "superintelligence" being used and I had to look for a definition.

In AI terms, there are three kinds of intelligence. "Artificial Narrow Intelligence" is what we have now. It is "superhuman" at specific tasks like playing Go or translating languages. ChatGPT, Gemini, CoPilot and Meta AI, et al fit in there at the moment.

"Artificial General Intelligence (AGI)" is human-level across the board and can learn anything a person can learn. We’re not quite there yet as of May 2026.

"Artificial Superintelligence (ASI)" is far beyond human level. Philosopher Nick Bostrom popularized the term: and defined it as “any intellect that greatly exceeds the cognitive performance of humans in virtually all domains of interest.”

ASI is what people worry about — or get excited about — when talking about advanced AI. But AGI isn't quite the same as superintelligence. With AGI, you clone the best human brain in software, but with superintelligence that clone keeps upgrading itself until it’s as far beyond us.

Two new tools are moving closer to the next level.

Google has released TurboQuant, a new compression method that makes AI models cheaper to run and faster to respond. In Google’s reported tests, it reduced the key-value cache, the model’s short-term working memory while it responds, by at least 6x and improved performance by up to 8x on H100 chips, Nvidia’s high-end AI processors used in data centres, while keeping benchmark performance, or standard test performance, close to the original model. That is a serious technical result with a clear business consequence: one of the biggest cost pressures in modern AI may begin to ease. For the past two years, the default logic has been simple. The best AI stayed in the cloud because that is where companies could absorb the cost of running it. TurboQuant starts to weaken that logic.

Meta TRIBE v2 is a foundation AI model that acts like a “digital twin” of the human brain. In plain terms, it’s an AI trained on real brain scan data so it can predict how a person’s brain will respond to things they see, hear, or read. It takes in video, audio, and text, then maps that to about 70,000 areas of the brain to simulate neural activity. Meta itself says that you can think of it as Meta teaching an AI to “think” more as humans do, by learning directly from brain responses instead of just internet text.

Where did I get information anout Meta's products and path? From their own Muse Spark. That is Meta’s latest (well, as of today) AI assistant model.

Marian Croak: A Force Behind Modern Communication

ronkowitz@gmail.com (Kenneth Ronkowitz) — Tue, 18 Nov 2025 10:00:00 +0000

Marian Croak, a name that may not be familiar to many, has had a profound impact on the way we communicate today. As a renowned American engineer, Croak has spent her career pushing the boundaries of technology, particularly in the realm of Voice over Internet Protocol (VoIP). With over 200 patents to her name, Croak's work has enabled seamless communication over the internet, revolutionizing the way we connect.

Her U.S. Patent No. 7,599,359 for VoIP (Voice over Internet Protocol) Technology was ultimately used to create applications such as Zoom, WhatsApp and many others.

Born on May 14, 1955, in New York City, Croak's interest in technology was sparked by her father, who built her a chemistry set that led to her early exploration of the sciences. She pursued her passion for problem-solving at Princeton University, where she earned her undergraduate degree in 1977. Later, she received a PhD in Social Psychology and Quantitative Analysis from the University of Southern California.

Croak's career spans three decades at Bell Labs and AT&T, where she worked on digital messaging applications and VoIP technologies. Her team convinced AT&T to adopt the TCP/IP protocol, which allowed for standardized communication over the Internet. Croak's work on VoIP enabled the conversion of voice data into digital signals, making it possible to transmit voice, text, and video over the internet.

Another of Croak's notable achievements is her patent for text-based donations to charity. Developed in response to Hurricane Katrina, this technology allowed users to donate to organizations using text messaging. The technology was widely used after the 2010 Haiti earthquake, raising over $43 million for relief organizations. Croak received the 2013 Thomas Edison Patent Award for this innovation.

Croak's contributions extend beyond her technical expertise. As a leader at AT&T, she managed over 2,000 engineers and computer scientists, overseeing programs that impacted millions of customers. In 2014, she joined Google as Vice President of Engineering, focusing on expanding internet access and developing Responsible AI.

Throughout her career, Croak has received numerous accolades for her work. She was inducted into the Women in Technology International Hall of Fame in 2016 and the National Inventors Hall of Fame in 2022, becoming one of the first two Black women to receive this honor. She has also been inducted into the National Academy of Engineering and the American Academy of Arts and Sciences.

As Croak herself notes, "Inventors are usually people like you. Sometimes they're good at certain things, other times they're not, and that's ok. Just focus on what you want to change, and you become that change and can make that change happen."

Her legacy serves as a testament to the power of innovation and the impact one person can have on the world. As we continue to navigate the complexities of modern communication, we owe a debt of gratitude to pioneers like Marian Croak, who have worked tirelessly to bring people closer together.

Are We Any Closer To Quantum Computing?

ronkowitz@gmail.com (Kenneth Ronkowitz) — Fri, 18 Apr 2025 08:08:00 +0000

In 1981, American physicist and Nobel Laureate Richard Feynman gave a lecture at the Massachusetts Institute of Technology (MIT) in which he outlined a revolutionary idea. Feynman suggested that the strange physics of quantum mechanics could be used to perform calculations. Quantum computing was born. The illustration here shows what one might have imagined it to be back in 1981 - a lind of science-fiction computer.

Quantum computing is a revolutionary area of computing that uses the principles of quantum mechanics to process information in fundamentally different ways than classical computers. In classical computing, information is processed using bits, which are binary and can represent either a 0 or a 1. In quantum computing, however, the fundamental units of information are called qubits. Qubits can exist in a state of 0, 1, or both simultaneously, thanks to a quantum property called superposition. This allows quantum computers to perform multiple calculations at once.

I am not a physicist or computer engineer, so I don't want to go too deeply into that realm. Reading about this, I see the word "entanglement" and have some memory of Einstein referring to quantum entanglement as "spooky action at a distance." He was skeptical since it seemed to defy the principles of classical physics and his theory of relativity. Einstein doubted entanglement, but modern experiments have confirmed its existence and shown that it is a fundamental aspect of quantum mechanics. In quantum computing, entanglement creates strong correlations between qubits, even when they are far apart.

Entanglement enables quantum computers to solve certain types of complex problems much faster than classical computers by leveraging these interconnected qubits. Quantum computers are particularly well-suited to tasks involving massive datasets, optimization problems, simulations, and cryptography. However, they are still in their early stages of development and face challenges such as error rates, stability, and scalability.

In the same way that AI is already in your daily life - even if you don't notice or acknowledge it - quantum computing could be used in everyday activities. It could revolutionize drug discovery and personalized medicine by simulating molecular interactions at an unprecedented speed, leading to faster development of cures and treatments. By solving complex optimization and learning problems, quantum computers could significantly enhance AI's capabilities, leading to smarter assistants and systems.

Cryptography and cybersecurity's current encryption methods could be broken by quantum computers, but they could also enable quantum-safe encryption, making online transactions and communications more secure. There's good and bad in almost every discovery.

In logistics, smarter traffic systems to more efficient delivery routes, quantum computing could optimize logistics, reducing fuel consumption, travel times, and costs.

And quantum computing could impact improved energy solutions, financial modeling, material design, and many things we haven't even considered yet.

Of course, there are challenges. Qubits are highly sensitive to their environment. Even minor disturbances like temperature fluctuations, vibrations, or electromagnetic interference can cause qubits to lose their quantum state—a phenomenon called decoherence. Maintaining stability long enough to perform calculations is a key challenge. Many quantum computers require extremely low temperatures (close to absolute zero) to operate, as qubits need highly controlled environments. Building and maintaining these cryogenic systems is both expensive and challenging.

Small-scale quantum computers exist, but scaling up to thousands or millions of qubits is a monumental task and requires massive infrastructure, advanced error correction mechanisms, and custom hardware, making them cost-prohibitive for widespread adoption.

On the education side of this, quantum computing sits at the intersection of physics, engineering, computer science, and more. A lack of cross-disciplinary expertise will slow down progress in this field.

Linking to the Wayback Machine

ronkowitz@gmail.com (Kenneth Ronkowitz) — Mon, 03 Feb 2025 10:22:00 +0000

Google Search has integrated a feature that links directly to the Wayback Machine, allowing users to access archived versions of webpages through search results.

The Wayback Machine is an online archive created by the Internet Archive, a non-profit organization. It allows users to access and view historical snapshots of web pages, dating back to the late 1990s. Essentially, it's like a digital time machine that lets you see how websites looked in the past. This can be useful for research, preserving digital history, or just satisfying curiosity.

By clicking the three dots next to a search result and selecting "More About This Page," users can view how a webpage appeared at different points in time. The collaboration enhances public access to web history, ensuring that digital records remain available for future generations.

Source https://blog.archive.org/2024/09/11/new-feature-alert-access-archived-webpages-directly-through-google-search/

Social Media Attribution

ronkowitz@gmail.com (Kenneth Ronkowitz) — Tue, 15 Oct 2024 10:39:00 +0000

When I first started consulting on social media in 2005, I was introducing blogs, wikis, podcasts and the newly -emerging social networks such as Facebook. Both with my academic colleagues and with clients, one of the persistent questions was "How do I know I'm getting any benefit from these social tools?"

Seeing the impact of your social marketing relies on attribution, which is similar to the older metric of ROI (return on investment). Both are sometimes difficult to quantify.

As someone who taught writing for many years, when I first heard the term attribution I thought of giving credit to the original source of information, ideas, images, or language used in a piece of writing. Attribution in writing is important because it shows respect for the work of others, helps to prevent plagiarism and those sources often provide additional information. (see my attribution at the end of this post)

That ROI (return on investment) is a much older dollars-and-cents measurement used well before the Internet and social media For example, you invested $1000 for an advertisement and it produced $5000 in sales. (Some might call that ROAS - Return on Ad Spend - but I'm being simpler here.) Or perhaps, you spent a $1000 on an ad and saw no increase in sales.

Attribution in the social media sense assigns value to the channels that drive an outcome. That might mean dollars but it coukd also be a measurement of a purchase, web visit, download, or subscribing to the site or a newsletter.

It is a bit of reverse engineering or backward design in that you are looking at the effect and trying to determine the cause.

My own tracking of the referring sites for posts on this site allows me to see if traffic to a post came from LinkedIn, Facebook, Twitter, one of my blogs or just a search engine. When someone finds me via Google, I can see what search terms they used. Those results can be surprising. I might get a surge of traffic from a search that found the mention of "Erik Satie" or "flat web design" or "social media attribution."

I have little control about search engine attributions, but I can control what I post on social media and how I word the posts.

Attribution is generally broken down as being in three modes:
Last-touch,
First-touch
Multi-touch attribution.
(Take a look at this diagram from digitalthought.me about more on multi-touch models called Even, Time Decay, Weighted, Algorithmic, etc.)

The first-touch attribution credits the first marketing touchpoint. For example, you run an ad and monitor how many contacts came from that ad.

Last-touch attribution credits the channel that a lead went through just before converting. Maybe you ran an ad on Facebook which someone later tweeted and the lead came from the Tweet that linked to your site for a purchase, so Twitter gets the attribution.

Last-touch is easier to measure, but both single-touch models fail to show the complete and sometimes circuitous customer journey. That's why multi-touch attribution is used. This gets much more complicated and more difficult to track. More complicated than the scope of this post. But as an example, the time decay attribution gives more weight to touchpoints closer to the final conversion event. If your original ad is the starting point but the final purchase came after a tweet that was retweeted and then posted as a link in someone's blog a week later, the blog gets more credit (as a personal endorsement) than the ad although obviously none of this would have happened without the ad.

Back to that question I started getting in 2005. It is important to remind clients that social media used for marketing and as engagement and brand-building may not always generate leads or sales directly but rather indirectly. Getting visitors to your site alone is a kind of success. It may not lead to sales (ROI) immediately, but it increases awareness of your brand for the future.

I will crosspost this on my business blog, Ronkowitz LLC, and measure which post gets the best results.

Attribution is more complicated than this primer, so you might want to check out these sources:

Applying Technology Laws

ronkowitz@gmail.com (Kenneth Ronkowitz) — Tue, 23 Jan 2024 08:08:00 +0000

Huang's Law and Moore's Law are technology "laws." Maybe it is more accurate to say they are observations, but "law" has become attached to these observations since they appear to remain true.

Moore's law is the observation that the number of transistors in an integrated circuit (IC) doubles about every two years. Moore's law is an observation and projection of a historical trend. Rather than a law of physics, it is an empirical relationship linked to gains from experience in production.

Gordon Moore, the co-founder of Fairchild Semiconductor and Intel (and former CEO of the latter), posited in 1965 posited the idea and projected this rate of growth would continue for at least another decade. In 1975, looking forward to the next decade, he revised the forecast to doubling every two years. His prediction has held since 1975 and has since become known as a "law".

Moore's prediction has been used in the semiconductor industry to guide long-term planning and to set targets for research and development, thus functioning to some extent as a self-fulfilling prophecy.

Huang’s Law has been called the new Moore’s Law. It seems that the law that the same dollar buys twice the computing power every 18 months is no longer true.

Huang's law is an observation in computer science and engineering that advancements in graphics processing units (GPUs) are growing at a rate much faster than with traditional central processing units (CPUs). The observation is in contrast to Moore's law as Huang's law states that the performance of GPUs will more than double every two years.

Jensen Huang was then CEO of Nvidia and at the 2018 GPU Technology Conference and observed that Nvidia’s GPUs were "25 times faster than five years ago" whereas Moore's law would have expected only a ten-fold increase. As microchip components became smaller, it became harder for chip advancement to meet the speed of Moore's law.

Huang's Law and Moore's Law are concepts primarily associated with the semiconductor industry and technology advancements. However, their principles can be extended and applied to various domains beyond technology.

You can extend Huang's Law to other fields where exponential growth or improvement is observed. For example, consider advancements in renewable energy efficiency, healthcare outcomes, or educational achievements. The idea is to identify areas where progress follows an exponential curve and apply the principles accordingly.

Both laws highlight the concept of scaling - either in computational power (Moore's Law) or AI efficiency (Huang's Law). You could apply this principle to other systems and processes where scaling can lead to significant improvements.

I am imagining a discussion (probably in a classroom setting) about ethical considerations, such as the impact of rapid advancements on society, and focus on responsible and ethical development in various fields. That certainly is true currently in discussions of AI.

Solving an Equation from 1907 and Liquid Neural Networks

ronkowitz@gmail.com (Kenneth Ronkowitz) — Wed, 07 Dec 2022 08:44:00 +0000

Last year, MIT researchers announced that they had built “liquid” neural networks, inspired by the brains of small species. This is a class of flexible, robust machine-learning models that learn on the job and can adapt to changing conditions. That is important for safety-critical tasks, like driving and flying.

The flexibility of these “liquid” neural nets are great but they are computationally expensive as their number of neurons and synapses increase and require inefficient computer programs to solve their underlying, complicated math.

Now, the same team of scientists has discovered a way to alleviate this bottleneck by solving the differential equation behind the interaction of two neurons through synapses. This unlocks a new type of fast and efficient artificial intelligence algorithm and is orders of magnitude faster, and scalable.

What I find interesting is that the equation that needed to be solved to do this has not had a known solution since 1907. That was the year that the differential equation of the neuron model was introduced. I recall when I was a student and when I was teaching at a university (in the humanities) hearing the complaints of students battling away in a course on differential equations.

These models are ideal for use in time-sensitive tasks like pacemaker monitoring, weather forecasting, investment forecasting, or autonomous vehicle navigation. On a medical prediction task, for example, the new models were 220 times faster on a sampling of 8,000 patients.

Specifically, the team solved, “the differential equation behind the interaction of two neurons through synapses… to unlock a new type of fast and efficient artificial intelligence algorithms.” “The new machine learning models we call ‘CfC’s’ [closed-form Continuous-time] replace the differential equation defining the computation of the neuron with a closed form approximation, preserving the beautiful properties of liquid networks without the need for numerical integration,” MIT professor and CSAIL Director Daniela Rus said. By solving this equation at the neuron-level, the team is hopeful that they’ll be able to construct models of the human brain that measure in the millions of neural connections, something not possible today. The team also notes that this CfC model might be able to take the visual training it learned in one environment and apply it to a wholly new situation without additional work, what’s known as out-of-distribution generalization. That’s not something current-gen models can really do and would prove to be a significant step toward the generalized AI systems of tomorrow.

Source https://news.mit.edu/2022/solving-brain-dynamics-gives-rise-flexible-machine-learning-models-1115